Guide
How many backlinks do you need?
Nobody can tell you, including us, and the pages that do are guessing in a confident font. That is the whole answer to the question as asked. It is also useless on its own, so the rest of this page is about what actually decides it, why the counts you are comparing are not measuring the same thing, and the question that is worth asking instead.
Why the question has no answer
Ranking is comparative, not absolute. There is no threshold you cross; there is a set of pages already holding the results you want, and the question is whether yours can outweigh them for a specific query, judged by a system that weighs relevance, intent, freshness and site quality alongside links. Change the query and the answer changes. Change the competitors and it changes again.
So a number can only ever be true of one page, in one market, on one day. An article that prints a figure has either surveyed a handful of sites and generalised it into advice, or it has invented it because a number reads better than a refusal.
The honest form of the question
Not "how many do I need" but "how many does the page I am trying to outrank have, and where did it get them". That version is answerable, specific to your situation, and it tells you what to go and do.
The counts you are comparing are not the same unit
Before chasing a number, know that most reported link counts are inflated in ways that have nothing to do with effort. A single footer or sidebar link on one site can appear on every page of it, and a tool will happily report that as thousands of backlinks. It is closer to one.
This is why people who do this for a living count referring domains, not backlinks: how many separate sites vouch for you, rather than how many times the same site does. When you see a big backlink number sitting next to a small domain number, the gap is repetition, and the small number is the one carrying the information.
The same applies to your own profile. Ten directory listings on ten domains is a different thing from ten pages on one directory, even though a naive count reads them identically.
What actually decides how many you need
Four things, none of which produce a number on their own, all of which you can observe:
- Who you are competing with. A term owned by established publications needs a different profile from one owned by other small products. This is the variable that moves the answer most, and it is the one generic advice cannot know.
- How specific the query is. The long tail is where a site with almost no links can place, because the competition there has almost none either. Head terms are a different sport.
- Whether the page deserves the click. Links do not rescue a page that answers the query worse than the ones above it. They compound something that already works.
- How many of yours are real. Which is the part almost nobody checks, and the only one on this list you can fix this afternoon.
The count you already have is probably wrong
Here is the part that makes target numbers actively harmful: the profile you are measuring against a target is not the profile you have. A link only does something if all of these hold at once, and each one fails silently:
- The link exists on the page. A submission that was accepted, or a listing that was published and later pruned, both leave you counting a link that is not there.
- It is dofollow. Read the rel on a page. A nofollow link is still worth having for the visitors it sends; it is simply not part of the count you are building.
- The page can be indexed. Check for a noindex. A dofollow link on a page Google drops is the most convincing false win in this whole subject.
- The page can be crawled. Check the robots.txt. A path a crawler may not fetch is a page whose links it never reads.
Before adding to the number, find out what the number is. Most maker link profiles have failures on that list nobody has ever looked for, and every one of them is a link being counted that is doing nothing. Check one you are counting, no account needed.
Why we will not give you a number, in plain terms
This page sits on a product that tracks backlinks, so the incentive to print a target is obvious: a number gives you a reason to buy something. We do not have one, and the reason is not modesty. Nobody outside Google can convert a count into a ranking, and nobody outside Ahrefs can convert one into a Domain Rating, which is why our DR page refuses the same question in its own form.
What we sell instead is the part that is knowable: whether each link you built exists, passes credit, sits on a page Google can read, and is still there this month. That is a verifiable claim, and you can check it yourself with the free tools above, which is rather the point of putting them there.
So what should you do instead
Replace the target with an order of operations. Verify what you already have, because it is the cheapest work available and it changes the number you think you are at. Then earn links that cost you nothing permanent before ones that do, which for most new products means one-way listings before reciprocal swaps. Then stop counting and go look at what the pages beating you actually have.
None of that produces a satisfying figure to aim at. It does produce a list of things to do, each with an outcome you can check, which is the closest honest substitute for the number you came here for.
Questions people actually ask
How many backlinks do I need to rank?
Nobody can tell you, and the number is not the variable people think it is. Ranking is comparative: you need enough to outweigh whoever currently holds the results you want, for the queries you want, in the eyes of a system that weighs relevance, intent and site quality alongside links. Two sites in different niches can sit at opposite ends of the count and rank identically. Any article giving you a figure either surveyed a handful of pages and generalised, or made it up.
Is there a minimum number of backlinks to get indexed?
Indexing and ranking are different questions, and this one has a real answer: zero. Google can and does index pages it reaches through a sitemap or an internal link, with no external links at all. Links help discovery and they help a page compete once indexed, but a new site with none is not locked out of the index.
Do 100 backlinks beat 10?
Only if the hundred are not a hundred copies of the same thing. Most link counts are inflated by repetition: a sitewide footer link on one domain can appear on thousands of pages and is closer to one link than to thousands. This is why practitioners count referring domains rather than backlinks, and why a tool reporting a huge backlink number and a small domain number is telling you the same story twice.
How many backlinks do I need for a new site?
The useful answer for a launch is not a target but an order: get the ones that are certain before the ones that are speculative. A handful of real, indexable, dofollow listings that verifiably exist beats a spreadsheet of a hundred submissions with unknown outcomes, because you can act on the first and can only hope about the second.
How long until backlinks start working?
Longer than the count suggests, and the delay has stages: the page has to be published, crawled, indexed, and only then can the link contribute anything. Any of those can stall silently, which is the reason a link you are counting may not exist as far as Google is concerned. That is a separate question from how many you need, and it is the more useful one to ask first.
Should I buy backlinks to hit a number?
Chasing a count is exactly what makes bought links tempting, which is the strongest practical argument against setting a target at all. Paid links that pass rank are against Google’s spam policies, and the failure mode is not usually a dramatic penalty: it is money spent on links that quietly pass nothing, which is indistinguishable from being scammed. If you cannot tell the difference afterwards, you were not buying a measurable thing.
What is the difference between referring domains and backlinks?
Backlinks counts every individual link; referring domains counts the separate sites they come from. The gap between the two numbers is repetition: one sitewide footer or sidebar link can appear on thousands of pages of the same site and is reported as thousands of backlinks while being, in effect, one vote. This is why people who do this seriously watch the domain count, and why a profile boasting a huge backlink number next to a small domain number is telling you the same story twice.
What should I measure instead of a target number?
Whether the links you already have are real. For each one: it exists on the page, it is dofollow, the page can be crawled and indexed, and it is still there this month. Most maker link profiles have failures on that list they have never seen, and fixing those is a known quantity of work with a known result, which is more than any target number can offer.
Counting links is easy. Knowing which ones are real is the work.
ShipDR reads every listing you track and reports the four things a count hides: the link is there, it is dofollow, the page can be crawled and indexed, and it was all still true this morning. When one of them changes, you hear about it that day instead of discovering it a quarter later.
- Every listing rechecked on a schedule, not once
- Live, dofollow and indexable, read from the page itself
- Told the day a backlink drops or flips
Adding a website and charting its Domain Rating is free, no card. The backlink engine is $10 once, per website, with no subscription.