Best backlink sites, and why a free list is the wrong place to start
Every free list of backlink sites is a list somebody else has already worked through. Here is what a site has to be worth a link, the categories that still work, and the test to apply to any list.
Published by AI Knows Us · Updated 21 October 2026
A published list of the best backlink sites has one guaranteed property: everybody who searched the same phrase has the same list. That is not a reason to ignore it, but it does tell you what the list is worth. The value in a link comes partly from the page being relevant and partly from it being hard to get, and a list circulating freely has already given away the second half.
So this page does two things instead of printing another list. It sets out what makes a site worth a link, on checks you can run yourself, and it names the categories that still work and why. The criterion throughout: can you verify the thing before you spend time on it?
What makes a site worth a link, in order
| Check | How to run it | Why it is first |
|---|---|---|
| Does the page have readers? | Look for comments, shares, a recent publishing date, a named author with a history. A site with no audience is a page that exists only for links. | A link from a page nobody reads sends nobody. Everything else on this list is downstream of this. |
| Is the topic actually yours? | Read the three most recent articles. If none of them would plausibly mention a business like yours, the match is wrong whatever the metrics say. | Relevance is the part that still works when tactics stop working. |
| Is the page indexed? | Search its exact title in quotes. | A page outside the index passes nothing on, and this takes ten seconds. |
| Is the link in the rendered HTML? | Load the page with scripts disabled and look for the link. | A link that only exists after JavaScript runs is a link a crawler may never see. |
| What does the vendor metric say, last? | Domain Rating, Authority Score, Trust Flow and Citation Flow each come from their own vendor's crawl, and Google uses none of them. | It sorts a list. It does not decide whether a link is worth having, and putting it first is how people end up on sites with good scores and no readers. |
The categories that still earn links
Five kinds of site reliably publish links to businesses and are not on any free list, because getting onto them is work rather than submission.
Trade publications in your actual industry. There are usually three or four, they are read by your buyers, and they publish data, comment and corrections from practitioners.
Association and membership pages. If you belong to a trade body, a chamber or a licensing register, those pages exist already and half of them have your listing without a link because nobody asked.
Supplier and partner pages. The companies whose products you use, resell or integrate with often maintain customer and partner lists. This is the warmest outreach there is and it is almost always unworked.
Comparison and roundup pages in your category. Search your category phrase with best, top and alternatives. Those pages are the highest intent link targets that exist and they are maintained by people who want to be accurate.
Pages that already mention you without linking. Search your brand name in quotes and your founder's name. Every unlinked mention is a publication that has already decided you are relevant.
How to judge a free list before using it
Take the first five entries and run the reader test and the index test above. If three of the five are sites with no visible audience, the list was assembled from metrics rather than from reading and the remaining ninety entries will be the same.
Then check whether the list is a submission list in disguise. A site that accepts anything is worth what acceptance costs, and Google's spam policies page names using automated programs or services to create links to your site as an example of link spam, which is what working through a submission list at volume amounts to.
Where the tool vendors beat us, said plainly
We publish no list of backlink sites, no link index and no crawler. On the subject of this page the vendors with crawls are the ones who can tell you anything about a site's link profile: Ahrefs publishes 493 B pages in its index refreshed every 15 minutes, Semrush publishes more than 43 trillion backlinks refreshed at least hourly, and Majestic publishes Trust Flow and Citation Flow against a Fresh Index covering the past 120 days. All three are better placed than us on this page's own subject.
Our rendered pricing page also carries no SSO, no white label, no SOC 2 and no ISO line, and covers one business per plan outside the agency tier, which covers ten.
The list of sites that now decides who gets named
There is a second list worth more attention than any free backlink list, and it is specific to your business rather than general. When a buyer asks an assistant who to use, the assistant reads a handful of sources and names two or three companies. That handful is the only list that mattered in that moment, and it is usually short, often surprising, and knowable.
AI Knows Us reads it. We ask the buyer questions you choose across ChatGPT, Gemini, Perplexity, Claude and Google AI Overviews, report which businesses each answer named, and list the sources behind the answer, which is how you find out which publications are actually feeding the recommendation in your category. The published arithmetic sits on the rendered pricing page: one credit is one buyer question asked on one AI app once, most checks use one credit, a Claude check uses 3 because Claude costs more to run, and checking 10 questions on 4 AI apps uses 40 credits. llms.txt publishes the rule that keeps the result meaningful: the buyer questions never name the business being measured. The US floor is $149 a month for SEO Start, and llms.txt states that plan watches 0 buyer questions on 0 engines, so Growth at $299 is the cheapest plan reading all five surfaces.
Ahrefs sells this too and is ahead of us on two published facts: Brand Radar states each prompt is checked daily on every selected platform, where ours is weekly on priority questions, and it reads AI Mode and Copilot, which we do not. Rankscale lists nine engines against our five and publishes a REST API and the removal of its branding from shared dashboards, which we do not. The only criterion on which our published figure is the most exact is the weight of the expensive engine: Ahrefs publishes 8 checks per update for Claude and Rankscale publishes a flat typical 0.25 credits per engine per prompt with no per engine weight.
Buying anything from a best of list has the same flaw as this one, and it is worth understanding before you act on any ranking, including ours. The argument is set out in our piece on the best rated SEO tool for B2B, and why the rating is the wrong thing to buy on. We measured durable inbound links ourselves on 21 October 2026, reading 2,169 stored bodies (1,797 drafts plus 372 fetched) against all 2,082 pages tracked on this site: 932 of them, 44.8% of 2,082, carry no inbound link written into another page's body. No other page on this site links to that one.
Questions buyers ask about the best sites for backlinks
Is there a definitive list of the best backlink sites?
Not a useful one, because a list everybody has is a list everybody has worked. The sites worth a link in your category are the trade publications, associations, suppliers and roundups specific to it, and finding them takes an afternoon of searching rather than a download.
Do free backlink site lists still work?
They produce links. Whether those links do anything is a different question, and the reader test answers it: a site with no visible audience sends no readers and is on the list because it accepts everybody.
What is the single best type of site to target?
A page that already mentions you without linking. The publication has decided you are relevant, somebody there wrote your name, and the ask is a correction rather than a pitch.
How many good backlink sites do I need?
Fewer than the lists imply. Twenty relevant, read, indexed pages in your category outperform two hundred directory entries, and twenty is a number one person can work through properly in a quarter.
Which list should I be checking that I am probably not?
The sources an assistant cites when it answers your buyer's question. It is short, it is specific to your category, and being on it is what decides whether your name appears in the answer.