104 Link-Building Statistics You Can Put to Work
104 sourced link-building findings with clear SEO actions. Compare backlinks, rankings and DR, test internal links, calculate campaign costs and download the evidence.
66.5%
Historical links that had decayed
- Population
- Links to 2,062,173 websites
- Window
- Tracked from January 2013; exact endpoint undisclosed
- Read as
- Cumulative decay, not annual loss
A useful link-building statistic should help you decide what to do next. Does a page need more backlinks, a better answer or stronger internal links? Is a high-DR placement worth its price? What should an agency show a client after the links go live?
This hub brings together 104 findings from 41 studies and research releases, with sources checked on 9 September 2026. Each finding includes the result, a plain-language explanation, practical actions and the limit that matters when you use it. The research periods are shown beside the numbers; older studies stay visibly dated.
Choose your route: find a statistic, compare your page with competitors, understand DR and DA, or calculate campaign costs. You can export findings with their citations and download blank worksheets for your own SEO work.
Start with the decision you need to make
You do not need to read every statistic. Choose the task closest to your situation, complete one check and use the supporting findings when you need to explain your recommendation.
From evidence to action
What are you trying to improve?
Choose a task to see the first check and the working document to keep. These are suggested workflows based on the research.
Start with your campaign
Select an issue to see what to check and what evidence to keep.
Start with one page or campaign. The guide turns the research into a manageable next step.
Does your page need more backlinks?
Compare the pages that compete for the same search before setting a link target. Backlinko found that the top result averaged 3.8 times as many backlinks as positions two to ten in its January 2020 dataset, after excluding pages with no backlinks. That is a useful pattern, but it cannot tell you how many links your page needs. Better rankings can also make a page easier for writers to discover and cite. Read the finding and its limits.
In a different study, Ahrefs found a 0.255 correlation between page-level referring domains and rankings. Think of that as a modest relationship, not a 25.5% ranking boost. Neither study tested a rule such as “add ten links and move up three positions.” See the Ahrefs ranking analysis.
Try this with one important page:
- Choose one search query that could bring the right visitors. Use a consistent country and device when checking the results.
- Open the five most relevant competing pages. Note what kind of answer they offer: a guide, product category, local service, comparison or tool.
- Compare the websites linking to each exact page, then inspect several of those sources. Keep the domain score in a separate column.
- Check your page's missing answers and internal links. Choose the clearest next improvement and record why it comes first.
| What you find | A useful next move | What to put in the brief |
|---|---|---|
| Searchers want a comparison, but your page is a sales pitch | Improve the page's purpose and coverage | The question it will answer and the information it lacks |
| Comparable pages have few links, but yours is hard to reach on your own site | Improve relevant internal links | Source pages, destination and descriptive link text |
| Your answer is competitive, but useful industry sources consistently cite other pages | Develop a specific reason to cite yours | The evidence, example or resource you will offer those audiences |
| Useful backlinks point to a missing or outdated page | Repair or update the destination | The old URL, appropriate live destination and owner |
This is a suggested diagnostic workflow, not a ranking formula. Download the page-comparison worksheet.
Improve internal links before expanding outreach
Internal links connect pages on your own website. They can help visitors reach a useful resource and make important pages easier to discover. Ahrefs found that 66.2% of audited domains had at least one page with only one followed incoming internal link. That is a site-level finding, not the percentage of all pages with this issue. See the crawl study.
Start with a useful destination and find related pages where a link would help the reader. Use clear text such as “compare delivery options” when that describes the destination. Avoid adding a long list of unrelated links just to increase a count.
The experiment evidence is mixed in a useful way. A 2026 store-page test estimated a 3% increase in organic sessions after adding an expandable category-link block. Another test estimated an 11% gain from more related-article links but did not reach the publisher's usual 95% confidence threshold. Removing a breadcrumb link produced a 5.5% decline on a different site. These are reasons to test a relevant change and measure both the linking and linked-to pages. Category-link test, related-article test, breadcrumb test.
What DR, DA and page authority can tell you
A domain score describes the website. It does not describe every page on it. DR, DA and Authority Score are metrics made by SEO software companies. They are useful for comparisons within the same tool, but they are not interchangeable or Google approval scores. Ahrefs explains the distinction.
| Measure | What it describes | How to use it |
|---|---|---|
| Referring domains to a page | Distinct websites linking to that exact URL | Compare pages answering the same search; inspect the sources behind the count |
| DR — Ahrefs | The relative strength of a domain's backlink profile | Add context to a competitor or publisher review |
| DA — Moz | Moz's domain-level authority metric | Compare domains using the same tool and date; keep it separate from DR |
| UR — Ahrefs | The link profile of one page, including internal and external links | Review the exact destination or placement URL; a homepage UR is not the whole site's score |
| Authority Score — Semrush | Semrush's metric for the domain or page being checked | State which level you measured and compare it with the same metric |
Ahrefs' DR definition and UR definition explain their different scopes. Our SEO metrics guide puts these measures alongside traffic and keyword data.
A useful benchmark depends on your competitors. Ahrefs' illustrative sample of 600 SaaS companies had a median DR of 62. That describes the selected companies, not a score every SaaS business must reach. The same article showed a DR 29 domain among the top results for a query whose top-ten average was DR 76. A lower score can coexist with a competitive page. Peer-group example, lower-DR ranking example.
Evaluate the exact page before setting a link target
For each proposed placement, write one sentence explaining why this publication's reader would find the reference useful. Then read the publication and check the intended article, surrounding links, recent activity and relevant audience clues.
A traffic estimate helps, but a hard cutoff can hide useful specialist sources. Many paid listings in one marketplace had little estimated Google traffic, while an older ranking study showed that a referring page's traffic threshold was not a universal requirement. The practical answer is to inspect the opportunity, including audiences the SEO tool cannot see. Marketplace evidence, referring-page traffic study.
Put in the shortlist: the publication, proposed article or topic, target URL, audience fit, page-level checks, price and reason to accept or reject. A high score alone is an incomplete explanation.
Plan, measure and maintain the campaign
The workflows below are SEOlutions' interpretation of the evidence. Use them to turn research into a brief your team can deliver and a report a client can understand.
Give outreach a finding worth discussing
A useful pitch tells the writer what happened, who is affected and what evidence they can inspect. In Muck Rack's 2026 survey, 40% wanted original data or research and 58% wanted access to relevant interview sources. Make those easy to use: a short methods note, a clean table and someone who can answer questions. Research finding, interview-access finding.
If you compiled other people's research, call it a sourced reference. If you collected new data, explain your sample and method. Keep the pitch focused on what the reader learns, with a specific reason it fits the journalist's coverage.
Put in the brief: the finding, affected audience, target contacts, evidence available and what counts as a useful reply. Record coverage, links and unlinked mentions separately.
Make the budget explain what will be measured
A quoted placement price and a complete digital PR campaign cost cover different work. A provider survey found that 64% charged for unlinked mentions, so even “cost per result” can hide different deliverables. Define the accepted outputs and full cost before comparing providers. See the billing comparison.
Use your own campaign numbers
What did each retained link cost?
Include fees, production and staff time. An accepted link meets the brief; a retained link is still live at your review date. Keep the same definition for both counts. No data is sent or saved.
Enter all three numbers to compare the original delivery cost with the cost of links that remain live. Or load the example to see how the calculation works.
Use the calculator to understand delivery costs. Then evaluate what the work contributed. A retained link is not proof of a ranking increase, and a ranking increase is not automatically a qualified lead.
| Report line | What to record |
|---|---|
| Delivery | Accepted coverage URLs, link types and distinct referring domains |
| Retention | Links still live, destination status and review date |
| Search performance | Relevant queries, target-page impressions and clicks against a baseline |
| Audience response | Referral visits, useful onward actions and qualified inquiries |
| Other changes | Content edits, technical releases, seasonality and changes in search results |
| Next decision | Continue, change the approach, repair a problem or investigate further |
Download the campaign-report worksheet. Use our guest-post pricing guide when comparing placement scope.
Give link maintenance an owner
Save the source URL, intended destination, observed link and last check date. A lost-link alert may mean the link was removed, the page moved or a crawler temporarily could not access it. Confirm the problem before asking a publisher to fix it. See the loss-reason comparison.
On your own website, prioritize broken destinations that still receive relevant backlinks. Restore useful content or redirect to a genuinely matching replacement. Keep valuable research pages current so writers have a reason to continue citing them.
Put in the handoff: the owner, review dates and the action for each confirmed problem. The historical 66.5% link-decay finding spans about nine years; it is not an annual loss rate. Read the historical study.
Separate AI visibility from website visits
A brand mention, a citation to a publication and a visit to your website are different outcomes. In Pew's March 2025 browsing sample, users clicked a link inside a Google AI summary on 1% of visits with a summary. Citation visibility can exist without a visit. See the browsing evidence.
For AI checks, save the platform, question, date, answer and cited URL. Keep your own website citations separate from third-party coverage. Mentions on well-linked pages had different relationships with different AI systems in one study, so avoid turning one platform's result into a general promise. See the platform comparison.
The current inventory includes 527 placement offers across 510 distinct publisher domains with more than 1,000 recorded AI citations each. We use DataForSEO LLM Mentions to count how often a publisher domain appears as a source in the provider’s ChatGPT and Google AI Overview dataset for the United States, in English. The AI citations total adds the two platform counts. Open a publisher row for the platform breakdown and check date. How we measure AI citations.
Explore all 104 link-building statistics
Search for a topic, publisher or phrase, then export the findings that help with your task. Copy citation includes the claim, original source, study period and a key limitation. Read the methods disclosure when the sample or study design could change your decision.
Topic 01 10 findings
Backlinks and Google rankings
Decide whether links are the likely next step for a page. Start with the search question and the links to the exact pages that rank.
Put this section to work: Compare the top five relevant pages for one query: their referring domains, content format and missing answers. A link gap is a clue, not a required purchase quantity.
Links and rankings have a modest relationship
0.255 / 0.248 correlation
In Ahrefs’ January 2025 US keyword analysis, referring-domain count had a 0.255 rank correlation and backlink count 0.248.
What it means
Pages with more linking websites tended to rank higher, but the relationship was modest. A coefficient near zero means a weak pattern; it is not a percentage increase. Strong rankings can also help a page earn links naturally.
Try this
- Pick one valuable search query and compare the referring domains of the exact pages in the top results.
- Check intent, content and internal links alongside the backlink gap. Use the gap to investigate, not to calculate how many links to buy.
Use with care: 0.255 is a correlation, not 25.5% more traffic or Google’s backlink weighting.
Read the original study: Ahrefs (opens in a new tab)See how the study was done for finding 1
- Research
- Google Says “Links Matter Less”—We Looked at 1,000,000 SERPs to See if It’s True
- How much was studied
- 1,000,000 keywords
- Who or what was studied
- Top 20 Google results for the one million highest-volume US keywords
- When the data came from
- not reported; published January 2025
- Where the number appears
- Correlation table; introduction and methodology
- Keep in mind
- Observational Spearman correlations; high-volume US queries are not representative of all searches. Correlation cannot establish ranking causation.
Reference S001 · Original source checked 9 September 2026
Most indexed pages had no estimated traffic
96.55% of pages
Ahrefs’ 2023 analysis estimated zero Google organic traffic for 96.55% of roughly 14 billion indexed pages.
What it means
Ahrefs estimated no Google visits for most pages in its index. Some may still serve customers or receive traffic elsewhere. A page needs a clear purpose before you invest in attracting links to it.
Try this
- Use Search Console to check whether the intended destination is indexed and appears for relevant searches.
- Improve missing answers or target a clearer query before promoting the page. Keep useful support and conversion pages even if their organic traffic is small.
Use with care: Observed pattern in the stated sample. It does not isolate what caused the result.
Read the original study: Ahrefs (opens in a new tab)See how the study was done for finding 2
- Research
- 96.55% of Content Gets No Traffic From Google. Here’s How to Be in the Other 3.45%
- How much was studied
- approximately 14 billion pages; separate subset of about 20 million pages with no referring domains
- Who or what was studied
- Pages in the Ahrefs index, with estimated Google organic traffic
- When the data came from
- not reported; 2023 report
- Where the number appears
- Opening findings; study limitations
- Keep in mind
- Traffic is estimated from Ahrefs keyword coverage, not measured analytics. Indexed sample is quality-biased and does not represent every page on the web.
Reference S002 · Original source checked 9 September 2026
Some unlinked pages still attracted search visits
2,997 of about 20 million
In a separate subset in Ahrefs’ 2023 traffic study, 2,997 of about 20 million pages with no referring domains exceeded 1,000 estimated monthly Google visits.
What it means
A few pages received substantial estimated Google traffic despite having no referring domains detected by Ahrefs. They may still benefit from internal links or links the tool missed. External-link count alone cannot tell you whether a page can rank.
Try this
- Look for comparable pages that rank with few detected referring domains and read them closely.
- Compare their search intent, specialist detail and internal links with yours. Use this to find content opportunities before setting a link target.
Use with care: Observed pattern in the stated sample. It does not isolate what caused the result.
Read the original study: Ahrefs (opens in a new tab)See how the study was done for finding 3
- Research
- 96.55% of Content Gets No Traffic From Google. Here’s How to Be in the Other 3.45%
- How much was studied
- approximately 14 billion pages; separate subset of about 20 million pages with no referring domains
- Who or what was studied
- Pages in the Ahrefs index, with estimated Google organic traffic
- When the data came from
- not reported; 2023 report
- Where the number appears
- Section discussing pages without backlinks
- Keep in mind
- Traffic is estimated from Ahrefs keyword coverage, not measured analytics. Indexed sample is quality-biased and does not represent every page on the web.
Reference S003 · Original source checked 9 September 2026
Local queries had a stronger link relationship
0.33 correlation
For local queries in Ahrefs’ 2025 analysis, backlinks and referring domains each had a 0.33 ranking correlation.
What it means
Links had a stronger relationship with rankings for the local-query subset than in the whole sample. Local competitors often differ from national ones. Your useful comparison is the actual local search results your customers see.
Try this
- Compare organic results for the same service, location and device. Record page-level links as well as business relevance.
- Find credible local or industry sources that customers would use. Keep Google Maps performance separate from the organic ranking analysis.
Use with care: Organic top-twenty results. This is not a Google Maps ranking study.
Read the original study: Ahrefs (opens in a new tab)See how the study was done for finding 4
- Research
- Google Says “Links Matter Less”—We Looked at 1,000,000 SERPs to See if It’s True
- How much was studied
- Local-query subset of one million keywords; subset size undisclosed
- Who or what was studied
- Top-20 results for the local queries in the parent US keyword sample
- When the data came from
- not reported; published January 2025
- Where the number appears
- Local and non-local query comparison
- Keep in mind
- Observational Spearman correlations; high-volume US queries are not representative of all searches. Correlation cannot establish ranking causation. The local-query subset size is not disclosed.
Reference S085 · Original source checked 9 September 2026
The top result had more backlinks on average
3.8× as many backlinks
In Backlinko’s January 2020 analysis, the number-one result averaged 3.8 times the backlinks of positions two to ten; pages with zero backlinks were excluded from this comparison.
What it means
This is a broad pattern across search results. It does not mean your page needs 3.8 times a competitor’s links. A page that ranks well can also attract links because more people discover it.
Try this
- Compare the exact pages ranking for your target query, using the same backlink tool and date.
- Inspect which links provide useful references. A large count from repeated sitewide links can hide a much smaller number of linking websites.
Use with care: Historical association. The sample excluded pages without backlinks; 3.8× is not a target or a forecast.
Read the original study: Backlinko / Ahrefs (opens in a new tab)See how the study was done for finding 5
- Research
- We Analyzed 11.8 Million Google Search Results. Here’s What We Learned About SEO
- How much was studied
- 1,183,628 cleaned keywords; 11,835,086 ranking results, including 10,052,136 unique URLs
- Who or what was studied
- Top-ten Google results for keywords sampled from the Ahrefs database
- When the data came from
- Data extracted in the first week of January 2020; article updated April 2025
- Where the number appears
- Pages With More Backlinks Tend to Rank Higher; methods PDF, page 4: January 2020 extraction
- Keep in mind
- Historical correlation study, not a link-building experiment. Subsets vary by metric. Some large sites were omitted from certain analyses. The 2025 page update is not new fieldwork.
Reference S139 · Original source checked 9 September 2026
Answering the query belonged in the link-building discussion
0.47 relevance / 0.18 referring domains
In Semrush’s 2024 top-twenty study, text relevance had a 0.47 ranking correlation; the number of referring domains to a URL had a 0.18 correlation.
What it means
Pages that matched the search question more closely tended to rank higher in this sample. Before spending on links, check whether your page gives searchers the kind of answer they want. Relevance here was a modeled text measure, not a human quality review.
Try this
- Open the leading results and note whether they are guides, category pages, tools or comparisons.
