How many backlinks do I need? How to work out your own number

The short answer

You need about as many referring domains as the pages you are competing with. Count the links to those pages, not to their whole website. When your page answers the search better, you can often get by with fewer.

Three parts of that answer do most of the work.

• Referring domains, not backlinks. One site that links to you 40 times from its footer is still one website vouching for you. Tools show both numbers. The one to plan with is referring domains.
• The page, not the domain. A big software brand might have links from 30,000 websites, but the one article that ranks for your keyword may have links from 12. That 12 is what you are up against.
• Comparable pages. Some results rank on the strength of the brand alone. You cannot match those by counting links, and you do not need to.

What Google says about links

Google does not publish a target number, and any article that gives you one as a rule is guessing. What Google does confirm is that links still matter. Its ranking systems guide says Google has systems that understand how pages link to each other to work out what pages are about and which are most helpful, and that PageRank is still part of its core ranking systems. Its SEO starter guide adds that the vast majority of new pages Google finds are discovered through links.

Everything else in this guide, including the method, is SEO practice for estimating a target. It is not a Google rule.

How to work out your number, step by step

You need a backlink tool that shows referring domains for a single URL. I use Ahrefs; Semrush and Moz show the same thing, and most have a limited free version. It takes one keyword at a time.

1. Pick one keyword for one page

The page you want to rank and the one search you most want it to win. If you try to plan links for a page against five different keywords, you will get five different answers.

2. Check that your page is the right kind of result

Search the keyword in an incognito window and look at the top ten. If they are all guides and your page is a product page, links will not close that gap, because Google has decided people want something else. Fix the format first, or pick a keyword where results like yours already rank.

3. Pull referring domains for each ranking URL

Paste each of the top ten URLs into your tool, exactly as it appears in the results, and note the referring domains to that URL. Write down the site's Domain Rating next to it. You will need both numbers in the next step.

4. Set aside the results you cannot learn from

Mark any result that ranks mainly because of who published it: a huge brand with almost no links to the page, a review site like G2, a Reddit or Quora thread, a Wikipedia entry. They tell you Google trusts that site, not how many links a page needs.

5. Take the middle of what is left

Look at the remaining pages. The median of their referring domains is a sensible target. The lowest one is your floor: it shows a page can rank there with that many links. The highest one is usually not worth chasing.

6. Discount links that do not look relevant

Open the list of sites linking to the lowest two or three competitors. If half of them are unrelated directories or spam, the real number of links that count is lower than the tool shows. Links from sites your buyers actually read carry more weight, and that is the kind worth adding to your own count.

7. Subtract what you already have

Check the referring domains your own page already has. The gap between that and your target is the number you are really planning for.

A worked example

The numbers below are illustrative, made up to show the method. They are not from a real search.

Say you sell scheduling software and want your feature page to rank for "shift scheduling app". Here is what the top five might look like after step 3:

The usable pages have 31, 14 and 9. The median is 14 and the floor is 9. If your page already has 3 referring domains, the plan is roughly 6 to 11 relevant new sites, with 11 as the safer target. That is a very different conversation from "you need 100 backlinks".

Why the number changes by page type

For SaaS sites, the same domain can need very different link counts for different pages.

• Blog posts on specific questions often rank with few or no links of their own, especially when the results are forums or old articles. Internal links from your stronger pages matter more here.
• Feature and solution pages compete with other vendors, and those pages usually have links. This is where the method above pays off.
• "Alternatives" and comparison pages often compete with review sites that rank on brand strength. Count only the vendor pages in those results.
• Your homepage for a broad category term is the hardest. Its target is usually the domain-level link profile of your competitors, which grows over years, not one campaign.

How many backlinks per month is safe?

Google has not published a safe pace, so treat any fixed figure as opinion. What I watch is whether the link profile looks like the ones already ranking in the category. In practice that means three things:

• Pace close to your competitors. Most backlink tools show a chart of new referring domains over time. If the pages you compete with gained a few a month, gaining dozens in one month stands out.
• Varied anchor text. A burst of links that all use the same commercial phrase is the pattern that looks bought. Mix branded and descriptive anchors.
• Relevant sites only. A link from a site in your category is one a reader would expect to find. Dozens from general sites that link to everything mostly add risk.

If you are buying placements, be clear about the risk. Google's spam policies count paid links that pass ranking value as link spam. I set out what that means for link insertions on my service page, and I do not tell anyone it is risk-free.

When more links will not help

Before you plan any links, check that the page is worth linking to. More links will not fix these problems:

• The page does not match what the searcher wants (step 2 above).
• It is thin, or says the same thing as every other result.
• It is not indexed, or Google has picked a different URL as the canonical version.
• Your own site barely links to it, so it sits several clicks deep with no internal links.

Fix these first. They are cheaper than links, and every link you add afterwards works harder.

Once you have a number, the question becomes which sites. I wrote up the 13 checks I run on every site before a link goes on it, because a count only means something if each link comes from a site people actually read.