Link Building · 11 min read

How Many Backlinks Do I Need to Rank? The Data, and a Method to Find Your Own Number

Published · Updated · by 3rd Unicorn editorial team

Order on the marketplace → Or have our agency do it

Branded graphic: how many backlinks do I need to rank — the data and a method to find your own number

There is no single number of backlinks that ranks a page. The best public benchmark, WebFX’s 2026 study, puts the median page-one site at 907 referring domains, but that ranges from 76 to 3,027 by industry, and it measures whole domains, not the page competing for your keyword. The number that matters is the count of quality referring domains held by the pages already ranking for your keyword, and you can estimate it in about an hour with the method below.

This guide covers what the studies actually say (including a figure that is widely misattributed), what Google says about links, and a step-by-step method with a table template you can copy.

Four studies are worth knowing. Each measures something different, so read the “what it measures” column before quoting any of them.

StudySampleKey findingWhat it measures
WebFX, 20261,462 page-one domains, 150 commercial keywords, 15 industries, data from March 2026Median 907 referring domains; 76 (Apparel) to 3,027 (Finance & Insurance); average 48 new referring domains a monthWhole domains, not pages
Backlinko11.8 million Google results, Ahrefs dataThe #1 result has an average of 3.8x more backlinks than positions #2–#10, and about 3x more referring domainsPages, plus a domain-level correlation
Ahrefs search traffic study, Dec 2023About 14 billion pages96.55% of pages get zero traffic from Google; of about 20 million pages with no referring domains, only 2,997 get 1,000+ visits a monthPages
Ahrefs backlink growth study, 2018About 200,000 pages, top 20 for 10,000 keywordsMost #1 pages gained followed referring domains at +5% to +14.5% a month; the top 3 gain more new referring domains than the restPages, over time

A note on the 3.8x figure: it comes from Backlinko’s 11.8-million-result study. You will often see it credited to Ahrefs, sometimes with a link to the Ahrefs backlink growth study, which contains no such claim. Ahrefs supplied Backlinko’s data, which is probably where the mix-up started. If you quote it, quote Backlinko.

What WebFX’s industry medians look like

WebFX’s medians give you a sense of scale before you look at your own keyword. A selection from the study:

IndustryMedian referring domainsNew referring domains a month
Finance & Insurance3,027101
Real Estate2,03259
Health1,91571
Computers & Electronics1,11723
Business & Industrial93657
Home & Garden46148
Food & Drink28439
Apparel7615

Source: WebFX 2026 backlink study. WebFX also found 92.2% of the links were editorial, which matters later.

Why these averages can’t give you your number

Three things limit every benchmark above.

  • Domain vs page. WebFX counts referring domains to the whole site. A site with 900 referring domains might have 15 pointing at the page that actually ranks. Your competitor is a page, and the site behind it.
  • Your keyword is not an industry. WebFX notes Computers & Electronics has the widest spread, with a 25th percentile of 262 and a 75th of 12,813. A median sitting inside a range that wide tells you very little about one query.
  • Correlation, not cause. Backlinko describes its work as a correlation study, and Ahrefs says the same of its traffic data. Pages that rank well also attract links because people find them. The numbers show what ranking pages have, not a recipe.

Google is clear that links still count. Its ranking systems guide says of PageRank that it “continues to be part of our core ranking systems”. Its link best practices say Google “uses links as a signal when determining the relevancy of pages and to find new pages to crawl”.

Google is equally clear that links are not the only thing. Its SEO starter guide says “there’s much more to Google Search than just links”. And its spam policies define link spam as creating links “primarily for the purpose of manipulating search rankings”, with examples that include buying links that pass ranking credit and “using automated programs or services to create links to your site”. Paid placements are fine when they are qualified with rel=“sponsored” or rel=“nofollow”.

So the practical rule: the number you are trying to match is a count of genuine, followed, editorial links from real sites, not a count of anything a tool reports.