- Fix a mismatch in purpose or missing information before using a backlink gap to justify more spending.
Use with care: Observational scores are not Google’s factor weights. This does not prove content is 2.6 times more important than links.
Read the original study: Semrush (opens in a new tab)See how the study was done for finding 6
- Research
- Ranking Factors [Semrush Study]
- How much was studied
- 16,298 English keywords; about 300,000 ranking positions
- Who or what was studied
- Top-twenty results for sampled English keywords with more than 100 monthly searches
- When the data came from
- January 2024 release; collection window undisclosed in the public article
- Where the number appears
- Factors Influence, Top-20 Results chart: Text relevance and Number of referring domains (URL)
- Keep in mind
- Correlations are not Google weights or isolated causal effects. Query mix, correlated metrics and model choices affect the scores. Semrush Authority Score is distinct from Moz DA and Ahrefs DR.
Reference S144 · Original source checked 9 September 2026
Most new pages did not reach the top ten within a year
6.11% in the English-content sample
Ahrefs found that 6.11% of two million new, non-empty English pages from October 2023 reached the top ten within a year; a separate broader September cohort of one million URLs returned 1.74%.
What it means
Even the better-filtered sample had relatively few new pages reach the first page. This is not the success rate of a managed SEO campaign. It is a reminder to check demand, indexing, content and competition instead of promising a ranking date after a link goes live.
Try this
- Separate the date a link is published from the date the target page is discovered, indexed and starts receiving relevant impressions.
- Agree regular review points for the target page. Investigate stalled progress rather than assuming waiting alone will solve it.
Use with care: Two different page samples, not an improvement from 1.74% to 6.11%. The study did not measure time for a backlink to work.
Read the original study: Ahrefs (opens in a new tab)See how the study was done for finding 7
- Research
- How Long Does It Take to Rank in Google? And How Old Are Top Ranking Pages?
- How much was studied
- Two million non-empty English URLs and a separate one-million-URL broader cohort
- Who or what was studied
- New URLs in Ahrefs’ September and October 2023 cohorts
- When the data came from
- New-URL cohorts from September and October 2023 followed for one year; separate ranking-age snapshot published May 2025
- Where the number appears
- How long does it take to rank in Google? First three paragraphs
- Keep in mind
- Different questions use different samples. First seen by the crawler is a proxy for age, not a verified publication date. The study does not isolate backlinks, page quality or effort spent on promotion.
Reference S145 · Original source checked 9 September 2026
Established pages occupied much of the top ten
72.9% more than three years old
In Ahrefs’ 2025 analysis of 1.3 million US keywords, 72.9% of top-ten pages were more than three years old and 13.7% were under one year old, based on crawler first-seen dates.
What it means
Many ranking pages had been around for years. That time can include updates, links and brand growth. Age alone is not an SEO strategy, but an existing useful page may be a better investment than another new URL covering the same subject.
Try this
- Before creating a new campaign page, check whether an existing URL already has relevant links and search visibility.
- Update that page when it still fits the need. Preserve useful URLs and links when consolidating overlapping content.
Use with care: Age was estimated from first discovery by Ahrefs. The study cannot isolate an effect of age or links.
Read the original study: Ahrefs (opens in a new tab)See how the study was done for finding 8
- Research
- How Long Does It Take to Rank in Google? And How Old Are Top Ranking Pages?
- How much was studied
- Top-ten results for 1.3 million random US keywords
- Who or what was studied
- Ranking pages in the separate US keyword sample
- When the data came from
- New-URL cohorts from September and October 2023 followed for one year; separate ranking-age snapshot published May 2025
- Where the number appears
- How old are the top-ranking pages? Age distribution
- Keep in mind
- Different questions use different samples. First seen by the crawler is a proxy for age, not a verified publication date. The study does not isolate backlinks, page quality or effort spent on promotion.
Reference S146 · Original source checked 9 September 2026
The average number-one page was years old
Five years old on average
Ahrefs’ 2025 ranking-age study reported an average first-seen age of five years for number-one results, compared with two years in its 2017 study.
What it means
This is the average age of pages that already won the top position. It is not a five-year waiting time for your page. Winners had different histories, topics and promotion, and the two editions used different samples.
Try this
- Use existing top-ranking pages to study how strong resources develop: their coverage, updates and earned references.
- Give clients milestones they can assess now, such as fixing content gaps and earning relevant coverage, alongside longer-term search outcomes.
Use with care: Average age of winners is not time-to-rank. Sample differences limit comparisons with the 2017 edition.
Read the original study: Ahrefs (opens in a new tab)See how the study was done for finding 9
- Research
- How Long Does It Take to Rank in Google? And How Old Are Top Ranking Pages?
- How much was studied
- Number-one results within the 1.3-million-keyword US sample
- Who or what was studied
- Number-one ranking pages, with 2017 results used as historical context
- When the data came from
- New-URL cohorts from September and October 2023 followed for one year; separate ranking-age snapshot published May 2025
- Where the number appears
- How old are the top-ranking pages? Final paragraph
- Keep in mind
- Different questions use different samples. First seen by the crawler is a proxy for age, not a verified publication date. The study does not isolate backlinks, page quality or effort spent on promotion.
Reference S147 · Original source checked 9 September 2026
Removing link signals coincided with a visibility decline
13.3% estimated traffic drop
In Ahrefs’ 2021 disavow experiment, its SEO pricing article lost 13.3% of estimated search traffic during the disavow period and returned to 99% of its prior level after the file was removed.
What it means
Disavowing asks Google to disregard specified links. This page lost estimated search traffic and then largely recovered. The example supports taking existing links seriously, while the absence of a control group limits what it proves about cause.
Try this
- Protect useful referring links when changing URLs: map old pages to a genuinely matching destination.
- Investigate large losses in relevant backlinks alongside ranking changes. Record other site and search changes before attributing a decline to one cause.
Use with care: One page in an uncontrolled three-page experiment. This is not a forecast and not a test to repeat casually.
Read the original study: Ahrefs (opens in a new tab)See how the study was done for finding 10
- Research
- Do Links Still Matter for Rankings? A Study by Ahrefs
- How much was studied
- One reported page within a three-post experiment
- Who or what was studied
- Ahrefs’ SEO pricing article
- When the data came from
- 17 August–14 September 2021 disavow period, with subsequent recovery observation
- Where the number appears
- The test pages: Example 2, SEO Pricing
- Keep in mind
- Before-and-after experiment without an untreated control group. Competitors, recrawl delays and other changes can affect results. Traffic here is an Ahrefs estimate based on rankings, not counted sessions.
Reference S148 · Original source checked 9 September 2026
Topic 03 14 findings
Internal links you can improve
Find changes on your own site, with positive, negative and uncertain test results kept together.
Put this section to work: Choose an important page that is hard to reach. Add useful links from related pages and monitor both sides of the change. Compare with unchanged pages when traffic allows.
Many domains had sparsely linked pages
66.2% of domains
In Ahrefs’ 2023 audit sample, 66.2% of domains had at least one page with only one followed incoming internal link.
What it means
A page can be published yet receive very few links from the rest of its website. That can make it hard to find. Internal links are often a practical first check because your team controls them.
Try this
- Use a crawl to list important indexable pages with few incoming internal links, then verify their business and search purpose.
- Add helpful links from related articles, categories or service pages. Use descriptive words so visitors know what opens next.
Use with care: Observed pattern in the stated sample. It does not isolate what caused the result.
Read the original study: Ahrefs (opens in a new tab)See how the study was done for finding 18
- Research
- We Crawled Over 1 Million Domains to Find the Most Common Technical SEO Issues
- How much was studied
- more than 1 million domains
- Who or what was studied
- Domains processed through Ahrefs Site Audit; a domain counts if at least one page has the issue
- When the data came from
- not reported; August 2023 report
- Where the number appears
- Most common issues; incoming internal links
- Keep in mind
- Site Audit sample is not a random sample of the entire web. Percentages apply to domains with at least one affected page, not the proportion of all pages.
Reference S042 · Original source checked 9 September 2026
Redirect links were common in crawls
62.7% of domains
In Ahrefs’ 2023 audit sample, 62.7% of domains had at least one page linking to a redirect.
What it means
Many sites linked through a redirect instead of the final page. Updating the link can make the journey cleaner.
Try this
- Find internal links that first pass through a redirect.
- Point important links straight to the final page while keeping useful external redirects.
Use with care: Observed pattern in the stated sample. It does not isolate what caused the result.
Read the original study: Ahrefs (opens in a new tab)See how the study was done for finding 19
- Research
- We Crawled Over 1 Million Domains to Find the Most Common Technical SEO Issues
- How much was studied
- more than 1 million domains
- Who or what was studied
- Domains processed through Ahrefs Site Audit; a domain counts if at least one page has the issue
- When the data came from
- not reported; August 2023 report
- Where the number appears
- Most common issues; links to redirects
- Keep in mind
- Site Audit sample is not a random sample of the entire web. Percentages apply to domains with at least one affected page, not the proportion of all pages.
Reference S043 · Original source checked 9 September 2026
More internal links correlated with more clicks
4× average Google clicks
In Zyppy’s undated internal-link study, URLs with 40–44 incoming internal links averaged four times the Google clicks of URLs with zero to four.
What it means
The study grouped pages by how many internal links they received and compared Google clicks. More important pages may naturally attract both. The fourfold difference is a relationship in the sample, not an expected return from adding links.
Try this
- Compare similar pages on your site by purpose, internal links and Search Console clicks to find neglected destinations.
- Try a relevant linking improvement and record other changes. Where traffic allows, compare with a similar unchanged group.
Use with care: Observed pattern in the stated sample. It does not isolate what caused the result.
Read the original study: Zyppy (opens in a new tab)See how the study was done for finding 20
- Research
- 23 Million Internal Links – SEO Case Study
- How much was studied
- 23 million internal links, 1,800 websites, approximately 520,000 URLs
- Who or what was studied
- URLs on participating websites, with internal-link counts compared to Google Search Console clicks
- When the data came from
- Collection window undisclosed; article updated 23 February 2026
- Where the number appears
- More internal links correlate with higher traffic
- Keep in mind
- Study and measurement dates undisclosed; page updated February 23, 2026. Observational averages can reflect page importance and other confounders. Results do not identify an optimal link count or causal anchor-text effect.
Reference S045 · Original source checked 9 September 2026
Keyword-matching internal anchors had a traffic association
At least 5× search traffic
In Zyppy’s undated study, pages with at least one exact-match internal anchor received at least five times the search traffic of pages without one.
What it means
An exact-match anchor uses the search phrase a page targets as the clickable words. Pages with at least one such internal link had more traffic in this sample. Better-optimized or more important pages may explain the difference; it does not establish a fivefold benefit from one link.
Try this
- Use a natural, accurate description of the destination when adding an internal link. Include its topic when that helps readers.
- Review your important pages for unclear labels. Avoid forcing the same keyword into every link or treating this association as a traffic forecast.
Use with care: At least one exact-match internal anchor, not descriptive wording in general. Correlation does not establish cause.
Read the original study: Zyppy (opens in a new tab)See how the study was done for finding 21
- Research
- 23 Million Internal Links – SEO Case Study
- How much was studied
- 23 million internal links, 1,800 websites, approximately 520,000 URLs
- Who or what was studied
- URLs on participating websites, with internal-link counts compared to Google Search Console clicks
- When the data came from
- Collection window undisclosed; article updated 23 February 2026
- Where the number appears
- Exact Match Anchors and Google Search Traffic
- Keep in mind
- Study and measurement dates undisclosed; page updated February 23, 2026. Observational averages can reflect page importance and other confounders. Results do not identify an optimal link count or causal anchor-text effect.
Reference S046 · Original source checked 9 September 2026
A category-link test improved traffic
+25% estimated traffic
SearchPilot’s August 2020 grocery-site case study estimated a 25% organic-traffic increase across the tested category pages after adding links between category levels.
What it means
This ecommerce test added links between related category levels and measured both pages giving and receiving links. The combined gain was substantial on that site. It gives you a test idea where useful categories are difficult to reach.
Try this
- Choose one group of related category pages and add links that make a genuine shopping task easier.
- Measure source and destination pages together and separately against an unchanged group. Include onward clicks or conversions when available.
Use with care: One website and one implementation. Use the result as a test idea, with its confidence limits intact.
Read the original study: SearchPilot (opens in a new tab)See how the study was done for finding 22
- Research
- SEO Split Test Lessons: Increasing Internal Linking
- How much was studied
- Not disclosed
- Who or what was studied
- Category pages on one grocery website; links added between category levels
- When the data came from
- not reported; case study published August 2020
- Where the number appears
- Results of increasing internal linking experiment
- Keep in mind
- One client experiment using page groups and a modeled control/counterfactual. The estimated effect does not establish an expected uplift for other sites; page count and experiment dates not reported.
Reference S047 · Original source checked 9 September 2026
A positive estimate can remain uncertain
+10.2%; inconclusive at 95%
SearchPilot’s April 2024 link-block experiment estimated a 10.2% organic-traffic increase, but the result was inconclusive at 95% confidence.
What it means
A positive best estimate can still be uncertain: the plausible range can include no improvement. That is different from proving the change had no effect. The reader needs both the estimate and its uncertainty to judge the test.
Try this
- Agree the success measure and acceptable uncertainty before changing an internal-link component.
- Report the estimate and confidence threshold together. Keep inconclusive outcomes in the learning log instead of presenting them as guaranteed wins.
Use with care: One website and one implementation. Use the result as a test idea, with its confidence limits intact.
Read the original study: SearchPilot (opens in a new tab)See how the study was done for finding 23
- Research
- SEO Split Test Lessons: Reducing a Link Block
- How much was studied
- Not disclosed
- Who or what was studied
- One website’s desktop internal-link block, shortened with a Show More control
- When the data came from
- not reported; case study published April 2024
- Where the number appears
- Results of reducing link block experiment
- Keep in mind
- Estimated result is inconclusive at 95% confidence, although positive at 90%. Do not describe it as a demonstrated gain or general advice to hide links; test sample not reported.
Reference S048 · Original source checked 9 September 2026
Internal links also show an association
0.117 correlation
Incoming internal links had a 0.117 ranking correlation in the same Ahrefs sample.
What it means
Pages with more incoming internal links tended to rank a little better. This was a relationship, not proof.
Try this
- Ask a site-audit tool to find important pages with few internal links.
- Add links from related strong pages, then watch whether discovery and search visits improve.
Use with care: Observed pattern in the stated sample. It does not isolate what caused the result.
Read the original study: Ahrefs (opens in a new tab)See how the study was done for finding 24
- Research
- Google Says “Links Matter Less”—We Looked at 1,000,000 SERPs to See if It’s True
- How much was studied
- 1,000,000 keywords
- Who or what was studied
- Top 20 Google results for the one million highest-volume US keywords
- When the data came from
- not reported; published January 2025
- Where the number appears
- Correlation table: Internal links
- Keep in mind
- Observational Spearman correlations; high-volume US queries are not representative of all searches. Correlation cannot establish ranking causation.
Reference S084 · Original source checked 9 September 2026
URL anchors had a traffic association
Almost 50% more traffic
In Zyppy’s undated study, pages with naked-URL internal anchors had almost 50% more traffic than pages without them.
What it means
Pages with URL-shaped link text got more traffic on average. That does not mean raw URLs are better for readers.
Try this
- Find internal links that use the full web address as their text.
- Keep them only when the address helps readers; otherwise use descriptive words.
Use with care: Observed pattern in the stated sample. It does not isolate what caused the result.
Read the original study: Zyppy (opens in a new tab)See how the study was done for finding 25
- Research
- 23 Million Internal Links – SEO Case Study
- How much was studied
- 23 million internal links, 1,800 websites, approximately 520,000 URLs
- Who or what was studied
- URLs on participating websites, with internal-link counts compared to Google Search Console clicks
- When the data came from
- Collection window undisclosed; article updated 23 February 2026
- Where the number appears
- Naked URL anchors
- Keep in mind
- Study and measurement dates undisclosed; page updated February 23, 2026. Observational averages can reflect page importance and other confounders. Results do not identify an optimal link count or causal anchor-text effect.
Reference S087 · Original source checked 9 September 2026
Some internal links had no anchor text
Over 6% of links
More than 6% of internal links in Zyppy’s sample had no anchor text.
What it means
Some internal links had no words. Many may have been image links, so each case needs checking.
Try this
- Find links with no text and check whether linked images have useful labels.
- Fix broken or invisible controls first, and keep legitimate image links.
Use with care: Observed pattern in the stated sample. It does not isolate what caused the result.
