Do Backlinks Still Matter in 2026? What Changed and What Did Not

The short answer

Backlinks still matter. They are not the only thing that matters, and they are not the most important thing for every query, but a site with no relevant links will struggle to rank for anything competitive.

What has changed is the type of link that counts. Ten years ago, volume of links could move rankings even when the links were low quality. Today, Google's systems are designed to ignore or neutralize most manipulative links, so large numbers of cheap links mostly do nothing, while a smaller number of relevant, editorial links still make a clear difference.

What Google has said

A few points from Google's own documentation and public statements:

• Google's guidance on how Search works explains that links help it discover pages and are among the signals it uses to understand relevance and importance. The original PageRank idea, that a link is a kind of vote, is still part of how Google talks about links.
• Google's spam policies treat link schemes as spam, including buying or selling links for ranking purposes, excessive link exchanges, and automated link creation. If links did not matter, these policies would be unnecessary.
• Google representatives have said in recent years that links are less important than they used to be and that people overestimate them. That is consistent with the other points: links matter, but they are one signal among many and content quality is essential.
• Google has said its systems, including SpamBrain, can identify and neutralize spammy links, which means many purchased links simply give zero benefit.

Links that still move rankings

In our experience the links that make a measurable difference share these traits:

• Topically relevant: the linking site and page are about the same subject as yours.
• Editorial: placed in the body of real content because the link helps the reader.
• From sites with real traffic: pages people actually visit, not sites that exist only to sell links.
• Unique referring domains: one link each from 20 different relevant sites beats 20 links from one site.
• Pointing to the right page: links to the specific page you want to rank, not only the homepage.

A link from a niche newsletter's article that sends ten visitors a month is often worth more than a link from a generic high-DA site no one reads.

Links that are mostly wasted

• Comment and forum spam: usually nofollow or ugc, often deleted, and ignored for ranking.
• Private blog networks (PBNs): sites built only to link out; a classic link scheme.
• Mass directory submissions: hundreds of low-quality directories that accept anything.
• Sitewide footer or sidebar links from unrelated sites.
• Links on pages Google does not index: if the linking page is not indexed, it is unlikely to pass value.
• Bulk "DA 50+" packages where every link sits on unrelated articles across sites that sell to anyone.

These can be risky as well as useless; see buying backlinks: risks and Google's link spam policy.

Why links matter for AI search too

AI answer engines like ChatGPT search, Perplexity and Google's AI Overviews mostly answer by retrieving pages from a search index and summarizing them. Which pages get retrieved depends heavily on how well they rank for the query and its sub-questions.

So links influence AI visibility indirectly: they help your pages rank, which makes them more likely to be retrieved and cited. Mentions of your brand on third-party sites, linked or not, also shape how AI systems describe you. See what GEO is for the full picture.

When links matter most

Links have the biggest impact in a few situations:

• New domains that need to be discovered and trusted.
• Competitive queries where many pages have strong content and links break the tie.
• Commercial pages (product, pricing, comparison) that rarely earn links naturally.
• Sites recovering visibility after content improvements, where authority is the missing piece.

They matter least for very specific long-tail questions where few good answers exist; there, content alone often ranks.

Unlinked brand mentions: the overlooked signal

Not every valuable reference is a link. When a newsletter, podcast transcript, forum thread or review mentions your product by name without linking, it still does work:

• It helps search engines and AI systems connect your brand with a category and a set of problems.
• It drives branded searches, which tend to convert well.
• It gives you an easy outreach opportunity.

To turn mentions into links, set up alerts for your brand and product names, check each new mention, and send a short note to the author: thank them, and ask whether they would add a link so readers can find you. Many will, because it makes their article more useful. In our experience this is one of the highest-response forms of link outreach, since the author already chose to write about you.

A balanced strategy

For most startups, a sensible split of effort is roughly:

• Content first: pages that genuinely answer buyer questions better than what ranks.
• A steady trickle of relevant links: 5-15 new referring domains a month from listings, outreach, PR, partnerships and, where appropriate, transparent placements.
• Internal links that pass authority from your strongest pages to new ones.
• Measurement: track rankings for target pages before and after links are added.

If you want the link side handled, see our backlinks service and pricing.

How to test whether links move your rankings

You do not have to take anyone's word for it, including ours. Run a simple test on your own site:

1. Pick 4-6 pages that rank between positions 8 and 30 for clear target queries, with stable rankings for the past month.
2. Split them into two groups with similar positions and similar content quality.
3. Build links to group A only: 3-5 relevant, editorial referring domains per page over two to three weeks. Leave group B alone.
4. Change nothing else on either group: no content edits, no new internal links.
5. Track positions weekly for 8-12 weeks in Search Console (average position for the target query) and a rank tracker.

If group A moves and group B does not, links are doing work for your site in your niche. If neither moves, look at content, intent and technical issues first. Small tests like this are noisy, so treat the result as directional, but it beats guessing.