Sometimes, for low-competition, long-tail queries. But the odds are poor for anything with real search volume. In Ahrefs’ study, about 20 million pages had no referring domains and only 2,997 of them earned more than 1,000 search visits a month. That’s roughly one in 6,700.

If the top 10 for your keyword includes several pages with zero or near-zero referring domains, that’s a sign the query is winnable on content and internal links alone. The method below will show you whether that’s the case.

This method estimates the number of quality referring domains you need for one keyword. It builds on two ideas from the ranking pages: WebFX’s gap formula (target minus current, divided by time, plus velocity) and Robbie Richards’ practice of filtering a competitor’s links to quality ones (in his example, 85 referring domains shrank to 22 once he kept only dofollow links from DR30+ sites with real traffic). What it adds is looking at all ten results at page level and domain level, so one outlier doesn’t set your target.

You need any backlink tool that reports referring domains per URL.

Step 1: Record the top 10 for the exact keyword

Search the keyword in the market you care about and list the ten organic URLs. Skip ads and SERP features. Note any result that is a different content type from yours (a forum thread, a tool, a homepage), because it competes on different terms.

Step 2: Pull page-level and domain-level referring domains

For each URL, record:

  • Page RDs (all): referring domains to that exact URL.
  • Domain RDs: referring domains to the whole site.

You need both. Backlinko found a site’s overall link authority correlated more strongly with rankings than the page’s own. A big site can rank a page with very few links of its own.

Step 3: Filter to quality referring domains

For each page, keep only referring domains that are followed (not nofollow, sponsored or UGC), from sites with real organic traffic, and topically relevant. Record that as Page RDs (quality). This is the column your target comes from. Raw counts include scraper sites, directories and spam that won’t help you, and you shouldn’t try to match them.

Record new referring domains to each page over the last 90 days. This tells you whether the target is moving. Ahrefs found most #1 pages keep gaining followed referring domains every month, so a target you measure today will be higher by the time you reach it.

Step 5: Take medians, not averages

Calculate:

  • Page-one entry target: the median quality page RDs across all ten results.
  • Top-3 target: the median quality page RDs of positions 1–3.
  • Domain context: the median domain RDs across all ten.

Medians stop one heavily linked page from skewing the answer.

Step 6: Calculate your gap and monthly pace

Subtract your page’s current quality referring domains from each target. Then use WebFX’s formula: divide the gap by the number of months you have, and add the competitors’ monthly velocity.

Monthly target = (target quality RDs − your quality RDs) ÷ months + competitors’ new quality RDs a month

The table template

Copy this into a spreadsheet, one row per result:

PosURLContent type matches yours?Page RDs (all)Page RDs (quality)Domain RDsNew page RDs, last 90 days
1
2
…
10
Median, all 10
Median, top 3
Your page

A worked example (hypothetical numbers)

The numbers below are invented for illustration. They are not client data and not a real SERP. They show how the arithmetic works.

Imagine a B2B keyword with moderate competition. Your page has 6 quality referring domains and your site has about 300 domain-level referring domains. The top 10 looks like this:

PosPage RDs (all)Page RDs (quality)Domain RDsNew page RDs, 90 days
1140482,1009
295311,4504
3210625,80012
438146201
564229803
61259,4000
751194102
827113501
944167602
10941200

Working through it:

  1. Page-one entry target. Sorted quality RDs: 4, 5, 11, 14, 16, 19, 22, 31, 48, 62. The median is (16 + 19) ÷ 2 = 17.5, so about 18.
  2. Top-3 target. Positions 1–3 have 48, 31 and 62. The median is 48.
  3. Gaps. To page one: 18 − 6 = 12. To the top 3: 48 − 6 = 42.
  4. Velocity. The top 3 gained 9, 4 and 12 new referring domains in 90 days, about 8.3 per quarter or 2.8 a month. Across all ten, the median is 2 per quarter, about 0.7 a month.
  5. Monthly pace. Page one in six months: 12 ÷ 6 + 0.7 ≈ 3 quality referring domains a month. Top 3 in twelve months: 42 ÷ 12 + 2.8 ≈ 6–7 a month.
  6. Domain check. The median domain RDs is 870, against your 300. Position 6 shows why that matters: it ranks with 5 quality page RDs because its site has 9,400. You are not that site, so plan to sit at or slightly above the page-level median, and build some links to your homepage and supporting content, not only the target page.