Read the original study: Zyppy (opens in a new tab)See how the study was done for finding 26
- Research
- 23 Million Internal Links – SEO Case Study
- How much was studied
- 23 million internal links, 1,800 websites, approximately 520,000 URLs
- Who or what was studied
- URLs on participating websites, with internal-link counts compared to Google Search Console clicks
- When the data came from
- Collection window undisclosed; article updated 23 February 2026
- Where the number appears
- Anchor text analysis
- Keep in mind
- Study and measurement dates undisclosed; page updated February 23, 2026. Observational averages can reflect page importance and other confounders. Results do not identify an optimal link count or causal anchor-text effect. The total study covered about 23 million internal links. Missing text can include image links; it is not automatically an SEO penalty.
Reference S088 · Original source checked 9 September 2026
A footer-link test differed by device
+10% desktop; +5% mobile
SearchPilot’s 2021 footer-link experiment estimated a 10% desktop traffic lift at 95% confidence and 5% mobile lift at 90% confidence.
What it means
The same footer-link change gave different levels of confidence on desktop and mobile. Looking only at the overall average could hide that difference. A link module also needs to remain usable on a small screen.
Try this
- Review traffic and navigation behavior by device after an internal-link change.
- Keep the combined result alongside device results and their confidence levels. Check that extra links remain readable and useful on mobile.
Use with care: One website and one implementation. Use the result as a test idea, with its confidence limits intact.
Read the original study: SearchPilot (opens in a new tab)See how the study was done for finding 27
- Research
- Does adding internal links to your home page’s footer improve organic traffic?
- How much was studied
- One controlled page-split experiment; page count undisclosed
- Who or what was studied
- Linked-to pages on one customer site after expanding homepage footer links
- When the data came from
- Test reported September 2021; exact dates not reported
- Where the number appears
- Device-level results after expanding roughly 30 footer links to 100+
- Keep in mind
- Desktop lift met 95% confidence; mobile met 90%, not 95%. Combined estimate was inconclusive at 95%. One customer and one implementation.
Reference S127 · Original source checked 9 September 2026
A breadcrumb change reduced traffic
−5.5% organic traffic
A SearchPilot 2024 ecommerce test reported a statistically significant 5.5% traffic decline after removing self-referential breadcrumb links.
What it means
Removing a breadcrumb’s link to its own page was followed by a decline in this controlled test. The visible navigation changed but its structured data did not. A seemingly tidy change can affect a real site in unexpected ways.
Try this
- Review visible breadcrumbs and their structured data together before changing a template.
- Test on comparable pages and keep a rollback ready. Do not use this single result to declare every self-link necessary.
Use with care: One website and one implementation. Use the result as a test idea, with its confidence limits intact.
Read the original study: SearchPilot (opens in a new tab)See how the study was done for finding 28
- Research
- Will removing self-referential breadcrumb links improve organic traffic?
- How much was studied
- One controlled page-split experiment; page count undisclosed
- Who or what was studied
- Product listing pages on one ecommerce customer site
- When the data came from
- Test reported July 2024; exact dates not reported
- Where the number appears
- Results; implementation and schema mismatch
- Keep in mind
- The visible breadcrumb changed while its schema did not. The author’s proposed explanation is not separately tested; this cannot isolate a universal effect of self-links.
Reference S128 · Original source checked 9 September 2026
Connecting nearby location pages helped one site
+7% organic traffic
SearchPilot reported a 7% organic-traffic uplift for regional pages receiving new links from nearby regions in a controlled test published in January 2021.
What it means
A visitor interested in one area may also need a nearby location. The site made those connections easier to follow, and the linked pages gained traffic against the control group. The starting point was a site without a consistent linking system for these pages.
Try this
- For a multi-location business, identify genuinely nearby alternatives a visitor could use and connect those pages.
- Measure both source and destination pages where possible, with a comparable unchanged group. Avoid adding hundreds of irrelevant city links.
Use with care: One website. The +7% result is evidence for this implementation, not an expected lift for every local site.
Read the original study: SearchPilot (opens in a new tab)See how the study was done for finding 29
- Research
- Internal linking case study: can links to nearby location pages help?
- How much was studied
- Around 8,000 regional pages on one website; exact group sizes undisclosed
- Who or what was studied
- Regional US pages listing a company’s nearby physical locations
- When the data came from
- January 2021 case publication; test dates undisclosed
- Where the number appears
- The Case Study: six nearest neighbouring regional pages; final result
- Keep in mind
- One customer with limited systematic internal linking before the test. The reported change applies to linked-to pages. Confidence bounds are not stated in the article text.
Reference S151 · Original source checked 9 September 2026
More related-article links produced an uncertain overall gain
+11% estimate; below 95% confidence
Increasing related-article links from two to four in one SearchPilot test produced an estimated 11% combined organic-traffic lift, positive at 90% confidence but inconclusive at 95%.
What it means
The combined result was promising but less certain than a standard 95% finding. Pages adding links showed a clearer improvement than pages receiving them. That is why a review should examine both sides of an internal-link change.
Try this
- Add a small, relevant related-reading block where it helps visitors take the next step.
- Track the linking pages and linked-to pages separately as well as together; report an inconclusive result as uncertain, not zero or a proven gain.
Use with care: Single controlled test. The combined estimate did not meet the publisher’s usual 95% threshold.
Read the original study: SearchPilot (opens in a new tab)See how the study was done for finding 30
- Research
- Will increasing the number of related article links help for SEO?
- How much was studied
- One ecommerce content hub; page counts undisclosed
- Who or what was studied
- Content-hub pages linking to related articles, with donor and recipient groups
- When the data came from
- June 2021 case publication; test dates undisclosed
- Where the number appears
- The Case Study: combined, donor and recipient result paragraphs
- Keep in mind
- The combined +11% estimate was positive at 90% confidence, not 95%. Donor pages had a +16% estimate at 95%; recipient results were inconclusive. This is one experiment, counted once.
Reference S152 · Original source checked 9 September 2026
A compact category-link block improved store-page traffic
+3% organic sessions
SearchPilot’s January 2026 store-page accordion test estimated a 3% increase in daily Google organic sessions, with a 95% credible interval from +0.2% to +5.7%.
What it means
The site added an expandable list of useful categories and services. The most likely gain was modest, with a range of possible effects. This is a practical example of improving navigation while keeping a busy page readable.
Try this
- On store or service pages, identify category links that answer the visitor’s likely next question.
- Test a clear, accessible block on similar pages and monitor visits plus useful onward clicks. Use the observed range when explaining uncertainty to a client.
Use with care: One ecommerce site. The chart’s range is uncertainty about this test, not a predicted range for your site.
Read the original study: SearchPilot (opens in a new tab)See how the study was done for finding 31
- Research
- Can a Categories Accordion Boost Organic Traffic?
- How much was studied
- One ecommerce site; page count undisclosed
- Who or what was studied
- Store pages receiving a collapsible block of category and service links
- When the data came from
- January 2026 case publication; test dates undisclosed
- Where the number appears
- Results chart: Range of Outcomes on Daily Google Organic Sessions; values visually inspected
- Keep in mind
- Single-site result. The chart gives a +3% best estimate and +0.2% to +5.7% at the 95% credible interval. The test combines navigation, link and presentation changes, so their individual contributions are not isolated.
Reference S153 · Original source checked 9 September 2026
Topic 04 9 findings
Choosing worthwhile placements
Check the article, its audience and the reason for the reference before relying on a score or traffic filter.
Put this section to work: Read recent articles and inspect the proposed placement. Record topic fit, real audience clues, surrounding links and the useful destination.
Reciprocal linking was common in the sample
73.6% of domains
In Ahrefs’ historical study of 140,592 domains with search traffic, 73.6% had at least one reciprocal linking relationship.
What it means
Many websites linked to sites that also linked back. This can happen naturally between related publishers.
Try this
- Check whether each two-way link is genuinely helpful to readers.
- Review repeated link swaps between the same small group of websites.
Use with care: Observed pattern in the stated sample. It does not isolate what caused the result.
Read the original study: Ahrefs (opens in a new tab)See how the study was done for finding 32
- Research
- Reciprocal Links: Do They Help or Hurt Your SEO?
- How much was studied
- 140,592 domains
- Who or what was studied
- Domains with search traffic; followed incoming and outgoing links only
- When the data came from
- not reported
- Where the number appears
- How common are reciprocal links on the web?
- Keep in mind
- Historical study; page updated March 11, 2020, but original publication and fieldwork dates are not established. Traffic-selected sample has survivor bias. Reciprocity is not proof of an exchange scheme or its safety.
Reference S007 · Original source checked 9 September 2026
Substantial two-way overlap also appeared
27.4% of domains
In the same undated Ahrefs sample, 27.4% of domains had at least 15% overlap between the domains they linked to and those linking back.
What it means
More than a quarter of the sites had a lot of two-way links. The number describes what existed, not what you should copy.
Try this
- Find which websites exchange the most links with your site.
- Keep natural partnerships, but do not aim for this study’s percentage.
Use with care: Observed pattern in the stated sample. It does not isolate what caused the result.
Read the original study: Ahrefs (opens in a new tab)See how the study was done for finding 33
- Research
- Reciprocal Links: Do They Help or Hurt Your SEO?
- How much was studied
- 140,592 domains
- Who or what was studied
- Domains with search traffic; followed incoming and outgoing links only
- When the data came from
- not reported
- Where the number appears
- Distribution of reciprocal-link overlap
- Keep in mind
- Historical study; page updated March 11, 2020, but original publication and fieldwork dates are not established. Traffic-selected sample has survivor bias. Reciprocity is not proof of an exchange scheme or its safety.
Reference S008 · Original source checked 9 September 2026
Reciprocal links appear on ranking pages
43.7% of pages
Ahrefs found reciprocal links on 43.7% of 112,440 top-ranking pages in a separate historical sample.
What it means
Two-way links also appeared on high-ranking pages. That does not prove that planned link swaps are helpful or safe.
Try this
- Record why each site links to the other and how it helps readers.
- Review repeated swaps, copied link text or links to unrelated pages.
Use with care: Observed pattern in the stated sample. It does not isolate what caused the result.
Read the original study: Ahrefs (opens in a new tab)See how the study was done for finding 34
- Research
- Reciprocal Links: Do They Help or Hurt Your SEO?
- How much was studied
- 112,440 top-ranking pages from about 10,000 selected queries
- Who or what was studied
- Top-ranking pages for non-branded queries with keyword difficulty 40–60
- When the data came from
- not reported
- Where the number appears
- How much reciprocal linking happens in the top 10?
- Keep in mind
- Historical study; page updated March 11, 2020, but original publication and fieldwork dates are not established. Traffic-selected sample has survivor bias. Reciprocity is not proof of an exchange scheme or its safety. This subset used about 10,000 non-branded queries with keyword difficulty 40–60; it does not establish that exchange schemes are safe.
Reference S086 · Original source checked 9 September 2026
Many listings had little estimated traffic
39.6% zero; 78% below 1,000
In BuzzStream’s 2026 marketplace sample, 39.6% of sites had zero estimated organic traffic; 78% had fewer than 1,000 monthly visits.
What it means
Many sites in this marketplace had little estimated Google traffic. That makes a large inventory less impressive than it first sounds. The estimate still misses other audiences, so a low number is a reason to investigate the site more closely.
Try this
- Check the publisher’s topic, recent articles, search trend and evidence of a relevant audience before comparing prices.
- Record the traffic tool, date and country. Reject an opportunity for a clear lack of reader value rather than applying an unexplained traffic cutoff.
Use with care: One marketplace. Listings and tool estimates do not represent every publisher or a completed transaction.
Read the original study: BuzzStream (opens in a new tab)See how the study was done for finding 35
- Research
- Cost of Guest Posts, Based on 500k Sites [Updated 2026 Data]
- How much was studied
- 257,267 sites with price, Domain Rating and estimated traffic data
- Who or what was studied
- Complete listings in one unnamed link marketplace
- When the data came from
- June 2026 release; collection dates undisclosed
- Where the number appears
- Key takeaways: traffic distribution
- Keep in mind
- One marketplace, not the whole web. Listed quotes are not completed transactions. Traffic is an SEO-tool estimate; incomplete listings were excluded. Both traffic bands describe the same sample and count as one distribution finding.
Reference S104 · Original source checked 9 September 2026
Few listings passed the study’s top tiers
1.37% of listed sites
Only 1.37% of sites in BuzzStream’s 2026 sample met its “High Quality” or “Top Tier” metric thresholds.
What it means
Very few listings met this study’s strictest quality rules. Always ask what a seller’s quality label actually means.
Try this
- Ask sellers exactly what labels such as “premium” mean.
- Judge every site using your own relevance, quality and audience checks.
Use with care: One marketplace. Listings and tool estimates do not represent every publisher or a completed transaction.
Read the original study: BuzzStream (opens in a new tab)See how the study was done for finding 36
- Research
- Cost of Guest Posts, Based on 500k Sites [Updated 2026 Data]
- How much was studied
- 257,267 sites with price, Domain Rating and estimated traffic data
- Who or what was studied
- Complete listings in one unnamed link marketplace
- When the data came from
- June 2026 release; collection dates undisclosed
- Where the number appears
- Quality tier distribution; threshold definitions
- Keep in mind
- One marketplace, not the whole web. Listed quotes are not completed transactions. Traffic is an SEO-tool estimate; incomplete listings were excluded. These are the author’s DR/traffic tiers, not an editorial audit: DR 71–80 with 50,000+ traffic, or DR 81+ with 100,000+ traffic.
Reference S105 · Original source checked 9 September 2026
Niche labels do not establish an audience
70% of gambling listings
About 70% of roughly 10,000 gambling sites in BuzzStream’s 2026 marketplace subset had zero estimated organic traffic.
What it means
Most gambling-category sites in this marketplace had no estimated search traffic. A niche label can be misleading.
Try this
- Read recent articles and check search visibility before trusting a niche label.
- Remove sites whose real content or audience does not match your campaign.
Use with care: One marketplace. Listings and tool estimates do not represent every publisher or a completed transaction.
Read the original study: BuzzStream (opens in a new tab)See how the study was done for finding 37
- Research
- Cost of Guest Posts, Based on 500k Sites [Updated 2026 Data]
- How much was studied
- Roughly 10,000 gambling-site listings
- Who or what was studied
- Gambling-category subset of one marketplace’s complete listings
- When the data came from
- June 2026 release; collection dates undisclosed
- Where the number appears
- Gambling deep dive
- Keep in mind
- One marketplace, not the whole web. Listed quotes are not completed transactions. Traffic is an SEO-tool estimate; incomplete listings were excluded. A subset of one marketplace; not a statement about all gambling publishers.
Reference S106 · Original source checked 9 September 2026
Page relevance and domain relevance differ
62% vs 52%
Aira’s 2022 respondents assessed relevance using the linking page’s topic (62%) and the domain’s topic (52%).
What it means
A website can cover your topic while the actual placement page does not. You need to check both.
Try this
- Check whether both the website and the planned article fit your topic.
- Choose the page that helps readers most, even on a broader website.
Use with care: Survey answers describe this group of respondents. They do not prove a tactic improves rankings.
Read the original study: Aira (opens in a new tab)See how the study was done for finding 38
- Research
- 11 Takeaways from the 2022 State of Link Building Report
- How much was studied
- 270 respondents; branching questions have different denominators
- Who or what was studied
- Self-selected SEO professionals answering the annual link-building survey
- When the data came from
- 16 March–2 May 2022
- Where the number appears
- Takeaway 7; relevance criteria
- Keep in mind
- Historical self-reported practice and opinions. Multiple-choice questions may allow several answers; responses are not measured ranking effects.
Reference S117 · Original source checked 9 September 2026
A “toxic” label did not identify an easy traffic win
7.1% fewer Search Console clicks
Across three Ahrefs posts, Search Console traffic fell 7.1% in the 20 days after 129 tool-flagged backlink URLs were disavowed in July 2024; the author saw little or no overall effect in estimated search visibility.
What it means
The label came from SEO software. Removing those links did not produce a clear win, and the two traffic measures told different stories. A tool warning should start an investigation, not automatically create a disavow file.
Try this
- Review the actual linking pages and your Search Console manual-action report before deciding whether action is needed.
- Follow Google’s disavow guidance for substantial unnatural-link problems with an actual or likely manual action; a tool score alone is insufficient.
Use with care: Uncontrolled before-and-after observation. The 7.1% fall does not prove that disavowing caused harm.
Read the original study: Ahrefs (opens in a new tab)See how the study was done for finding 39
- Research
- I Disavowed “Toxic Backlinks”: Here’s What Happened
- How much was studied
- Three Ahrefs posts; 129 URLs classified by another SEO tool
- Who or what was studied
- Three selected Ahrefs posts with tool-flagged backlinks
- When the data came from
- 20 days before and after the 26 July 2024 disavow; ended before the August core update
- Where the number appears
- Opening result, experiment duration sidenote and interpretation before Test pages
- Keep in mind
- No control group and a short observation window. Search Console clicks and Ahrefs visibility estimates told different stories. The author judged the overall visibility effect small or absent. The click decline cannot be attributed to disavowing alone.
