How Many Backlinks Do I Need to Rank? A Study of 231 Keywords

How many backlinks do I need? For most keywords, fewer than you have been told, and the right number sits on your own page one. We pulled the first page of Google results for 231 US keywords and counted how many different websites link to each ranking page. The typical page-one result had links from 22 websites. The #1 result had 77. And 86% of the results pages had at least one page ranking with links from fewer than 10 sites.

The number grows with search volume. For keywords searched 50 to 499 times a month, the #1 result had a median of 21 linking websites; above 50,000 searches, it had 340. Links to the page itself did far more to separate the top three from the rest than the strength of the whole site did. Below is the data, how to read it for your own keyword, and why a fixed monthly link quota is the wrong way to plan.

How we measured page one

On 2 November 2026 we used the Ahrefs API to build a US sample from 12 fixed topics that span local services, finance, software, retail, food, health and travel: dentist, lawyer, HVAC, mortgage, insurance, CRM, SEO, running shoes, recipe, diabetes, hotel and laptop. For each topic we took five keywords in each of four volume bands (50–499, 500–4,999, 5,000–49,999 and 50,000+ searches a month), leaving out branded and navigational searches: 231 keywords in all.

For each keyword we took the organic results in the first 15 positions Ahrefs recorded between 28 October and 2 November 2026. AI Overviews, People Also Ask boxes and video carousels count as positions, so that held a median of eight organic results per keyword, 1,913 in all. We count referring domains, the number of different websites linking to a page, rather than raw backlinks, because one site can link to a page hundreds of times. The full method is at the end.

How many backlinks do I need to rank on page one?

The #1 spot carries the most links, by a wide margin. The median #1 result had 77 referring domains. Positions 2 to 10 together had a median of 18, so the top result had about four times as many. That matches the direction of Backlinko’s study of 11.8 million Google results (last updated April 2025), which found the #1 result had several times more backlinks and referring domains than positions 2 to 10.

Below the top three, counts flatten out. From position 4 down, the median page had 8 to 25 referring domains, and 38% to 51% of pages at each position had fewer than 10.

The spread is enormous. The middle half of #1 pages ran from 7 to 767 referring domains. 56 of the 203 #1 pages we could measure (28%) had fewer than 10, and 21 had none.

Most results pages have a weak spot. On the typical results page, the least-linked result had one referring domain, and in 128 of 227 (56%) one of the top three had fewer than 10. So the honest answer to “how many backlinks do I need to rank” is: count the links on the pages ranking for your keyword now.

The results for this very question make the point. When Ahrefs last checked the US results for “how many backlinks do I need” on 27 October 2026, two of the seven organic results were Reddit and Quora threads. The other five had 19, 17, 3, 1 and 1 referring domains.

The number of backlinks depends on the keyword

US searches a month	Keywords	Median page-one result	Median #1 result	Results pages with a page under 10
50–499	60	5 referring domains	21	90% (52 of 58)
500–4,999	58	13	31	90% (52 of 58)
5,000–49,999	60	25.5	68	87% (52 of 60)
50,000+	53	113	340	75% (40 of 53)

Most small businesses compete in the first two rows. For a keyword with a few hundred searches a month, the median page-one result had 5 referring domains, and nine in ten results pages had a page ranking with fewer than 10. Those are small numbers, which is why our small business SEO guide puts long-tail service and location pages first.

The 50,000+ band is a different game. Its #1 pages had a median of 340 referring domains, and the middle half ran from 66 to 1,458. Even there, three in four results pages had a page with fewer than 10 referring domains somewhere on page one.

By search intent and industry

Local searches had the fewest links on page one: a median of 12 referring domains per result across 76 keywords, against 27 for commercial and 25 for informational searches. They also had the weakest sites, with a median DR of 51 against 84.

By topic (about 20 keywords each, so treat these as rough), page-one results for lawyer and HVAC keywords had a median of 2 referring domains and running shoes 7. Diabetes (150), mortgage (137.5) and CRM (59.5) were far harder. Dentist searches were dominated by small practice sites with a median DR of 12; our guide to getting more dental patients covers the local signals that matter alongside links.

What about keyword difficulty?