The headline: this hypothetical page needs about 12 more quality referring domains to compete for page one, not the 907 a site-wide average might suggest, and about 42 for the top 3.

  • Counting backlinks instead of referring domains. WebFX puts it plainly: one site linking from 200 pages still counts as one referring domain. Backlinko’s data points the same way.
  • Copying the #1 result. The top result often has years of links and a strong domain. Aim for the median first.
  • Ignoring link type. Sponsored and nofollow links are the correct, compliant way to label paid or untrusted links, and they can send traffic and visibility. Leave them out of the quality count you are matching, and don’t expect them to close a ranking gap.
  • Measuring once. Re-run the table every quarter. Ahrefs found top pages keep growing.
  • Forgetting the page has to deserve it. If your page is weaker than the top 10, links will be expensive and slow. Fix the content first.

How do you turn the number into a plan?

Once you have a monthly target, the questions are where the links come from and what they cost. Editorial outreach, digital PR, guest contributions on relevant sites and sponsored placements all have a role. Spread links across your target page and the pages that support it, and keep anchors mostly branded, as covered in our anchor text guide.

If you’d rather hand the counting and the building to someone else, that is what a link building agency should do: run this analysis per keyword, set a quality floor for every link, and report against the gap. Our editorial links start at $99 (DR50+, 5,000+ monthly visits, 7–10 days), and the full set of packages is on our pricing page. If you want to choose placements yourself, the 3rd Unicorn Marketplace lists sponsored articles and other placements with each publisher’s metrics and link type shown. For a wider view of budgets, see how much link building costs.

The bottom line: the right number of backlinks is the median count of quality referring domains held by the pages ranking for your keyword, plus whatever they will add while you catch up. Measure it per keyword, count domains not links, and use the benchmarks for context only.

Want the gap worked out for your keywords? Our managed link building team will run the top-10 analysis and tell you honestly how many links each page needs.

Frequently asked questions

How many backlinks do you need to rank on the first page of Google?

There is no universal number. WebFX's 2026 study of 1,462 page-one domains found a median of 907 referring domains per site, ranging from 76 in Apparel to 3,027 in Finance & Insurance. For a specific keyword, the useful number is the median count of quality referring domains pointing at the pages already ranking for it, which is often far lower than the site-wide figure.

Is it possible to rank without backlinks?

For low-competition queries, yes, but it is rare for anything with real traffic. Ahrefs found about 20 million pages in its index with no referring domains, and only 2,997 of them got more than 1,000 search visits a month.

How many backlinks per month should I build?

Enough to close your gap in the time you have, plus enough to keep pace with competitors who are still earning links. WebFX measured an average of 48 new referring domains a month for page-one sites, and suggests dividing your gap by the months available and adding your industry's velocity.

Is 100 backlinks good?

It depends what they are and where they come from. 100 backlinks from three sites count as three referring domains. Count unique referring domains with followed, editorial links from sites that have real traffic, then compare that number with the pages ranking for your keyword.

Does the number of backlinks matter more than their quality?

Neither works alone. Backlinko found the number of referring domains correlated with rankings, but Google's spam policies mean links built mainly to manipulate rankings can be ignored or count against you. Count quality referring domains, not raw links.

3rd Unicorn

3rd Unicorn editorial team

Link building & SEO desk

Written and fact-checked by the team that vets publishers, audits placements and runs the 50 publications we own. Sources are named in the text; corrections go to support@3rdunicorn.com.

Place it yourself, or have it done for you

Browse live listings on the 3rd Unicorn Marketplace, with payment held until the link is live, or book a call and our agency team will plan the work with you.