Reference S149 · Original source checked 9 September 2026
A traffic threshold can hide useful link opportunities
About one in five search results sets
In Ahrefs’ 2019 sample of 44,589 queries, roughly one fifth of result sets had no top-ten page with a backlink meeting the study’s 100+ monthly organic-visit threshold for the referring page.
What it means
Those pages still ranked even without a backlink that passed this traffic filter. A source can reach people through email, direct visits or a specialist community, which an organic-traffic estimate misses. Traffic is a useful clue, not a universal pass mark.
Try this
- Separate estimated Google traffic from evidence of a real audience, such as relevant recent coverage and active distribution.
- Inspect the exact source page and why someone would use its link. Do not reject a useful niche source solely for falling below an arbitrary traffic cutoff.
Use with care: 2019 observations with a specific organic-traffic threshold. This does not make irrelevant or abandoned sites good prospects.
Read the original study: Ahrefs (opens in a new tab)See how the study was done for finding 40
- Research
- Do Links From Pages With Traffic Help You Rank Higher? [Ahrefs Study]
- How much was studied
- 44,589 non-branded search queries
- Who or what was studied
- Top-ten results for queries with 2,000–5,000 monthly searches
- When the data came from
- February 2019 release; exact extraction dates undisclosed
- Where the number appears
- How common are “links with traffic” across the studied SERPs?; threshold discussion in Study 1
- Keep in mind
- Historical data and estimated Google traffic only. The study’s referring-page traffic threshold was 100+ monthly visits. Low estimated organic traffic does not mean no readers, no referral value or no SEO value.
Reference S150 · Original source checked 9 September 2026
Topic 05 12 findings
Content that earns links
Give writers something useful to cite: original evidence, a clear explanation, a reference table or a practical tool.
Put this section to work: Find three relevant pages that cite similar resources. Identify what they needed from the source, then improve that specific job and make it easy to reference.
Data-led pitches were widely used
95.9% of respondents
In BuzzStream’s 2026 survey of more than 150 digital PR professionals, 95.9% used pitches built around data.
What it means
Almost all respondents used data-led pitches. Data helps only when it answers a question people care about.
Try this
- Choose one trustworthy number that answers a current audience question.
- Test the story with a few relevant journalists before doing more work.
Use with care: Self-reported tactics or outcomes. Campaigns, audiences and promotion were not held equal.
Read the original study: BuzzStream (opens in a new tab)See how the study was done for finding 41
- Research
- State of Digital PR Report (2026)
- How much was studied
- 150+ professionals
- Who or what was studied
- Digital PR professionals responding to the publisher’s survey
- When the data came from
- not reported; 2026 release
- Where the number appears
- Digital PR Tactics, question 5
- Keep in mind
- Self-reported vendor survey, not representative of all campaigns. Costs, success definitions and times to results differ across respondents; fieldwork dates undisclosed.
Reference S022 · Original source checked 9 September 2026
Original research was part of many content plans
49% of respondents
In Orbit Media’s August 2025 survey of 808 content marketers, 49% had created and published original research in the preceding year.
What it means
About half used original research. First-party evidence can give other writers something new to cite.
Try this
- List useful company data, customer questions and repeated observations.
- Check privacy, definitions and whether you can update the study later.
Use with care: Self-reported tactics or outcomes. Campaigns, audiences and promotion were not held equal.
Read the original study: Orbit Media (opens in a new tab)See how the study was done for finding 42
- Research
- Annual Blogger Survey 2025
- How much was studied
- 808 respondents
- Who or what was studied
- Individual bloggers/content marketers; sample skews toward US, B2B and the author’s LinkedIn network
- When the data came from
- August 2025
- Where the number appears
- Question 20 chart, original research adoption; methodology
- Keep in mind
- Voluntary survey with no incentives. Strong results are self-defined, not a uniform SEO or revenue metric. Cross-sectional associations are not causal. Selected chart values visually inspected.
Reference S049 · Original source checked 9 September 2026
Guest bloggers reported stronger outcomes
30% vs 15% strong results
In Orbit Media’s August 2025 survey, 30% of guest bloggers reported strong results, compared with 15% of those who did not guest blog.
What it means
Guest bloggers were more likely to report strong results. The survey does not show that the link caused those results.
Try this
- Choose guest articles for the right audience and editorial fit.
- Measure visits, links and relationships separately, and avoid claiming one article changed rankings.
Use with care: Self-reported tactics or outcomes. Campaigns, audiences and promotion were not held equal.
Read the original study: Orbit Media (opens in a new tab)See how the study was done for finding 43
- Research
- Annual Blogger Survey 2025
- How much was studied
- 808 respondents
- Who or what was studied
- Individual bloggers/content marketers; sample skews toward US, B2B and the author’s LinkedIn network
- When the data came from
- August 2025
- Where the number appears
- Question 17 chart, guest blogging and results
- Keep in mind
- Voluntary survey with no incentives. Strong results are self-defined, not a uniform SEO or revenue metric. Cross-sectional associations are not causal. Selected chart values visually inspected.
Reference S051 · Original source checked 9 September 2026
Blogs and product pages play different roles
68% vs 41%
In BuzzStream and Citation Labs’ 2025 survey, 68% prioritized blog posts for links and 41% prioritized product pages.
What it means
Teams prioritized links to useful blog content more often than product pages. A research resource can answer a question that a sales page does not. The two pages need a sensible connection if the campaign is meant to support the business.
Try this
- Choose a resource topic that overlaps with a real customer question and a relevant product or service.
- Add a natural internal link to the next useful step. Measure the resource’s earned links and the visits or inquiries it assists.
Use with care: Self-reported tactics or outcomes. Campaigns, audiences and promotion were not held equal.
Read the original study: BuzzStream and Citation Labs (opens in a new tab)See how the study was done for finding 44
- Research
- Link Building Trends Report (2025)
- How much was studied
- Not disclosed
- Who or what was studied
- Agency, freelance, consultancy and in-house link-building professionals
- When the data came from
- not reported; 2025 release
- Where the number appears
- Types of pages prioritized for backlinks
- Keep in mind
- Self-reported, non-probability survey; full sample size not stated. The approximately 150 respondents cited in the management section are the in-house subset only. Multiple page types could be selected.
Reference S093 · Original source checked 9 September 2026
Longer articles attracted more referring domains
77.2% more linking domains
In the 2017–2018 content sample, articles over 3,000 words averaged 77.2% more referring domains than articles below 1,000 words.
What it means
Longer articles attracted more referring domains on average in this historical dataset. Length can come with deeper research or better promotion, so adding words alone is not the lesson. Make your page more useful to someone who needs a reference.
Try this
- Look at why writers cite competing resources: a statistic, comparison, definition, table or original example.
- Add the missing evidence and make it easy to locate. Remove repetition even if it reduces the word count.
Use with care: Observed pattern in the stated sample. It does not isolate what caused the result.
Read the original study: Backlinko / BuzzSumo (opens in a new tab)See how the study was done for finding 45
- Research
- We Analyzed 912 Million Blog Posts. Here’s What We Learned About Content Marketing
- How much was studied
- 912.5 million articles
- Who or what was studied
- Articles in the BuzzSumo index
- When the data came from
- Articles published 1 October 2017–1 October 2018
- Where the number appears
- Long-form content gets more backlinks; methods PDF
- Keep in mind
- Historical, observational data. Links are unique referring domains in the methods. Means can be skewed by outliers; formats were classified automatically.
Reference S110 · Original source checked 9 September 2026
Most articles attracted no external links
94% had no linking domains
In the historical Backlinko/BuzzSumo sample, 94% of articles had no external referring domains; only 2.2% had links from multiple sites.
What it means
Most articles earned no links from other websites. Good content still needs a useful topic and active promotion.
Try this
- Before writing, find who links to similar resources and why.
- Plan promotion, internal links and future updates before publishing.
Use with care: Observed pattern in the stated sample. It does not isolate what caused the result.
Read the original study: Backlinko / BuzzSumo (opens in a new tab)See how the study was done for finding 46
- Research
- We Analyzed 912 Million Blog Posts. Here’s What We Learned About Content Marketing
- How much was studied
- 912.5 million articles
- Who or what was studied
- Articles in the BuzzSumo index
- When the data came from
- Articles published 1 October 2017–1 October 2018
- Where the number appears
- 94% of blog posts have zero external links
- Keep in mind
- Historical, observational data. Links are unique referring domains in the methods. Means can be skewed by outliers; formats were classified automatically.
Reference S111 · Original source checked 9 September 2026
Social shares were a weak link proxy
0.078 correlation
Backlinko and BuzzSumo reported a 0.078 Pearson correlation between social shares and backlinks in their historical content dataset.
What it means
A story can be widely shared without becoming a source that other websites link to. Social engagement and backlinks are different outcomes, so an agency should avoid using one as proof of the other.
Try this
- Keep social shares, referral visits and new referring domains on separate lines in the campaign report.
- Study the specific facts, tools or explanations that earned citations, not just the topics that attracted reactions.
Use with care: Observed pattern in the stated sample. It does not isolate what caused the result.
Read the original study: Backlinko / BuzzSumo (opens in a new tab)See how the study was done for finding 47
- Research
- We Analyzed 912 Million Blog Posts. Here’s What We Learned About Content Marketing
- How much was studied
- 912.5 million articles
- Who or what was studied
- Articles in the BuzzSumo index
- When the data came from
- Articles published 1 October 2017–1 October 2018
- Where the number appears
- No correlation between social shares and backlinks
- Keep in mind
- Historical, observational data. Links are unique referring domains in the methods. Means can be skewed by outliers; formats were classified automatically.
Reference S112 · Original source checked 9 September 2026
“Why” and “what” articles were useful reference formats
25.8% more linking domains
Why posts, what posts and infographics averaged 25.8% more links than video and how-to posts in that historical sample.
What it means
The formats associated with more links were “why” articles, “what” articles and infographics. They outperformed how-to articles and videos on average in this dataset. That does not mean you should change a useful tutorial into an unsuitable format.
Try this
- Choose the format that answers the reader’s actual question: an explanation, definition, visual comparison or instructions.
- Make the specific evidence people might cite easy to find. Compare outcomes with similar content and account for promotion differences.
Use with care: Observed pattern in the stated sample. It does not isolate what caused the result.
Read the original study: Backlinko / BuzzSumo (opens in a new tab)See how the study was done for finding 48
- Research
- We Analyzed 912 Million Blog Posts. Here’s What We Learned About Content Marketing
- How much was studied
- 912.5 million articles
- Who or what was studied
- Articles in the BuzzSumo index
- When the data came from
- Articles published 1 October 2017–1 October 2018
- Where the number appears
- Content formats and backlinks
- Keep in mind
- Historical, observational data. Links are unique referring domains in the methods. Means can be skewed by outliers; formats were classified automatically. Grouped categories; this does not estimate the effect of changing the format of one article.
Reference S113 · Original source checked 9 September 2026
A statistics page earned links in one campaign
515 emails → 32 linking sites
Ahrefs’ 2020 statistics-page campaign reported 36 backlinks from 32 websites after 515 outreach emails.
What it means
Ahrefs reported 32 linking websites after sending 515 outreach emails for one statistics page. The useful model is a maintained resource that solves a citation problem. An established brand and a selected successful case make the result a poor universal target.
Try this
- Find pages citing outdated or hard-to-check statistics and publish a clearer, well-sourced reference for that audience.
- Approach relevant writers with the specific correction or useful addition. Track delivered emails, useful replies and actual linking websites separately.
Use with care: One publisher’s campaign experience. Treat this as an example, not a forecast for your campaign.
Read the original study: Ahrefs (opens in a new tab)See how the study was done for finding 49
- Research
- Link Building Case Study: How We Built Backlinks With a Stats Page
- How much was studied
- 515 emails
- Who or what was studied
- One Ahrefs statistics-page outreach campaign
- When the data came from
- Campaign reported August 2020; precise outreach window undisclosed
- Where the number appears
- Campaign results in the introduction
- Keep in mind
- One successful, publisher-selected case with an established brand. No control group; the results are not a general outreach forecast.
Reference S114 · Original source checked 9 September 2026
Reports were rated ahead of interactives
56% vs 42%
In Aira’s 2022 survey, 56% considered reports and long-form content effective, versus 42% for interactive content.
What it means
Respondents rated research reports highly as linkable assets. A report works when its evidence is genuinely useful.
Try this
- Start with a report when original evidence is the main value.
- Add clear charts, tables and summaries that publishers can reuse accurately.
Use with care: Self-reported tactics or outcomes. Campaigns, audiences and promotion were not held equal.
Read the original study: Aira (opens in a new tab)See how the study was done for finding 50
- Research
- 11 Takeaways from the 2022 State of Link Building Report
- How much was studied
- 270 respondents; branching questions have different denominators
- Who or what was studied
- Self-selected SEO professionals answering the annual link-building survey
- When the data came from
- 16 March–2 May 2022
- Where the number appears
- Takeaway 11; PDF page 47: effective content formats
- Keep in mind
- Historical self-reported practice and opinions. Multiple-choice questions may allow several answers; responses are not measured ranking effects. Practitioner opinions about effectiveness, not measured link yields.
Reference S119 · Original source checked 9 September 2026
Link earning per visit varied by industry
14.4 vs 2.5 domains
Siege’s 2024–2025 real estate/property/home-services group averaged 14.4 new referring domains per 10,000 organic visits, versus 2.5 for B2B SaaS.
What it means
New referring domains per 10,000 search visits varied across industries in this selected sample. A ratio lets different-sized sites be compared, but it can rise because links increased or because visits fell. Check both numbers before calling it progress.
Try this
- Compare new referring domains and organic visits over the same period, using consistent tools and definitions.
- Choose similar businesses for context, then inspect which pages earn links. Keep the raw link and visit counts visible beside the ratio.
Use with care: Observed pattern in the stated sample. It does not isolate what caused the result.
Read the original study: Siege Media (opens in a new tab)See how the study was done for finding 51
- Research
- Organic Link Ratios by Industry: What 100 Sites Reveal About Link Acquisition
- How much was studied
- 100 sites; ten per industry
- Who or what was studied
- Sites selected for organic prominence across ten industries
- When the data came from
- January 2024–December 2025
- Where the number appears
- Industry comparison table; two-year averages
- Keep in mind
- Organic link ratio means new referring domains per 10,000 organic visits. Selected successful sites are not a random sample. Falling traffic can raise the ratio without an increase in referring domains.
Reference S123 · Original source checked 9 September 2026
Zillow’s link-to-traffic ratio increased
31.9 → 53.9
Zillow’s reported organic link ratio rose from 31.9 in 2024 to 53.9 in 2025 in Siege’s analysis.
What it means
Zillow’s ratio rose, but that can happen because links grow, traffic falls, or both.
Try this
- Show new linking websites and search visits as separate lines.
- Find which pages earned links before copying a single company’s result.
Use with care: Observed pattern in the stated sample. It does not isolate what caused the result.
Read the original study: Siege Media (opens in a new tab)See how the study was done for finding 52
- Research
- Organic Link Ratios by Industry: What 100 Sites Reveal About Link Acquisition
- How much was studied
- One website within a 100-site study
- Who or what was studied
- Zillow in the 2024–2025 study
- When the data came from
- January 2024–December 2025
- Where the number appears
- Real estate: Zillow
- Keep in mind
- Organic link ratio means new referring domains per 10,000 organic visits. Selected successful sites are not a random sample. Falling traffic can raise the ratio without an increase in referring domains. Units: new referring domains per 10,000 organic visits; not a raw backlink growth rate.
Reference S124 · Original source checked 9 September 2026
Topic 06 12 findings
Outreach and follow-up
Reach the right people with a specific contribution. Separate replies, interested replies, coverage and earned links.
Put this section to work: Test one clear story with a small relevant group. Record delivered emails and useful replies before expanding the list. Stop contacting someone who declines.
Media-list data often needed correction
62.8% of respondents
In BuzzStream’s 2026 survey, 62.8% named outdated or incorrect journalist information as their main challenge with media-list tools.
What it means
Contact databases often contained wrong or old information. Check the person before you send.
Try this
- Check each important contact’s latest articles and current employer before emailing.
- Remove or update contacts you cannot verify.
Use with care: Survey answers describe this group of respondents. They do not prove a tactic improves rankings.
Read the original study: BuzzStream (opens in a new tab)See how the study was done for finding 53
- Research
- State of Digital PR Report (2026)
- How much was studied
- 150+ professionals
- Who or what was studied
- Digital PR professionals responding to the publisher’s survey
- When the data came from
- not reported; 2026 release
- Where the number appears
- Media List Building, question 8
- Keep in mind
- Self-reported vendor survey, not representative of all campaigns. Costs, success definitions and times to results differ across respondents; fieldwork dates undisclosed.
