Are backlinks still relevant in 2026?

Short answer

Yes. Links are still one of the main ways Google discovers pages and judges trust, and its AI features run on the same ranking systems. What changed is which links count: relevant links from pages people read still matter, while bought and bulk links are mostly ignored. And links are less often the real constraint than people assume.

Every year someone declares that links no longer matter, and every year the sites ranking for competitive searches turn out to have strong link profiles. The question is worth asking properly in 2026, though, because AI search has given the argument a new twist.

What the evidence says

Links remain central to how Google finds content. Its own explanation of search describes discovery as happening largely when Google extracts a link from a known page to a new page.

They also still correlate with rankings. Ahrefs studied around 200,000 pages and found pages in the top position gained referring domains at roughly 5 to 14.5% a month, faster than pages ranking lower. Part of that is cause and part is effect: pages that rank get seen and linked more. But the relationship is consistent, and nobody working on competitive searches thinks it has gone away.

What about AI search?

Google says its generative AI features are rooted in its core ranking and quality systems, and that optimising for them is still SEO. So the signals that make a page rank in ordinary results, links among them, feed into what AI answers draw on. There is no separate system where links stop counting.

Other assistants work differently and publish less about how they choose sources. But the sources they cite tend to be pages that are well established and widely referenced, which is to say pages with links and mentions.

What changed

Which links count. Google has spent years learning to ignore links that exist only to manipulate rankings: paid placements, link networks, large-scale guest posting with keyword-rich anchors, directories and profile links. Those mostly do nothing now. They are usually discounted rather than punished, which is why they are still sold. The buyer sees no damage and no benefit.

What still works is links with a reason to exist: a supplier listing, a local paper story, an industry site citing your data. Ahrefs’ study of 44,589 keywords found the traffic of the linking pages correlated with rankings better than the count of links. Links on pages people read are the ones that move things.

When links are not your problem

This is the part most link building pitches skip. On a lot of the sites I audit, links are not the constraint at all.

The clearest case I have: a marketplace whose audit, which I wrote, concluded that investing in links would yield a much better return than content. The same audit found 8,317 of its 9,220 indexed pages could not be reached by clicking through the site. Its main competitor had more referring domains, 845 against 671, but a twelfth of the pages and four times the visits. Links were never the gap. Structure was. I changed my position publicly in 671 links against 29,812, and the number that actually mattered.

So before investing in links, compare your profile with the sites ranking above you. If they are similar, your gap is elsewhere: content, intent, internal linking or technical issues.

When links are the problem

When your pages are as good as or better than the competition and the sites above you have clearly stronger, more relevant link profiles. In competitive markets, legal, finance, home services in big metros, that is common, and no amount of on-page work closes the gap alone.

It is also the problem when a site has a history of bought links that is still weighing on it. Sometimes that needs cleaning up before new links can help.

How to approach links in 2026

Fix internal linking first. Point the strength you already have at the pages that earn money.
Audit what you have. Know which links help, which are ignored, and whether anything needs removing.
Earn fewer, better links. From sites your customers read, with a reason to point at you.
Be patient. Links build over months. A plan that expects movement in weeks is planning for disappointment.

A quick comparison you can run

Pick your three most important search terms. For each, note the three sites ranking above you. Look up how many different websites link to each of them and to you, using any backlink tool, and then open ten of their links at random. Are they from sites in your industry or area? Pages people would actually read? Or directories and thin blogs?

If their links are similar in number and quality to yours, stop thinking about links and look at the pages themselves. If theirs are clearly stronger and more relevant, you have found your gap, and you also have a ready-made list of the kinds of sites worth approaching. That hour of comparison is worth more than any general answer to whether links still matter.

If you are not sure whether links are your real gap, my link building service starts by comparing your profile with the sites above you, and I’ll tell you if the problem is somewhere else.