Ahrefs’ Keyword Difficulty score tracks these numbers closely, but that is by design, not a finding: Ahrefs says KD is based on the referring domains of the top 10 ranking pages (help article, September 2026). For scale, keywords with KD 0–10 (94 of ours) had a median page-one result with 2 referring domains and a #1 with 8. Keywords with KD 71–100 (23) had 766 and 3,241. Use KD to sort a keyword list, then open the results page itself before you commit.

Page links or Domain Rating: what gets you into the top three?

Domain Rating scores the link strength of a whole website; page-level referring domains count the sites linking to the one URL that ranks. We compared them inside each results page, where the keyword and competition are the same for every result.

In each of 187 results pages with link data for at least six results, we found the page with the most referring domains and the site with the highest DR. The most-linked page ranked in the top three in 53.5% of them, about one and a half times the 36.1% a page picked at random would reach. The highest-DR site made the top three in 34.7%, no better than chance. Rank correlations agree: the median was −0.27 for page links and −0.07 for DR.

Backlinko found the opposite: site-level DR correlated with rankings more strongly than page-level URL Rating. The difference is method. Pooling results across keywords mixes easy and hard searches, and big sites dominate the hard ones. Within one results page, most results already come from strong sites (median DR 81 in the top three, 77 below it), so DR stops separating them.

DR still decides which pages you can realistically join. 101 of our 231 keywords (44%) had at least one site with DR under 30 on page one, and 62 (27%) had one in the top three. Those are the keywords a young or small site can win.

One caution: this is correlation. Pages in the top three also attract links because people find and cite them. Google’s guide to ranking systems (last updated December 2025) confirms PageRank is still part of its core ranking systems, but our data can’t say how much of the top-three advantage the links caused.

Why a backlink count can mislead you

Local homepages are full of junk links. Homepages were 9.4% of page-one results and 22% of results for local searches. They had a median DR of just 19 but a median of 898 referring domains, against 14 for inner pages. Many domains and no weight usually means low-quality links.

We checked 12 of them: every sixth of the 71 local business homepages on page one with DR under 20 and 300 or more referring domains. Ahrefs counted a median of 760 referring domains each, but only a median of 31 were live, not flagged as spam and from sites with DR 10 or higher, about 4% of the total. On the #1 result for “dentist 78253”, a dental practice homepage, most of the highest-rated linking sites were spam-flagged .shop and .store domains with names built around “SEO” and “backlinks”. If you copy a local competitor’s link count, you may be copying spam.

Big platforms take one result in six. 300 of the 1,832 page-one results we could identify (16%) were on platforms such as Reddit (64 results), YouTube (48), Facebook (30) and Yelp (27). Their median DR was 95, and Ahrefs had no page-level link data for 47% of them. You cannot out-link Reddit. You can write the better answer, or take part in those threads and listings yourself.

How many backlinks per month do you need?

A snapshot of page one can’t tell you a monthly rate, and Google’s guidance gives none. What Google does publish is a list of link practices it treats as spam. Its spam policies (last updated August 2026) define link spam as links created “primarily for the purpose of manipulating search rankings”, and the examples include buying or selling links, automated link building, and links from low-quality directories. A plan built around a fixed number of links a month can drift into exactly those tactics.

What we could measure is how fast pages that already rank #1 are gaining links. We took 20 #1 pages from each volume band (inner pages only, no platforms), which gave 75 distinct URLs, and pulled Ahrefs’ monthly referring-domain history from November 2025 to November 2026.

• The median #1 page gained a net 3 referring domains in 12 months, about one every four months. The middle half ran from 0 to 39 a year.
• 27 of the 75 (36%) gained nothing or lost links, and 14 had no referring domains all year while ranking #1.
• Only 16 (21%) gained five or more a month, and none of those ranked for a keyword with fewer than 500 searches.

These are net figures, new links minus lost ones, as Ahrefs sees them. They show what holding #1 looks like, not what it took to get there. The lesson: plan links around a gap, not a quota. Work out how far your page is behind the results you want to pass, then plan the months to close it with links you would want anyway: suppliers, trade associations, local press, partners and data people cite. Our article on how long SEO takes covers realistic timelines, and what SEO costs covers what link work adds to a budget.