Reference S027 · Original source checked 9 September 2026
Most outreach emails received no reply
8.5% reply rate
Backlinko and Pitchbox’s 2019 analysis of 12 million outreach emails found an 8.5% reply rate.
What it means
The historical study counted replies, not links. An 8.5% reply rate would mean roughly eight or nine replies from 100 delivered emails if your campaign matched it; some replies would still be rejections. That arithmetic is an illustration, not a forecast.
Try this
- Measure positive replies separately from all replies, and count the referring domains you actually earn.
- Test a relevant story with a small group before expanding outreach. Compare results with your own similar campaigns.
Use with care: Observed pattern in the stated sample. It does not isolate what caused the result.
Read the original study: Backlinko and Pitchbox (opens in a new tab)See how the study was done for finding 54
- Research
- We Analyzed 12 Million Outreach Emails. Here’s What We Learned
- How much was studied
- 12 million outreach emails
- Who or what was studied
- Outreach emails supplied through Pitchbox
- When the data came from
- not reported; 2019 study
- Where the number appears
- Summary and most outreach emails are ignored
- Keep in mind
- Historical, platform-specific observational dataset. Reply is not equivalent to a positive reply, earned link or sale. Personalization findings are associations and relative changes.
Reference S035 · Original source checked 9 September 2026
Personalized subjects correlated with replies
30.5% higher response rate
In Backlinko and Pitchbox’s 2019 dataset, personalized subject lines were associated with a 30.5% higher response rate.
What it means
A 30.5% relative increase is different from adding 30.5 percentage points. If a reply rate were 5%, that relative improvement would make it about 6.5%. This example explains the math; the study does not predict your campaign’s rate.
Try this
- Write a subject line that shows why the finding fits the recipient’s beat, rather than adding a first name alone.
- Compare useful replies from similar groups and keep the story and sending approach consistent when testing the subject.
Use with care: Observed pattern in the stated sample. It does not isolate what caused the result.
Read the original study: Backlinko and Pitchbox (opens in a new tab)See how the study was done for finding 55
- Research
- We Analyzed 12 Million Outreach Emails. Here’s What We Learned
- How much was studied
- 12 million outreach emails
- Who or what was studied
- Outreach emails supplied through Pitchbox
- When the data came from
- not reported; 2019 study
- Where the number appears
- Personalized subject lines section
- Keep in mind
- Historical, platform-specific observational dataset. Reply is not equivalent to a positive reply, earned link or sale. Personalization findings are associations and relative changes.
Reference S036 · Original source checked 9 September 2026
Personalized bodies also correlated with replies
32.7% higher response rate
In Backlinko and Pitchbox’s 2019 dataset, personalized email bodies were associated with a 32.7% higher response rate.
What it means
Messages with a personalized body had more replies in the observed dataset. Personalization and better prospect selection may occur together. The practical aim is to show why this evidence belongs in this person’s work.
Try this
- Reference a relevant topic the writer actually covers, then explain what your evidence adds.
- Keep the ask clear and the source easy to check. Track interested replies and published links rather than treating any reply as success.
Use with care: Observed pattern in the stated sample. It does not isolate what caused the result.
Read the original study: Backlinko and Pitchbox (opens in a new tab)See how the study was done for finding 56
- Research
- We Analyzed 12 Million Outreach Emails. Here’s What We Learned
- How much was studied
- 12 million outreach emails
- Who or what was studied
- Outreach emails supplied through Pitchbox
- When the data came from
- not reported; 2019 study
- Where the number appears
- Personalized email body section
- Keep in mind
- Historical, platform-specific observational dataset. Reply is not equivalent to a positive reply, earned link or sale. Personalization findings are associations and relative changes.
Reference S037 · Original source checked 9 September 2026
First coverage can arrive quickly
81% of respondents
In BuzzStream’s 2026 digital PR survey, 81% typically saw their first coverage within a week of pitching.
What it means
Most surveyed teams saw their first coverage within a week. Search and business results usually take longer.
Try this
- Record how long each campaign takes to earn its first coverage.
- Treat early coverage as feedback, not proof of final SEO or business impact.
Use with care: Survey answers describe this group of respondents. They do not prove a tactic improves rankings.
Read the original study: BuzzStream (opens in a new tab)See how the study was done for finding 57
- Research
- State of Digital PR Report (2026)
- How much was studied
- 150+ professionals
- Who or what was studied
- Digital PR professionals responding to the publisher’s survey
- When the data came from
- not reported; 2026 release
- Where the number appears
- Time to first coverage
- Keep in mind
- Self-reported vendor survey, not representative of all campaigns. Costs, success definitions and times to results differ across respondents; fieldwork dates undisclosed. Self-reported typical timing; not a campaign guarantee.
Reference S095 · Original source checked 9 September 2026
Personalization is not universal
48% of respondents
In BuzzStream’s 2026 digital PR survey, 48% said they always personalized pitches.
What it means
Fewer than half always personalised pitches. A copied message is easy for a journalist to ignore.
Try this
- Add one real, person-specific reason to every pitch.
- Review sent emails and ignore simple name changes as personalisation.
Use with care: Survey answers describe this group of respondents. They do not prove a tactic improves rankings.
Read the original study: BuzzStream (opens in a new tab)See how the study was done for finding 58
- Research
- State of Digital PR Report (2026)
- How much was studied
- 150+ professionals
- Who or what was studied
- Digital PR professionals responding to the publisher’s survey
- When the data came from
- not reported; 2026 release
- Where the number appears
- Pitch personalization
- Keep in mind
- Self-reported vendor survey, not representative of all campaigns. Costs, success definitions and times to results differ across respondents; fieldwork dates undisclosed.
Reference S096 · Original source checked 9 September 2026
Contact checks are easy to skip
53.4% of respondents
In BuzzStream’s 2026 digital PR survey, 53.4% always checked that a journalist still worked at the publication.
What it means
Only about half always checked whether the journalist still worked there. A correct email can still reach the wrong person.
Try this
- Check the person’s employer, topic area and latest article before sending.
- Save the check date and remove old contacts.
Use with care: Survey answers describe this group of respondents. They do not prove a tactic improves rankings.
Read the original study: BuzzStream (opens in a new tab)See how the study was done for finding 59
- Research
- State of Digital PR Report (2026)
- How much was studied
- 150+ professionals
- Who or what was studied
- Digital PR professionals responding to the publisher’s survey
- When the data came from
- not reported; 2026 release
- Where the number appears
- Research before pitching
- Keep in mind
- Self-reported vendor survey, not representative of all campaigns. Costs, success definitions and times to results differ across respondents; fieldwork dates undisclosed.
Reference S097 · Original source checked 9 September 2026
One follow-up was associated with more replies
65.8% more replies
Backlinko and Pitchbox’s 2019 analysis associated one additional follow-up with 65.8% more replies.
What it means
One follow-up was linked with more replies. A helpful reminder can work, while repeated pressure can backfire.
Try this
- Follow up once only when you can add useful new information.
- Stop after a rejection, and track complaints and unsubscribe requests.
Use with care: Observed pattern in the stated sample. It does not isolate what caused the result.
Read the original study: Backlinko and Pitchbox (opens in a new tab)See how the study was done for finding 60
- Research
- We Analyzed 12 Million Outreach Emails. Here’s What We Learned
- How much was studied
- 12 million outreach emails
- Who or what was studied
- Outreach emails supplied through Pitchbox
- When the data came from
- not reported; 2019 study
- Where the number appears
- Following up significantly boosts response rates
- Keep in mind
- Historical, platform-specific observational dataset. Reply is not equivalent to a positive reply, earned link or sale. Personalization findings are associations and relative changes. Relative increase, not percentage points; observational comparison.
Reference S107 · Original source checked 9 September 2026
Finding the right contact can matter
93% higher reply rate
In the 2019 outreach dataset, contacting multiple people at an organization was associated with a 93% higher reply rate.
What it means
Contacting more than one relevant person was linked with more replies. It should still be done carefully.
Try this
- Choose the writer or editor whose recent work best fits the story.
- Keep one backup and contact them only if needed.
Use with care: Observed pattern in the stated sample. It does not isolate what caused the result.
Read the original study: Backlinko and Pitchbox (opens in a new tab)See how the study was done for finding 61
- Research
- We Analyzed 12 Million Outreach Emails. Here’s What We Learned
- How much was studied
- 12 million outreach emails
- Who or what was studied
- Outreach emails supplied through Pitchbox
- When the data came from
- not reported; 2019 study
- Where the number appears
- Multiple contacts at the same organization
- Keep in mind
- Historical, platform-specific observational dataset. Reply is not equivalent to a positive reply, earned link or sale. Personalization findings are associations and relative changes. Relative increase; contact choice and campaign quality may confound the comparison.
Reference S108 · Original source checked 9 September 2026
Social-profile links in signatures correlated with replies
8.2% / 11.5% / 23.4%
In the 2019 dataset, linked Twitter, LinkedIn and Instagram profiles were associated with 8.2%, 11.5% and 23.4% higher reply rates respectively.
What it means
The study compared signatures with and without social-profile links. Replies were more frequent when profiles were included, with different associations for Twitter, LinkedIn and Instagram. It did not test named senders against anonymous addresses.
Try this
- Include a relevant, accurate professional profile in your signature when it helps a recipient verify who is contacting them.
- Keep the message useful and easy to answer. Treat social-profile links as context for credibility, not as a promised reply-rate improvement.
Use with care: Relative differences associated with profile links, not absolute reply rates or proof of causation.
Read the original study: Backlinko and Pitchbox (opens in a new tab)See how the study was done for finding 62
- Research
- We Analyzed 12 Million Outreach Emails. Here’s What We Learned
- How much was studied
- 12 million outreach emails
- Who or what was studied
- Outreach emails supplied through Pitchbox
- When the data came from
- not reported; 2019 study
- Where the number appears
- Social profile links in email signatures
- Keep in mind
- Historical, platform-specific observational dataset. Reply is not equivalent to a positive reply, earned link or sale. Personalization findings are associations and relative changes. Relative associations, not percentage-point increases or controlled effects.
Reference S109 · Original source checked 9 September 2026
Question-style subjects had fewer opens
13% lower open rate
Reboot’s undated analysis associated question-style subject lines with a 13% lower open rate.
What it means
Question-style subject lines were opened less often in this sample. A direct subject may be easier to understand.
Try this
- Replace vague questions with a direct description of the news or finding.
- Compare similar groups and judge useful replies, not only opens.
Use with care: Observed pattern in the stated sample. It does not isolate what caused the result.
Read the original study: Reboot Online (opens in a new tab)See how the study was done for finding 63
- Research
- Digital PR Statistics: email subject-line analysis
- How much was studied
- More than 1,000 subject lines
- Who or what was studied
- Subject lines from Reboot’s own outreach analysis
- When the data came from
- Collection period not reported
- Where the number appears
- Email subject-line analysis
- Keep in mind
- Observational agency data with limited method disclosure. Opens are not replies or placements and can be distorted by mail-client privacy features. Relative difference; campaign assignment was not randomized.
Reference S121 · Original source checked 9 September 2026
Event and celebrity references correlated with opens
12% higher open rate
Reboot associated subjects containing buzzwords, such as celebrity names, events or special dates, with a 12% higher open rate.
What it means
Subjects mentioning celebrity names, events or special dates had higher open rates in this agency sample. These are the source’s “buzzwords.” A timely reference is useful only when it connects honestly to the story.
Try this
- Use an event or date in the subject only when it is central to the evidence you are offering.
- Judge useful replies and coverage as well as opens. Email privacy features can make open-rate comparisons unreliable.
Use with care: Observed pattern in the stated sample. It does not isolate what caused the result.
Read the original study: Reboot Online (opens in a new tab)See how the study was done for finding 64
- Research
- Digital PR Statistics: email subject-line analysis
- How much was studied
- More than 1,000 subject lines
- Who or what was studied
- Subject lines from Reboot’s own outreach analysis
- When the data came from
- Collection period not reported
- Where the number appears
- Email subject-line analysis: buzzwords
- Keep in mind
- Observational agency data with limited method disclosure. Opens are not replies or placements and can be distorted by mail-client privacy features. Relative difference; the term grouping and agency sample limit generalization.
Reference S122 · Original source checked 9 September 2026
Topic 07 14 findings
Earning coverage journalists can use
Make the audience connection and the evidence easy to check. These surveys describe reporting needs, not ranking effects.
Put this section to work: Put the finding, affected audience and source in the opening lines. Provide a short methods note and an available expert who can explain the limits.
Some inboxes receive heavy daily volume
14% received over 20 pitches
In Muck Rack’s January–March 2026 journalist survey, 14% said they received more than 20 PR pitches on a typical day.
What it means
Some journalists receive more than 20 pitches every day. They need to understand the story quickly.
Try this
- State the finding, affected audience and why it matters in the first two sentences.
- Make the story clear without requiring an attachment.
Use with care: Preferences reported by survey respondents, not a measured probability of coverage or an earned link.
Read the original study: Muck Rack (opens in a new tab)See how the study was done for finding 65
- Research
- The State of Journalism 2026
- How much was studied
- 897 retained responses from 1,044 collected
- Who or what was studied
- Journalists recruited by email from Muck Rack’s database; mostly US-based
- When the data came from
- 2026-01-30 to 2026-03-02
- Where the number appears
- PDF page 28, daily pitch volume
- Keep in mind
- Non-probability, self-administered survey; not representative of every newsroom. Question-specific response counts are not always shown. Use retained sample, not the larger collection count.
Reference S028 · Original source checked 9 September 2026
Relevant pitches were scarce for many writers
43% seldom; 4% never
In Muck Rack’s January–March 2026 survey, 43% of journalists seldom received relevant pitches, and 4% said they never did.
What it means
Many journalists said most pitches did not fit their work. A shorter, better-targeted list can beat a larger one.
Try this
- Read three recent articles to understand the journalist’s topics and audience.
- Send only when your evidence clearly fits their work.
Use with care: Preferences reported by survey respondents, not a measured probability of coverage or an earned link.
Read the original study: Muck Rack (opens in a new tab)See how the study was done for finding 66
- Research
- The State of Journalism 2026
- How much was studied
- 897 retained responses from 1,044 collected
- Who or what was studied
- Journalists recruited by email from Muck Rack’s database; mostly US-based
- When the data came from
- 2026-01-30 to 2026-03-02
- Where the number appears
- PDF page 29, pitch relevance frequency
- Keep in mind
- Non-probability, self-administered survey; not representative of every newsroom. Question-specific response counts are not always shown. Use retained sample, not the larger collection count.
Reference S029 · Original source checked 9 September 2026
Pitches still contributed to published stories
86% of journalists
In Muck Rack’s January–March 2026 survey, 86% of journalists said at least some of their stories began with a PR pitch.
What it means
Most journalists had used at least one pitch in a story. Good outreach can help when it provides useful information.
Try this
- Send a checkable finding, a short expert quote and the supporting source.
- Record which parts journalists actually use in their coverage.
Use with care: Preferences reported by survey respondents, not a measured probability of coverage or an earned link.
Read the original study: Muck Rack (opens in a new tab)See how the study was done for finding 67
- Research
- The State of Journalism 2026
- How much was studied
- 897 retained responses from 1,044 collected
- Who or what was studied
- Journalists recruited by email from Muck Rack’s database; mostly US-based
- When the data came from
- 2026-01-30 to 2026-03-02
- Where the number appears
- PDF page 30, stories originating from pitches
- Keep in mind
- Non-probability, self-administered survey; not representative of every newsroom. Question-specific response counts are not always shown. Use retained sample, not the larger collection count.
Reference S030 · Original source checked 9 September 2026
Original evidence was welcome
40% of journalists
In Muck Rack’s January–March 2026 survey, 40% of journalists wanted original data or research included with a story pitch.
What it means
Original data gives a journalist something new to report. It becomes usable when the writer can understand who was studied, when the measurement happened and what the result leaves out. A number without that context creates more work.
Try this
- Publish a short methods note beside the finding, including the sample, period, definitions and limitations.
- Offer a simple downloadable table and a real person who can answer questions. Describe a compilation of other studies as a compilation, not your own new survey.
Use with care: Preferences reported by survey respondents, not a measured probability of coverage or an earned link.
Read the original study: Muck Rack (opens in a new tab)See how the study was done for finding 68
- Research
- The State of Journalism 2026
- How much was studied
- 897 retained responses from 1,044 collected
- Who or what was studied
- Journalists recruited by email from Muck Rack’s database; mostly US-based
- When the data came from
- 2026-01-30 to 2026-03-02
- Where the number appears
- PDF page 31, ideal pitch assets
- Keep in mind
- Non-probability, self-administered survey; not representative of every newsroom. Question-specific response counts are not always shown. Use retained sample, not the larger collection count.
