Link building statistics: verified numbers on backlinks, cost and impact

Why most link building statistics deserve a second look

Plenty of numbers circulate around link building, few of them are properly sourced. This page collects verified link building statistics, each with source and year, grouped by the questions people actually ask. How strongly do backlinks work, how many do top-ranking pages hold, what does building them cost, how common is buying links, and when does it pay off.

While researching this page, every number was traced back to its original source. Three widely quoted values did not survive that check and are therefore absent from this collection. That is no accident. Statistics pages quote each other, and the year of collection and the methodology get lost along the way.

Backlink statistics on ranking impact

Whether backlinks still matter can be answered with data. If you need the basics first, read what backlinks are and how they work.

• 96.55 percent of all pages in the Ahrefs index receive no traffic from Google. Of roughly 20 million pages with no referring domain at all, only 2,997 attract more than 1,000 search visits per month, about one in 6,671. (Source: Ahrefs, 2023)
• The first result on Google holds 3.8 times more backlinks on average than the results in positions 2 to 10. (Source: Backlinko, collected 2020)
• Among domains that ranked in the top 100 for 13 consecutive months, 92.3 percent had at least one backlink. Conversely, 55.1 percent of the domains that never reached the top 10 had no backlinks at all. (Source: Semrush, 2023)

Two caveats belong alongside these. Ahrefs notes itself that the index analysed leans towards the quality side of the web. And Backlinko excluded pages without backlinks from that particular analysis, which lifts every value.

What Google says, and what Google stopped saying

More telling than the correlations is Google's own wording, because it has demonstrably shifted.

• Until February 2024, Google's spam policies stated that Google uses links as an "important factor" in determining the relevance of web pages. (Source: Google Search Central, as of February 2024)
• In early March 2024, coinciding with the March core update, Google removed that sentence from its documentation. It appears in the archived version from 29 February 2024 and is gone in the version from 5 March 2024. (Source: Google Search Central, archive comparison)
• In the current documentation links remain a quality signal. Google writes that one factor used to determine quality is whether other prominent websites link to the content, calling it a good sign that the information is trustworthy. (Source: Google, "How Search Works", 2026)

So the honest reading is neither that links are the top ranking factor nor that they stopped counting. Google still treats references as a trust signal and has grown more careful about weighting them.

Statistics on backlinks at top rankings

How many backlinks a page needs for the front page cannot be answered with a single figure. What the verified data shows above all is how far the values spread apart.

• Pages in the Google top 10 hold 112 referring domains on average. The median sits at 6. For backlinks overall, the mean is 2,418 and the median is 13. (Source: Semrush, Ranking Factors Study 2024, data from 2023)
• For keywords showing an AI Overview, a minimum of 13 referring domains sufficed in the analysis, against 41 for results without an AI Overview. (Source: Ahrefs, 2024)

The gap between 112 and 6 is the most important figure on this page. It means the average is useless as a target. Semrush explains the gap itself with a handful of very large websites such as Amazon or Apple, which have accumulated enormous link volumes over the years and occupy individual results.

Turning that into a target like "we need 112 referring domains" plans against reality. Looking at the actual competitors for your own keyword works better. How many links they hold is quickest to check with the right link building tools.

Link building statistics on cost

Prices for placements spread widely, and the spread has two separate causes. It depends on the tactic and on the quality of the target site.

• A link inserted directly into an existing article cost 361.44 US dollars on average. A paid guest post cost 77.80 US dollars. Both values come from the same study and the same niches. (Source: Ahrefs, 2022)
• Placements booked through GetMentioned cost 561 euros on average in 2026. The analysis reaches back to 2016, and the average price has risen over recent years. (Source: GetMentioned, internal analysis, 2026)
• For monthly budgets on external link building, the most common bracket was 10,000 to 25,000 US dollars, named by 25 percent of respondents. The second most common was 2,500 to 5,000 US dollars at 19 percent. (Source: Aira, State of Link Building Report, 2022)
• Only 8 percent of the agencies and freelancers surveyed bill per link. 45 percent work on a monthly retainer. (Source: Aira, 2022)

That last point rarely comes up and explains a good deal of confusion in sales conversations. Price per link and agency retainer are different quantities. A retainer usually cannot be divided sensibly by the number of links delivered, because research, coordination and copy are baked into it.

For guest posts there is an added catch that Ahrefs points out itself. The lower placement price is misleading, because producing the article is a cost of its own. On the GetMentioned placement marketplace you see publishers, SEO data and prices side by side before booking.

Statistics on how common buying links is

On this topic the spread between measurements is the finding. Two surveys measure nominally the same thing and land more than a factor of two apart.

• 30 percent of the SEO professionals surveyed name paid links among the tactics they use. (Source: Aira, 2022)
• 74.3 percent of the professional link builders surveyed say they pay for links. (Source: Authority Hacker, collected 2022)
• In a covert purchase attempt, 12.6 percent of the websites contacted were willing to sell a link. (Source: Ahrefs, 2022)

The distance between 30 and 74.3 percent has a methodological explanation. At Aira, "paid links" was one of some fifteen options in a question about tactics used. Authority Hacker asked anonymously and directly, and its sample consisted mostly of niche site owners. The third figure does not rest on self-reporting at all. It measures real responses to a purchase enquiry, though on the supply side.

A fourth number from the same survey completes the picture.

• 69 percent of respondents believe bought links can influence rankings positively. Only 30 percent say they use the tactic themselves. (Source: Aira, 2022)

Those 69 percent get passed around in many collections as a prevalence figure. That is wrong. The question asked whether buying links can work, not whether respondents do it. It measures belief, not behaviour.

Link building statistics on timing and return

Here two measurements deserve separating, because they get mixed up constantly. The time until a page ranks at all differs from the time until a single link takes effect.

• Only 1.74 percent of newly published pages reach the Google top 10 within a year. Among those that do, 40.82 percent get there in the first month. (Source: Ahrefs, 2025)
• 49 percent of the SEO professionals surveyed see the effect of link building within one to three months, another 30 percent within three to six months. (Source: Aira, 2022)
• Practitioners rate the effectiveness of link building for organic rankings at 7.8 out of 10 on average. At the same time, for 29 percent the most recent content-led campaign produced zero links. (Source: Aira, 2022)

The first figure illustrates this industry's ageing problem well. The value 5.7 percent from the same study series circulates widely. Ahrefs re-ran the research in 2025 and arrived at 1.74 percent, with the old address redirecting to the new version. Quoting 5.7 percent today means quoting a number the publisher itself has replaced.

One caveat matters. The Ahrefs study measures page rankings in general and is not conditioned on link building. It supports how hard rankings are, not how fast a single link works. For the latter there is only self-reported data from the Aira survey, whose question also merges rankings and traffic.

The contrast in the third line is the most honest value in the whole collection. A high effectiveness score alongside nearly three in ten campaigns that returned nothing. Planning beats placing individual links opportunistically. What that looks like is covered in our backlink strategy.

Turning statistics into action

Numbers show you what matters. MentionIQ takes the next step. It analyses your website and your competitors, shows your backlink gaps and your visibility gaps against the competition, both in classic SEO visibility and in visibility inside large language models and AI search. On that basis MentionIQ recommends suitable placements.

Through GetMentioned you book them directly on the placement marketplace. If you would rather hand over the whole build, the link building agency runs it as a full service.

What to take away from these numbers

Three things. The median tells you more than the average, because a few very large sites distort every mean. The year of collection belongs to every statistic, because part of this industry relabels datasets annually without collecting new data. And link building works over months, not days.