How to work out the number of backlinks for your keyword

1. List the organic results. Search your keyword in a private window from your target location. Skip ads, the map pack, videos and AI answers.
2. Look up page-level referring domains for each result in a backlink tool. Use the count for the ranking URL, not the whole domain.
3. Strip out the noise. Set aside Reddit, YouTube, Yelp and other platform pages, and treat local homepages with hundreds of referring domains but a DR under 20 as mostly junk.
4. Find two numbers: the least-linked page in the top three (the minimum that has worked) and the median for the whole page (what a solid position takes). Favor a few relevant links over a long list of directories.
5. Fix the page before you build links. Content, internal links and clean markup come first; our Drupal SEO guide covers the technical side for Drupal sites.
6. Decide who does the work. Earning links is slow, relationship-based work; our comparison of an SEO agency vs in-house team helps you decide where it sits.

Key statistics

• Across 231 US keywords, the median page-one result had 22 referring domains; the #1 result had 77, position 3 had 27 and position 9 had 8 (Gspikes, November 2026).
• 86% of results pages (196 of 229) had at least one page-one result with fewer than 10 referring domains, and 49% had one with none (Gspikes, November 2026).
• For keywords with 50 to 499 US searches a month, the #1 result had a median of 21 referring domains; for 50,000+ searches, 340 (Gspikes, November 2026).
• In 187 results pages, the most-linked page ranked in the top three 53.5% of the time, against 36.1% by chance; the highest-DR site did so 34.7% of the time (Gspikes, November 2026).
• Pages ranking #1 gained a median net 3 referring domains over 12 months, and 36% of 75 gained none (Gspikes, November 2026).
• Twelve low-DR local business homepages on page one had a median of 760 referring domains in Ahrefs, but only 31 from non-spam sites with DR 10 or higher (Gspikes, November 2026).

Methodology and data

Keyword sample. Ahrefs Keywords Explorer (US), matching-terms report, 2 November 2026. For each of the 12 seed topics and four volume bands, we excluded keywords Ahrefs tags as branded or navigational and took the five whose results Ahrefs had checked most recently (four cells had fewer than five). Of the 234 keywords this gave, we removed two malformed entries written with plus signs and one with no results data. Volume, Keyword Difficulty and intent tags are Ahrefs’. A keyword counts as local if Ahrefs tags it local, otherwise commercial if tagged commercial or transactional, otherwise informational.

Results data. Ahrefs SERP overview: organic results within the first 15 positions, numbered in order, keeping the first ten. 81 results had no usable URL (Google redirect links or blanks). 203 had no page-level link data, 141 of them on platforms such as Reddit, Facebook and Yelp, and are left out of the link medians; counting them as zero lowers the median page-one result from 22 to 12 referring domains. Some local searches, such as “dentist” and “hotel”, returned results for Wichita, Kansas, where Ahrefs checked them. Only 9 keywords had a tenth organic result, so the position chart stops at 9.

Comparisons. Homepages are URLs with a root path. Platforms are a fixed list of large social, user-generated, marketplace and directory sites. The top-three test uses the 187 results pages with link data for at least six results; tied pages share the credit, and the chance rate is the share of measurable results in the top three, averaged across pages. The homepage check took every sixth of 71 local homepages with DR under 20 and 300 or more referring domains (12 sites) and counted live referring domains with DR 10 or higher that Ahrefs does not flag as spam; one site hit the API’s 100-row limit, so its count is a floor. The velocity check took the first 20 #1 inner pages per volume band on non-platform sites (80 picks, 75 distinct URLs) and pulled each URL’s monthly referring-domain history from November 2025 to November 2026.

Limits. Ahrefs’ link index is not Google’s, and a count says nothing about relevance. Results move daily and vary by location. The “how many backlinks do I need” results quoted above came from a separate test query and are not part of the sample.

Data. Download the full results dataset (one row per page-one result, with URL, position and Ahrefs metrics) as a CSV file, the per-keyword summary as a second CSV, the 12-month link history of #1 pages as a third CSV, and the local homepage check as a fourth CSV. All are free to reuse with attribution to Gspikes.

Want the number for your own keywords? A page-one link analysis like this is where our SEO optimization service starts: which keywords you can win now, which pages need links, and how many. Gspikes has done SEO and Drupal work since 2018. Ask us for a free review and we’ll run it for your site.

Field Notes

Get the next one by email

Research on Drupal, WordPress and search. Measured, not guessed. New pieces straight to your inbox.

Company website

Email address Subscribe

Confirmed opt-in. One click to leave, any time.