Reference S031 · Original source checked 9 September 2026
Audience impact made a story relevant
78% of journalists
In Muck Rack’s January–March 2026 survey, 78% said direct effects on their audience’s community made a pitch relevant.
What it means
Journalists cared most about how a story affects their audience. Relevance should be obvious in the pitch.
Try this
- State who is affected, what changes and how large the change is.
- Pitch only publications whose readers have a clear reason to care.
Use with care: Preferences reported by survey respondents, not a measured probability of coverage or an earned link.
Read the original study: Muck Rack (opens in a new tab)See how the study was done for finding 69
- Research
- The State of Journalism 2026
- How much was studied
- 897 retained responses from 1,044 collected
- Who or what was studied
- Journalists recruited by email from Muck Rack’s database; mostly US-based
- When the data came from
- 2026-01-30 to 2026-03-02
- Where the number appears
- PDF page 33, audience relevance
- Keep in mind
- Non-probability, self-administered survey; not representative of every newsroom. Question-specific response counts are not always shown. Use retained sample, not the larger collection count.
Reference S032 · Original source checked 9 September 2026
An irrelevant pitch was easily rejected
82% of journalists
In Cision’s 2026 survey of more than 1,800 journalists, 82% identified irrelevance to their audience or beat as a reason to reject a pitch.
What it means
Irrelevant pitches were the most common reason for rejection. Sending to the wrong person wastes both sides’ time.
Try this
- Check topic fit, reader value and story format before approving a pitch.
- Do not send when you cannot point to a relevant recent article.
Use with care: Preferences reported by survey respondents, not a measured probability of coverage or an earned link.
Read the original study: Cision (opens in a new tab)See how the study was done for finding 70
- Research
- Why Do Journalists Ignore Your PR Pitch?
- How much was studied
- more than 1,800 journalists across 19 markets
- Who or what was studied
- Journalists across industries in the global State of the Media survey
- When the data came from
- not reported; 2026 survey release
- Where the number appears
- 1. Make your pitch relevant
- Keep in mind
- Original publisher’s public explanation of its own survey. Fieldwork dates and exact question-specific denominator not disclosed on this page; self-reported rejection reasons, not observed delivery outcomes. Full report landing page requires a form, so selected findings use accessible original-publisher reporting.
Reference S081 · Original source checked 9 September 2026
A pitch still needs a news angle
34% of journalists
In Cision’s 2026 journalist survey, 34% said they would reject a pitch without a clear news angle or point of view.
What it means
A third rejected pitches with no clear news. Data still needs a reason to matter now.
Try this
- Answer three questions: What is new? Why now? Why does it matter?
- Improve the story before adding more contacts if the answers are weak.
Use with care: Preferences reported by survey respondents, not a measured probability of coverage or an earned link.
Read the original study: Cision (opens in a new tab)See how the study was done for finding 71
- Research
- Why Do Journalists Ignore Your PR Pitch?
- How much was studied
- more than 1,800 journalists across 19 markets
- Who or what was studied
- Journalists across industries in the global State of the Media survey
- When the data came from
- not reported; 2026 survey release
- Where the number appears
- 6. Bring a clear news angle or point of view
- Keep in mind
- Original publisher’s public explanation of its own survey. Fieldwork dates and exact question-specific denominator not disclosed on this page; self-reported rejection reasons, not observed delivery outcomes. Full report landing page requires a form, so selected findings use accessible original-publisher reporting.
Reference S082 · Original source checked 9 September 2026
Some journalists face a large weekly queue
29% received over 50 pitches
In Cision’s 2026 survey, 29% of journalists received more than 50 pitches per week.
What it means
Some journalists receive hundreds of pitches a week. Your message must be easy to scan.
Try this
- Start with the small group of journalists who fit the story best.
- Use their questions to improve the pitch before sending it more widely.
Use with care: Preferences reported by survey respondents, not a measured probability of coverage or an earned link.
Read the original study: Cision (opens in a new tab)See how the study was done for finding 72
- Research
- Why Do Journalists Ignore Your PR Pitch?
- How much was studied
- more than 1,800 journalists across 19 markets
- Who or what was studied
- Journalists across industries in the global State of the Media survey
- When the data came from
- not reported; 2026 survey release
- Where the number appears
- Pitch volume
- Keep in mind
- Original publisher’s public explanation of its own survey. Fieldwork dates and exact question-specific denominator not disclosed on this page; self-reported rejection reasons, not observed delivery outcomes. Full report landing page requires a form, so selected findings use accessible original-publisher reporting.
Reference S129 · Original source checked 9 September 2026
Most pitches missed the mark for many recipients
72% of journalists
In Cision’s 2026 survey, 72% said fewer than a quarter of the pitches they received were relevant.
What it means
Many said most pitches missed the mark. Better targeting matters more than sending more.
Try this
- Sort missed pitches by wrong topic, audience, timing, format or evidence.
- Fix repeated targeting mistakes before adding more contacts.
Use with care: Preferences reported by survey respondents, not a measured probability of coverage or an earned link.
Read the original study: Cision (opens in a new tab)See how the study was done for finding 73
- Research
- Why Do Journalists Ignore Your PR Pitch?
- How much was studied
- more than 1,800 journalists across 19 markets
- Who or what was studied
- Journalists across industries in the global State of the Media survey
- When the data came from
- not reported; 2026 survey release
- Where the number appears
- Pitch relevance
- Keep in mind
- Original publisher’s public explanation of its own survey. Fieldwork dates and exact question-specific denominator not disclosed on this page; self-reported rejection reasons, not observed delivery outcomes. Full report landing page requires a form, so selected findings use accessible original-publisher reporting.
Reference S130 · Original source checked 9 September 2026
A promotional angle can lose the story
53% of journalists
In Cision’s 2026 survey, 53% would reject a pitch for being overly promotional.
What it means
A sales message is not automatically a news story. Lead with the useful finding.
Try this
- Lead with the useful finding or event, then explain the company’s role.
- Remove sales claims that do not help a journalist report the story.
Use with care: Preferences reported by survey respondents, not a measured probability of coverage or an earned link.
Read the original study: Cision (opens in a new tab)See how the study was done for finding 74
- Research
- Why Do Journalists Ignore Your PR Pitch?
- How much was studied
- more than 1,800 journalists across 19 markets
- Who or what was studied
- Journalists across industries in the global State of the Media survey
- When the data came from
- not reported; 2026 survey release
- Where the number appears
- Reasons to reject a pitch
- Keep in mind
- Original publisher’s public explanation of its own survey. Fieldwork dates and exact question-specific denominator not disclosed on this page; self-reported rejection reasons, not observed delivery outcomes. Full report landing page requires a form, so selected findings use accessible original-publisher reporting.
Reference S131 · Original source checked 9 September 2026
PR relationships still have value
53% of journalists
In Muck Rack’s 2026 survey, 53% rated relationships with PR professionals important or very important.
What it means
More than half valued relationships with PR people. Reliable help over time can make future contact easier.
Try this
- Keep short notes on each contact’s topics, preferences and past conversations.
- Answer requests quickly and accurately, including between campaigns.
Use with care: Preferences reported by survey respondents, not a measured probability of coverage or an earned link.
Read the original study: Muck Rack (opens in a new tab)See how the study was done for finding 75
- Research
- The State of Journalism 2026
- How much was studied
- 897 retained responses from 1,044 collected
- Who or what was studied
- Journalists recruited by email from Muck Rack’s database; mostly US-based
- When the data came from
- 2026-01-30 to 2026-03-02
- Where the number appears
- PDF page 27, printed page 26: PR relationships
- Keep in mind
- Non-probability, self-administered survey; not representative of every newsroom. Question-specific response counts are not always shown. Use retained sample, not the larger collection count.
Reference S132 · Original source checked 9 September 2026
A non-response is common
54% of journalists
In Muck Rack’s 2026 survey, 54% said they seldom or never replied to pitches.
What it means
More than half rarely or never reply. No answer does not always mean the email failed to arrive.
Try this
- Expect some silence and limit the number of follow-ups.
- Check the story and audience fit before sending more emails.
Use with care: Preferences reported by survey respondents, not a measured probability of coverage or an earned link.
Read the original study: Muck Rack (opens in a new tab)See how the study was done for finding 76
- Research
- The State of Journalism 2026
- How much was studied
- 897 retained responses from 1,044 collected
- Who or what was studied
- Journalists recruited by email from Muck Rack’s database; mostly US-based
- When the data came from
- 2026-01-30 to 2026-03-02
- Where the number appears
- PDF page 28: replies to pitches
- Keep in mind
- Non-probability, self-administered survey; not representative of every newsroom. Question-specific response counts are not always shown. Use retained sample, not the larger collection count.
Reference S133 · Original source checked 9 September 2026
Access to a person helps the reporting
58% of journalists
In Muck Rack’s 2026 survey, 58% wanted access to relevant interviewees with a pitch.
What it means
Many wanted access to someone they could interview. A real expert can make the evidence easier to use.
Try this
- Offer a real expert and explain what they can add.
- Prepare them to answer questions about the method and limits.
Use with care: Preferences reported by survey respondents, not a measured probability of coverage or an earned link.
Read the original study: Muck Rack (opens in a new tab)See how the study was done for finding 77
- Research
- The State of Journalism 2026
- How much was studied
- 897 retained responses from 1,044 collected
- Who or what was studied
- Journalists recruited by email from Muck Rack’s database; mostly US-based
- When the data came from
- 2026-01-30 to 2026-03-02
- Where the number appears
- PDF page 32, printed page 31: what to include
- Keep in mind
- Non-probability, self-administered survey; not representative of every newsroom. Question-specific response counts are not always shown. Use retained sample, not the larger collection count.
Reference S134 · Original source checked 9 September 2026
Repeated pitches can get deleted
46% of journalists
In Muck Rack’s 2026 survey, 46% would delete pitches sent repeatedly without a response.
What it means
Nearly half would delete repeated pitches. A follow-up should add something new.
Try this
- Stop after one useful follow-up unless you have genuinely new information.
- Record the result so another teammate does not resend the same pitch.
Use with care: Preferences reported by survey respondents, not a measured probability of coverage or an earned link.
Read the original study: Muck Rack (opens in a new tab)See how the study was done for finding 78
- Research
- The State of Journalism 2026
- How much was studied
- 897 retained responses from 1,044 collected
- Who or what was studied
- Journalists recruited by email from Muck Rack’s database; mostly US-based
- When the data came from
- 2026-01-30 to 2026-03-02
- Where the number appears
- PDF page 33: reasons to reject or delete
- Keep in mind
- Non-probability, self-administered survey; not representative of every newsroom. Question-specific response counts are not always shown. Use retained sample, not the larger collection count.
Reference S135 · Original source checked 9 September 2026
Topic 08 7 findings
Costs and budgets
Compare like-for-like work. A marketplace quote, a monthly retainer and a cost per accepted link are different measures.
Put this section to work: Agree accepted link types and included work. Divide the full campaign cost by accepted links, then by links still live at your review date.
Many teams did not know cost per link
19.6% at $300–$500; 39.2% unsure
In BuzzStream’s 2026 digital PR survey, 19.6% reported average link costs of $300–$500, while 39.2% did not know their cost per link.
What it means
A cost-per-link figure is hard to compare when many teams cannot calculate their own. The result changes if one team counts only fees and another includes research, writing and staff time.
Try this
- Add all campaign costs and define which live links meet the brief before dividing one by the other.
- Recalculate using links still live at the review date, and show the original cost and both counts beside the result.
Use with care: Self-reported provider costs and terms. Included work and definitions vary; this is not a quote for your brief.
Read the original study: BuzzStream (opens in a new tab)See how the study was done for finding 79
- Research
- State of Digital PR Report (2026)
- How much was studied
- 150+ professionals
- Who or what was studied
- Digital PR professionals responding to the publisher’s survey
- When the data came from
- not reported; 2026 release
- Where the number appears
- Digital PR Budgeting, question 3
- Keep in mind
- Self-reported vendor survey, not representative of all campaigns. Costs, success definitions and times to results differ across respondents; fieldwork dates undisclosed.
Reference S024 · Original source checked 9 September 2026
A retainer includes more than a link
US$5,458 per month
BuzzStream’s 2025 survey of PR providers reported a mean monthly digital PR contract of $5,458.
What it means
The average monthly fee covered a complete digital PR service, not just one link.
Try this
- Ask what research, content, outreach and reporting the monthly fee includes.
- Compare providers by the work included, not only the price.
Use with care: Self-reported provider costs and terms. Included work and definitions vary; this is not a quote for your brief.
Read the original study: BuzzStream (opens in a new tab)See how the study was done for finding 80
- Research
- How Much Does Digital PR Cost?
- How much was studied
- About 70 respondents
- Who or what was studied
- Agency and freelance digital PR leaders, predominantly UK and US
- When the data came from
- 2025 survey; fieldwork dates not disclosed; page updated June 2026
- Where the number appears
- Average monthly contract
- Keep in mind
- Self-reported service costs, not invoices. UK/US-heavy convenience sample; item-level response counts are undisclosed. Scope varies between retainers.
Reference S098 · Original source checked 9 September 2026
Cost per link needs a definition
US$597 per link
BuzzStream’s 2025 provider survey reported a $597 mean cost per link; roughly one-third did not know their figure.
What it means
The reported average is a survey benchmark, not a quote for your campaign. A provider producing a major study and one arranging simple coverage can have very different costs. Compare the deliverable and quality before the unit price.
Try this
- Ask for the included work, accepted placement types and reporting period behind any cost-per-link claim.
- For your campaign, show total spend divided by accepted links and by links still live. The article’s cost calculator uses your inputs, not the survey average.
Use with care: Self-reported provider costs and terms. Included work and definitions vary; this is not a quote for your brief.
Read the original study: BuzzStream (opens in a new tab)See how the study was done for finding 81
- Research
- How Much Does Digital PR Cost?
- How much was studied
- About 70 respondents
- Who or what was studied
- Agency and freelance digital PR leaders, predominantly UK and US
- When the data came from
- 2025 survey; fieldwork dates not disclosed; page updated June 2026
- Where the number appears
- Average cost per link
- Keep in mind
- Self-reported service costs, not invoices. UK/US-heavy convenience sample; item-level response counts are undisclosed. Scope varies between retainers. The mean reflects respondents able to give a cost; the exact denominator is not reported.
Reference S099 · Original source checked 9 September 2026
Provider type changes the quoted scope
US$6,357 vs US$4,200
In that 2025 survey, mean monthly retainers were $6,357 for agencies and $4,200 for freelancers.
What it means
Agencies charged more on average than freelancers, but the work and capacity may not have been the same.
Try this
- Give agencies and freelancers the same brief and ask what each includes.
- Compare the people, work and expected results beside the price.
Use with care: Self-reported provider costs and terms. Included work and definitions vary; this is not a quote for your brief.
Read the original study: BuzzStream (opens in a new tab)See how the study was done for finding 82
- Research
- How Much Does Digital PR Cost?
- How much was studied
- About 70 respondents
- Who or what was studied
- Agency and freelance digital PR leaders, predominantly UK and US
- When the data came from
- 2025 survey; fieldwork dates not disclosed; page updated June 2026
- Where the number appears
- Agency and freelancer comparison
- Keep in mind
- Self-reported service costs, not invoices. UK/US-heavy convenience sample; item-level response counts are undisclosed. Scope varies between retainers.
Reference S100 · Original source checked 9 September 2026
Longer commitments were common
About 75% of contracts
In BuzzStream’s 2025 provider survey, about 75% of contracts lasted longer than six months.
What it means
Most contracts lasted longer than six months. Buyers need review points before the full term is over.
Try this
- Add review dates after 30, 60 and 90 days to longer contracts.
- Agree when you can change the story, audience or budget.
Use with care: Self-reported provider costs and terms. Included work and definitions vary; this is not a quote for your brief.
Read the original study: BuzzStream (opens in a new tab)See how the study was done for finding 83
- Research
- How Much Does Digital PR Cost?
- How much was studied
- About 70 respondents
- Who or what was studied
- Agency and freelance digital PR leaders, predominantly UK and US
- When the data came from
- 2025 survey; fieldwork dates not disclosed; page updated June 2026
- Where the number appears
- Contract length
- Keep in mind
- Self-reported service costs, not invoices. UK/US-heavy convenience sample; item-level response counts are undisclosed. Scope varies between retainers.
Reference S101 · Original source checked 9 September 2026
Billable coverage can include unlinked mentions
74% / 68% / 64%
Surveyed PR providers charged for nofollow links (74%), syndicated coverage (68%) and unlinked mentions (64%).
What it means
Some PR providers billed for mentions without a link and for stories republished elsewhere. These can reach readers, but they do not all add new linking websites. Agree what each result counts toward before signing a brief.
Try this
- List followed links, nofollow or sponsored links, unlinked mentions and syndicated copies as separate deliverables.
- Report unique referring domains as well as coverage URLs. Do not count several copies on one domain as several new websites.
Use with care: Self-reported provider costs and terms. Included work and definitions vary; this is not a quote for your brief.
Read the original study: BuzzStream (opens in a new tab)See how the study was done for finding 84
- Research
- How Much Does Digital PR Cost?
- How much was studied
- About 70 respondents
- Who or what was studied
- Agency and freelance digital PR leaders, predominantly UK and US
- When the data came from
- 2025 survey; fieldwork dates not disclosed; page updated June 2026
- Where the number appears
- Charging for types of coverage
- Keep in mind
- Self-reported service costs, not invoices. UK/US-heavy convenience sample; item-level response counts are undisclosed. Scope varies between retainers.
Reference S102 · Original source checked 9 September 2026
Vendor quotes differed from the lowest quote
US$461 vs US$295
BuzzStream’s June 2026 marketplace analysis found a $461 average vendor quote versus a $295 lowest-vendor-price proxy.
What it means
Different sellers quoted different prices for the same publication. The cheapest number may not include the same service.
Try this
- Group duplicate publication quotes and record who offered each price.
- Compare service, review and replacement terms before choosing the cheapest option.
Use with care: One marketplace. Listings and tool estimates do not represent every publisher or a completed transaction.
Read the original study: BuzzStream (opens in a new tab)See how the study was done for finding 85
- Research
- Cost of Guest Posts, Based on 500k Sites [Updated 2026 Data]
- How much was studied
- 257,267 sites with price, Domain Rating and estimated traffic data
- Who or what was studied
- Complete listings in one unnamed link marketplace
- When the data came from
- June 2026 release; collection dates undisclosed
- Where the number appears
- Key takeaways; Vendor vs direct pricing
- Keep in mind
- One marketplace, not the whole web. Listed quotes are not completed transactions. Traffic is an SEO-tool estimate; incomplete listings were excluded. The source calls its lowest-quote proxy “direct price”; it is not independently verified publisher-direct pricing.
Reference S103 · Original source checked 9 September 2026
Topic 09 7 findings
Agency process and measurement
Make the work easy to manage and the results easy to explain. Delivery, search changes and business contribution need separate measures.
Put this section to work: Put the target page, baseline, owner, acceptance rules and review date in the brief. Record other changes that could affect the results.
Documented processes were uncommon
32% of respondents
In BuzzStream and Citation Labs’ 2025 survey, 32% of respondents followed a documented, repeatable link-building process.
What it means
A written process makes a campaign easier to hand over and improve. The survey shows this was missing in many teams. For an agency, clear decisions matter more than having a long process document.
Try this
- Create a short checklist for approving the destination, audience, prospect, pitch, live placement and follow-up check.
- Assign an owner and an acceptance rule to each step so a colleague can pick up the work without guessing.
Use with care: Self-reported team practice or experience. It does not measure the effect of following this approach.
Read the original study: BuzzStream and Citation Labs (opens in a new tab)See how the study was done for finding 86
- Research
- Link Building Trends Report (2025)
- How much was studied
- Not disclosed
- Who or what was studied
- Agency, freelance, consultancy and in-house link-building professionals
- When the data came from
- not reported; 2025 release
- Where the number appears
- Link Building Strategy, question 3
- Keep in mind
- Self-reported, non-probability survey; full sample size not stated. The approximately 150 respondents cited in the management section are the in-house subset only.
Reference S016 · Original source checked 9 September 2026
Benchmarking had limited adoption
About one quarter
In BuzzStream and Citation Labs’ 2025 survey, about one quarter used benchmarks for their link-building results.
What it means
Your own campaign history is useful only when you count results consistently. A reply rate from a previous survey campaign may not be a fair target for a product announcement sent to a different audience.
Try this
- Keep comparable campaigns together by story type, target market, outreach method and what counted as a useful reply.
- Record delivered messages, useful replies, linking websites and retained links. Compare the same measures across those campaigns.
Use with care: Self-reported team practice or experience. It does not measure the effect of following this approach.
Read the original study: BuzzStream and Citation Labs (opens in a new tab)See how the study was done for finding 87
- Research
- Link Building Trends Report (2025)
- How much was studied
- Not disclosed
- Who or what was studied
- Agency, freelance, consultancy and in-house link-building professionals
- When the data came from
- not reported; 2025 release
- Where the number appears
- Link Building Reporting, question 12
- Keep in mind
- Self-reported, non-probability survey; full sample size not stated. The approximately 150 respondents cited in the management section are the in-house subset only.
Reference S017 · Original source checked 9 September 2026
Reports often failed to explain value
40% of respondents
In BuzzStream and Citation Labs’ 2025 survey, 40% felt their reporting effectively communicated the value of link building.
What it means
A client needs to know what changed and what to do next. A list of new links can prove delivery, but it cannot by itself prove a traffic increase or business return.
Try this
- Connect every report to the supported pages: relevant search clicks, referral visits and qualified inquiries, with a before-campaign baseline.
- Explain other changes such as page updates or seasonality. State which next decision the evidence supports and where attribution remains uncertain.
Use with care: Self-reported team practice or experience. It does not measure the effect of following this approach.
Read the original study: BuzzStream and Citation Labs (opens in a new tab)See how the study was done for finding 88
- Research
- Link Building Trends Report (2025)
- How much was studied
- Not disclosed
- Who or what was studied
- Agency, freelance, consultancy and in-house link-building professionals
- When the data came from
- not reported; 2025 release
- Where the number appears
- Link Building Reporting, question 11
- Keep in mind
- Self-reported, non-probability survey; full sample size not stated. The approximately 150 respondents cited in the management section are the in-house subset only.
Reference S018 · Original source checked 9 September 2026
Quality-link count was a common KPI
85.1% of respondents
In BuzzStream’s 2026 digital PR survey, 85.1% measured campaign success using the number of quality links.
What it means
Most respondents counted high-quality links. The measure works only when quality is clearly defined.
Try this
- Write simple rules for relevance, article quality, audience, link type and retention.
- Count only links that meet those rules and explain any rejections.
Use with care: Self-reported team practice or experience. It does not measure the effect of following this approach.
Read the original study: BuzzStream (opens in a new tab)See how the study was done for finding 89
- Research
- State of Digital PR Report (2026)
- How much was studied
- 150+ professionals
- Who or what was studied
- Digital PR professionals responding to the publisher’s survey
- When the data came from
- not reported; 2026 release
- Where the number appears
- Measuring Digital PR; ways to measure success
- Keep in mind
- Self-reported vendor survey, not representative of all campaigns. Costs, success definitions and times to results differ across respondents; fieldwork dates undisclosed.
Reference S023 · Original source checked 9 September 2026
Results often required several months
85.2% said within 6 months
In BuzzStream’s 2026 digital PR survey, 85.2% said measurable campaign results took no more than six months.
What it means
Most respondents reported seeing results within six months, but “results” was self-reported. First coverage, higher rankings and qualified leads happen on different timelines. This number cannot promise when a new link will affect your site.
Try this
- Agree separate milestones for asset completion, coverage, link checks and target-page performance reviews.
- Show the starting data and what changed at each milestone. If rankings stall, revisit intent, content, technical access and competition as well as links.
Use with care: Self-reported team practice or experience. It does not measure the effect of following this approach.
Read the original study: BuzzStream (opens in a new tab)See how the study was done for finding 90
- Research
- State of Digital PR Report (2026)
- How much was studied
- 150+ professionals
- Who or what was studied
- Digital PR professionals responding to the publisher’s survey
- When the data came from
- not reported; 2026 release
- Where the number appears
- Measuring Digital PR; time to measurable results
- Keep in mind
- Self-reported vendor survey, not representative of all campaigns. Costs, success definitions and times to results differ across respondents; fieldwork dates undisclosed.
Reference S025 · Original source checked 9 September 2026
Targets often included more than followed links
62% of respondents
In Aira’s 2022 survey, 62% counted nofollow links, including sponsored and UGC attributes, toward overall link targets.
What it means
Many respondents counted nofollow links, including sponsored and user-generated-content attributes, toward their targets. A reported link total can therefore include different link types. Agree the categories before comparing providers or campaigns.
Try this
- Keep followed, nofollow, sponsored and user-generated-content links in separate report columns.
- Agree which types meet the brief. Track unlinked mentions separately because they are not part of this link count.
Use with care: Self-reported team practice or experience. It does not measure the effect of following this approach.
Read the original study: Aira (opens in a new tab)See how the study was done for finding 91
- Research
- 11 Takeaways from the 2022 State of Link Building Report
- How much was studied
- 270 respondents; branching questions have different denominators
- Who or what was studied
- Self-selected SEO professionals answering the annual link-building survey
- When the data came from
- 16 March–2 May 2022
- Where the number appears
- Takeaway 4; PDF page 17, exact survey question
- Keep in mind
- Historical self-reported practice and opinions. Multiple-choice questions may allow several answers; responses are not measured ranking effects. The summary also mentions brand mentions, but the PDF question specifically asks about link attributes. The selected claim follows that narrower wording.
Reference S116 · Original source checked 9 September 2026
Weak campaigns were part of the experience
53% low-link; 29% zero-link
In Aira’s 2022 survey, 53% had run a content campaign earning only 1–9 links, and 29% had run one earning none in the preceding year.
What it means
Many people had worked on campaigns that performed poorly. A useful review should explain why.
Try this
- Review weak campaigns and name the main problem: story, timing, audience or delivery.
- Test one clear change in the next campaign.
Use with care: Self-reported team practice or experience. It does not measure the effect of following this approach.
Read the original study: Aira (opens in a new tab)See how the study was done for finding 92
- Research
- 11 Takeaways from the 2022 State of Link Building Report
- How much was studied
- 270 respondents; branching questions have different denominators
- Who or what was studied
- Self-selected SEO professionals answering the annual link-building survey
- When the data came from
- 16 March–2 May 2022
- Where the number appears
- Takeaway 10; content campaign results
- Keep in mind
- Historical self-reported practice and opinions. Multiple-choice questions may allow several answers; responses are not measured ranking effects. Respondents could have experienced both outcomes.
Reference S118 · Original source checked 9 September 2026
Topic 10 5 findings
Keeping and recovering useful links
Protect the work already done. A removed reference, moved page and temporary crawl error need different responses.
Put this section to work: Check valuable lost links and linked-to pages returning errors. Repair destinations you own, then request specific corrections for confirmed publisher-side problems.
Links disappear over time
66.5% cumulative decay
In Ahrefs’ historical study of links to 2,062,173 websites, 66.5% of links tracked from January 2013 had decayed by the study endpoint.
What it means
The finding covers roughly nine years of accumulated link loss. It is not the share of links expected to vanish next year. It does show why a completed outreach campaign still needs a plan for checking the links and keeping its destination useful.
Try this
- Review important placements at agreed intervals, such as 30, 90 and 180 days, with a named owner.
- Record the source page, destination, live-link status and problem type. Prioritize broken destinations you control and relevant links you can realistically recover.
Use with care: Cumulative historical loss over about nine years; never present this as an annual decay rate.
Read the original study: Ahrefs (opens in a new tab)See how the study was done for finding 93
- Research
- At Least 66.5% of Links to Sites in the Last 9 Years Are Dead
- How much was studied
- 2,062,173 websites
- Who or what was studied
- Historical backlinks pointing to sampled websites
- When the data came from
- Links tracked from January 2013 across a described nine-year window; exact endpoint not reported
- Where the number appears
- Opening finding and link-rot methodology
- Keep in mind
- Page updated February 2, 2024; do not substitute that date for measurement or original publication. Cumulative decay is not an annual loss rate. Ahrefs crawler classifications determine status.
Reference S038 · Original source checked 9 September 2026
Original page addresses often stopped working
25% of sampled pages
Pew found that 25% of sampled pages that existed at some point during 2013–2023 were inaccessible at their original addresses in October 2023.
What it means
A page disappearing from its original address can break citations even if some of its content survives elsewhere. On your own site, URL changes deserve attention when those pages have earned useful links.
Try this
- Before moving or consolidating content, export linked-to URLs and map each to a relevant destination.
- Keep appropriate redirects working and update your internal links. Avoid redirecting unrelated old pages to the homepage just to retain a link count.
Use with care: Observed pattern in the stated sample. It does not isolate what caused the result.
Read the original study: Pew Research Center (opens in a new tab)See how the study was done for finding 94
- Research
- When Online Content Disappears
- How much was studied
- approximately 1 million historical pages; about 500,000 news and 500,000 government pages; 50,000 Wikipedia pages
- Who or what was studied
- Common Crawl pages; sampled news, government and English-language Wikipedia pages
- When the data came from
- 2013–2023 Common Crawl pages checked October 2023; news, government and Wikipedia samples from spring 2023
- Where the number appears
- A quarter of all webpages that existed at one point between 2013 and 2023 are no longer accessible
- Keep in mind
- Availability at the original address was checked at a particular time. Pages and links have different denominators; a moved page may count as inaccessible at its former address.
Reference S039 · Original source checked 9 September 2026
Broken links appeared on established publishers
23% news; 21% government pages
In Pew’s spring 2023 samples, 23% of news pages and 21% of government pages contained at least one broken link.
What it means
Broken links appeared even on trusted news and government sites. Famous domains still have page-level problems.
Try this
- Check that the page works, appears in Google and still contains the link.
- Check again after delivery, even on famous websites.
Use with care: Observed pattern in the stated sample. It does not isolate what caused the result.
Read the original study: Pew Research Center (opens in a new tab)See how the study was done for finding 95
- Research
- When Online Content Disappears
- How much was studied
- approximately 1 million historical pages; about 500,000 news and 500,000 government pages; 50,000 Wikipedia pages
- Who or what was studied
- Common Crawl pages; sampled news, government and English-language Wikipedia pages
- When the data came from
- 2013–2023 Common Crawl pages checked October 2023; news, government and Wikipedia samples from spring 2023
- Where the number appears
- Broken links are common on government and news websites
- Keep in mind
- Availability at the original address was checked at a particular time. Pages and links have different denominators; a moved page may count as inaccessible at its former address.
Reference S040 · Original source checked 9 September 2026
Reference lists also deteriorated
11% of Wikipedia references
In Pew’s spring 2023 sample of 50,000 English Wikipedia pages, 11% of reference links were inaccessible.
What it means
Some Wikipedia references also broke. Research pages need regular source checks.
Try this
- Check links on your research pages regularly, starting with popular pages.
- Replace dead sources and record what changed and when.
Use with care: Observed pattern in the stated sample. It does not isolate what caused the result.
Read the original study: Pew Research Center (opens in a new tab)See how the study was done for finding 96
- Research
- When Online Content Disappears
- How much was studied
- approximately 1 million historical pages; about 500,000 news and 500,000 government pages; 50,000 Wikipedia pages
- Who or what was studied
- Common Crawl pages; sampled news, government and English-language Wikipedia pages
- When the data came from
- 2013–2023 Common Crawl pages checked October 2023; news, government and Wikipedia samples from spring 2023
- Where the number appears
- Links on Wikipedia; reference-link analysis
- Keep in mind
- Availability at the original address was checked at a particular time. Pages and links have different denominators; a moved page may count as inaccessible at its former address.
Reference S041 · Original source checked 9 September 2026
A lost-link alert can describe several different problems
34.2% removed / 6.45% crawl errors
In Ahrefs’ historical breakdown of lost links, 34.2% were classified as removed links, 6.45% as crawl errors and 5.99% as redirects of the linking page.
What it means
A removed link needs a different response from a page that temporarily fails to load. Treat a lost-link export as a queue to check. Some links may still exist, while others moved or were deliberately removed.
Try this
- Open the source page and destination before contacting the publisher. Record whether the reference was removed, moved or temporarily unavailable.
- Fix destinations you control first; then make a specific, useful correction request for confirmed publisher-side problems.
Use with care: Historical tool classifications, not annual loss rates. A crawler error is not confirmation that a link disappeared.
Read the original study: Ahrefs (opens in a new tab)See how the study was done for finding 97
- Research
- At Least 66.5% of Links to Sites in the Last 9 Years Are Dead
- How much was studied
- 2,062,173 websites
- Who or what was studied
- Historical backlinks pointing to sampled websites
- When the data came from
- Links tracked from January 2013 across a described nine-year window; exact endpoint not reported
- Where the number appears
- Most common reasons that links are lost; Link removed, Crawl error and 301/302 sections
- Keep in mind
- Page updated February 2, 2024; do not substitute that date for measurement or original publication. Cumulative decay is not an annual loss rate. Ahrefs crawler classifications determine status.
Reference S155 · Original source checked 9 September 2026
Topic 11 7 findings
Links, coverage and AI visibility
Understand where link building and PR overlap with AI citations. Keep mentions, source links and visits separate.
Put this section to work: Save the platform, exact question, date, answer and cited URL. Separate mentions, links to your site, citations to third-party coverage and referral visits.
AI summaries coincided with fewer result clicks
8% vs 15% of visits
In Pew’s March 2025 US browsing study, traditional-result clicks occurred on 8% of Google visits with an AI summary, versus 15% without one.
What it means
People clicked normal Google results less often when an AI summary appeared. The searches were not identical, so treat this as a pattern.
Try this
- Track important searches with and without an AI summary.
- Compare similar searches before blaming the summary for fewer clicks.
Use with care: Observed pattern in the stated sample. It does not isolate what caused the result.
Read the original study: Pew Research Center (opens in a new tab)See how the study was done for finding 98
- Research
- Google Users Are Less Likely to Click on Links When an AI Summary Appears in the Results
- How much was studied
- 900 adults; 68,879 unique searches reconstructed
- Who or what was studied
- Google search visits in browsing histories of US adults in a tracking panel
- When the data came from
- Browsing March 1–31, 2025; search-result pages reconstructed April 7–17, 2025
- Where the number appears
- Clicking on traditional search results
- Keep in mind
- Observational comparison, with different query types in AI and non-AI groups. April reconstructions may differ from March results; US panel only. A browsing session ends when the user exits the browser for at least five seconds.
Reference S068 · Original source checked 9 September 2026
Clicks on summary citations were uncommon
1% of visits with a summary
In Pew’s March 2025 US browsing study, a link within the AI summary was clicked on 1% of visits with a summary.
What it means
People rarely clicked links inside AI summaries. Being cited and receiving a visit are different results.
Try this
- Record AI citations and visits from AI tools as separate results.
- Report a citation as visibility, even when it produces no visit.
Use with care: Observed pattern in the stated sample. It does not isolate what caused the result.
Read the original study: Pew Research Center (opens in a new tab)See how the study was done for finding 99
- Research
- Google Users Are Less Likely to Click on Links When an AI Summary Appears in the Results
- How much was studied
- 900 adults; 68,879 unique searches reconstructed
- Who or what was studied
- Google search visits in browsing histories of US adults in a tracking panel
- When the data came from
- Browsing March 1–31, 2025; search-result pages reconstructed April 7–17, 2025
- Where the number appears
- Clicking links within AI summaries
- Keep in mind
- Observational comparison, with different query types in AI and non-AI groups. April reconstructions may differ from March results; US panel only. A browsing session ends when the user exits the browser for at least five seconds.
Reference S069 · Original source checked 9 September 2026
Top-position CTR was lower than the model expected
58% lower estimated CTR
Ahrefs’ February 2026 update estimated that AI Overviews were associated with 58% lower desktop position-one CTR than expected without them.
What it means
The study estimated lower clicks to the number-one result than its model expected when an AI Overview appeared. Ranking first and receiving traffic are different outcomes. The modeled gap is not the expected loss for every keyword or campaign.
Try this
- Track relevant queries, positions, impressions and clicks together, noting which searches show AI Overviews.
- Build forecasts from your own query mix. Do not apply the study’s percentage across all organic traffic.
Use with care: Observed pattern in the stated sample. It does not isolate what caused the result.
Read the original study: Ahrefs (opens in a new tab)See how the study was done for finding 100
- Research
- AI Overviews Reduce Clicks by 58% (Updated Study)
- How much was studied
- 300,000 keywords, split into two groups of 150,000
- Who or what was studied
- Desktop Google Search Console data for AI Overview and non-AI informational keyword groups
- When the data came from
- December 2023 versus December 2025
- Where the number appears
- Updated CTR comparison and counterfactual method
- Keep in mind
- Trend-adjusted counterfactual from observational groups, not a randomized test. The 58% estimate is relative to expected CTR without AI Overviews, not a raw before/after difference.
Reference S071 · Original source checked 9 September 2026
Web mentions correlated with AI visibility
0.66–0.71 correlation
Ahrefs’ December 2025 analysis of 75,000 brands reported correlations of 0.66–0.71 between web mentions and AI brand visibility.
What it means
Brands mentioned more often online also appeared more often in AI answers. That does not prove that paid mentions will help.
Try this
- List accurate brand mentions on relevant news, industry and community sites.
- Check which sources AI answers use and improve missing facts.
Use with care: Observed pattern in the stated sample. It does not isolate what caused the result.
Read the original study: Ahrefs (opens in a new tab)See how the study was done for finding 101
- Research
- Top Brand Visibility Factors in ChatGPT, AI Mode, and AI Overviews (75k Brands Studied)
- How much was studied
- 75,000 brands and millions of AI responses
- Who or what was studied
- Brands identified from domains with DR above 40 and a top keyword with at least 800 monthly searches; AI responses in Brand Radar
- When the data came from
- Snapshot dates not reported; includes AI Overview results from earlier 2025 research
- Where the number appears
- Key takeaways; brand mentions; methodology
- Keep in mind
- Spearman correlations in an established-brand-biased sample; not ranking factors or evidence of causation. Web-mention and YouTube-mention measures partly overlap. Platform measurements are not all contemporaneous.
Reference S073 · Original source checked 9 September 2026
AI citations extended beyond the organic top ten
37.1% also in the top 10
In Ahrefs’ March 2026 study, 37.1% of AI Overview citation URLs also ranked among the top ten organic blue links for the same query.
What it means
Google’s AI Overviews cited many URLs outside the ten standard organic results for the measured query. Checking only the first page can miss relevant sources, but AI systems can also use related searches behind the scenes.
Try this
- Save the exact cited URLs for a small set of relevant questions and inspect what makes each useful.
- Compare those sources with the normal search results. Keep the observation date because citation choices and extraction methods change.
Use with care: Google AI Overviews, not all AI tools. Related searches and citation parsing affect measured overlap.
Read the original study: Ahrefs (opens in a new tab)See how the study was done for finding 102
- Research
- Update: 38% of AI Overview Citations Pull From the Top 10
- How much was studied
- 863,000 keyword SERPs and 4 million AI Overview URLs
- Who or what was studied
- Same-query Google SERPs and AI Overview citation URLs in Ahrefs Brand Radar
- When the data came from
- not reported; March 2026 release
- Where the number appears
- Organic position table, second test excluding ads and SERP features
- Keep in mind
- Organic blue-link top ten and first ten mixed SERP blocks are different definitions. Citation parsing changed since July 2025, so older 76.1% cannot support a like-for-like time trend. YouTube non-ranking figure uses the publisher’s broader results framing.
Reference S075 · Original source checked 9 September 2026
Paid content supplied a small citation share
0.3% of cited links
Muck Rack’s May 2026 study classified 0.3% of cited links as paid or advertorial content.
What it means
Paid or sponsored pages made up a very small share of citations. Buying a placement does not guarantee AI visibility.
Try this
- Report paid placements and AI citations as separate results.
- Check the same questions before and after publishing, without promising a change.
Use with care: Observed pattern in the stated sample. It does not isolate what caused the result.
Read the original study: Muck Rack / Generative Pulse (opens in a new tab)See how the study was done for finding 103
- Research
- What Is AI Reading? New insights for May 2026
- How much was studied
- More than 25 million citation links, per the companion May methodology article
- Who or what was studied
- AI responses to constructed, unbranded discovery prompts across 17 industries
- When the data came from
- May 2026 edition; collection window not disclosed in public write-ups
- Where the number appears
- Paid media section; companion methods
- Keep in mind
- Designed prompts are not a random sample of user queries. Transactional branded prompts were intentionally excluded. Model versions, answer settings and citation classification affect the result. The public summary repeats an older 1-million sample description; the companion methods explicitly gives 25 million for this edition.
Reference S136 · Original source checked 9 September 2026
Mentions on well-linked pages related differently to each AI system
0.70 / 0.12 / 0.40 correlations
For the top 50 domains in each system in June 2025, Ahrefs reported correlations between mentions on well-linked pages and AI visibility of 0.70 for Google AI Overviews, 0.12 for ChatGPT and 0.40 for Perplexity.
What it means
A brand appearing on pages that other sites cited was more closely associated with visibility in some AI systems than others. This connects PR coverage with AI research, but it does not show that a backlink or paid mention caused an AI recommendation.
Try this
- Study the relevant publications that AI answers already cite and identify where your expertise could add a useful contribution.
- Report each platform separately, saving the prompt, date and cited page. Separate a brand mention from a backlink to your own website.
Use with care: Small, highly prominent brand groups. The systems and their source choices have changed since June 2025.
Read the original study: Ahrefs (opens in a new tab)See how the study was done for finding 104
- Research
- Does Being Mentioned on Highly Linked Pages Influence AI Mentions?
- How much was studied
- Top 50 mentioned domains per system; drawn from about 76.7 million AI Overviews, 957,000 ChatGPT prompts and 953,500 Perplexity prompts
- Who or what was studied
- Prominent domains in Ahrefs Brand Radar and mentions on pages with referring domains
- When the data came from
- June 2025
- Where the number appears
- The nerdy data: Spearman rank correlation table
- Keep in mind
- Only the top 50 domains per system, not all brands. Prominence is a confounder. A mention on a linked-to page is not necessarily a backlink to the brand. These are historical platform snapshots, not causal effects or current model guarantees.
Reference S154 · Original source checked 9 September 2026
The original research and the refresh plan
This is a curated reference to original publishers' findings. We reviewed source articles, reports, tables and charts; we did not independently reproduce their datasets. The explanations, checklists and calculator examples are our interpretation and are not counted as additional research.
104 counts distinct findings, not every percentage on the page. Comparisons stay together. This revision adds 17 findings and retires 20 peripheral or repetitive entries from the previous 107-finding edition. It gives more space to rankings, authority metrics, internal links and practical link evaluation. Seven findings cover closely related AI visibility.
Some of the most useful studies are historical. Backlinko's ranking analysis used January 2020 data, even though its article was updated in 2025. The time-to-rank research also separates the age of pages that already rank from the chance that a new page will rank. The period and sample travel with each finding so these distinctions survive when you reuse it.
We hold back figures with unresolved conflicts or unclear denominators. Inconclusive tests remain labeled as uncertain, and different survey editions are not automatically treated as a trend. Stable evidence IDs preserve references to retained findings when the collection changes.
Browse all 41 studies and research releases
These are the original research pages behind the findings. One page can contain more than one study.
- Ahrefs — Google Says “Links Matter Less”—We Looked at 1,000,000 SERPs to See if It’s True
not reported; published January 2025 · 4 findings
- Ahrefs — 96.55% of Content Gets No Traffic From Google. Here’s How to Be in the Other 3.45%
not reported; 2023 report · 2 findings
- Backlinko / Ahrefs — We Analyzed 11.8 Million Google Search Results. Here’s What We Learned About SEO
Data extracted in the first week of January 2020; article updated April 2025 · 2 findings
- Semrush — Ranking Factors [Semrush Study]
January 2024 release; collection window undisclosed in the public article · 2 findings
- Ahrefs — How Long Does It Take to Rank in Google? And How Old Are Top Ranking Pages?
New-URL cohorts from September and October 2023 followed for one year; separate ranking-age snapshot published May 2025 · 3 findings
- Ahrefs — Do Links Still Matter for Rankings? A Study by Ahrefs
17 August–14 September 2021 disavow period, with subsequent recovery observation · 1 findings
- Aira — 11 Takeaways from the 2022 State of Link Building Report
16 March–2 May 2022 · 5 findings
- Reboot Online — Digital PR Statistics
Agency coverage acquired during 2024 · 1 findings
- Ahrefs — What Is a Good Domain Rating? (With Real Data)
June 2026 publication; metric extraction dates and search location not disclosed · 2 findings
- Ahrefs — We Crawled Over 1 Million Domains to Find the Most Common Technical SEO Issues
not reported; August 2023 report · 2 findings
- Zyppy — 23 Million Internal Links – SEO Case Study
Collection window undisclosed; article updated 23 February 2026 · 4 findings
- SearchPilot — SEO Split Test Lessons: Increasing Internal Linking
not reported; case study published August 2020 · 1 findings
- SearchPilot — SEO Split Test Lessons: Reducing a Link Block
not reported; case study published April 2024 · 1 findings
- SearchPilot — Does adding internal links to your home page’s footer improve organic traffic?
Test reported September 2021; exact dates not reported · 1 findings
- SearchPilot — Will removing self-referential breadcrumb links improve organic traffic?
Test reported July 2024; exact dates not reported · 1 findings
- SearchPilot — Internal linking case study: can links to nearby location pages help?
January 2021 case publication; test dates undisclosed · 1 findings
- SearchPilot — Will increasing the number of related article links help for SEO?
June 2021 case publication; test dates undisclosed · 1 findings
- SearchPilot — Can a Categories Accordion Boost Organic Traffic?
January 2026 case publication; test dates undisclosed · 1 findings
- Ahrefs — Reciprocal Links: Do They Help or Hurt Your SEO?
not reported · 3 findings
- BuzzStream — Cost of Guest Posts, Based on 500k Sites [Updated 2026 Data]
June 2026 release; collection dates undisclosed · 4 findings
- Ahrefs — I Disavowed “Toxic Backlinks”: Here’s What Happened
20 days before and after the 26 July 2024 disavow; ended before the August core update · 1 findings
- Ahrefs — Do Links From Pages With Traffic Help You Rank Higher? [Ahrefs Study]
February 2019 release; exact extraction dates undisclosed · 1 findings
- BuzzStream — State of Digital PR Report (2026)
not reported; 2026 release · 8 findings
- Orbit Media — Annual Blogger Survey 2025
August 2025 · 2 findings
- BuzzStream and Citation Labs — Link Building Trends Report (2025)
not reported; 2025 release · 4 findings
- Backlinko / BuzzSumo — We Analyzed 912 Million Blog Posts. Here’s What We Learned About Content Marketing
Articles published 1 October 2017–1 October 2018 · 4 findings
- Ahrefs — Link Building Case Study: How We Built Backlinks With a Stats Page
Campaign reported August 2020; precise outreach window undisclosed · 1 findings
- Siege Media — Organic Link Ratios by Industry: What 100 Sites Reveal About Link Acquisition
January 2024–December 2025 · 2 findings
- Backlinko and Pitchbox — We Analyzed 12 Million Outreach Emails. Here’s What We Learned
not reported; 2019 study · 6 findings
- Reboot Online — Digital PR Statistics: email subject-line analysis
Collection period not reported · 2 findings
- Muck Rack — The State of Journalism 2026
2026-01-30 to 2026-03-02 · 9 findings
- Cision — Why Do Journalists Ignore Your PR Pitch?
not reported; 2026 survey release · 5 findings
- BuzzStream — How Much Does Digital PR Cost?
2025 survey; fieldwork dates not disclosed; page updated June 2026 · 5 findings
- Ahrefs — At Least 66.5% of Links to Sites in the Last 9 Years Are Dead
Links tracked from January 2013 across a described nine-year window; exact endpoint not reported · 2 findings
- Pew Research Center — When Online Content Disappears
2013–2023 Common Crawl pages checked October 2023; news, government and Wikipedia samples from spring 2023 · 3 findings
- Pew Research Center — Google Users Are Less Likely to Click on Links When an AI Summary Appears in the Results
Browsing March 1–31, 2025; search-result pages reconstructed April 7–17, 2025 · 2 findings
- Ahrefs — AI Overviews Reduce Clicks by 58% (Updated Study)
December 2023 versus December 2025 · 1 findings
- Ahrefs — Top Brand Visibility Factors in ChatGPT, AI Mode, and AI Overviews (75k Brands Studied)
Snapshot dates not reported; includes AI Overview results from earlier 2025 research · 1 findings
- Ahrefs — Update: 38% of AI Overview Citations Pull From the Top 10
not reported; March 2026 release · 1 findings
- Muck Rack / Generative Pulse — What Is AI Reading? New insights for May 2026
May 2026 edition; collection window not disclosed in public write-ups · 1 findings
- Ahrefs — Does Being Mentioned on Highly Linked Pages Influence AI Mentions?
June 2025 · 1 findings
Next editorial review: 9 December 2026. Recheck original URLs, prioritize fast-changing AI and marketplace evidence, inspect new annual survey editions and verify any changed methods. Hold unsupported claims, recalculate the count and keep a dated change log. This is a maintenance plan; no automated monitoring has been scheduled.
Filter the noise before you fund it.
Compare vetted guest post, digital PR and brand-placement opportunities with visible metrics and exact-site selection. Approve the prepared content before publication. For higher-priced publications, we may require prepayment after confirming availability.
Browse placementsAbout the author
Janis Thies
Co-founder & Managing Director, SEOlutions GmbH
Janis Thies co-founded SEOlutions GmbH and leads its strategic direction, link building methodology and service development. He writes SEOlutions Insights on publisher evaluation, guest posting, digital PR and search visibility.
Continue your research