Do backlinks still matter for SEO in 2026?

I get some version of “are backlinks dead” in my inbox most weeks, usually forwarded from a client who just read a LinkedIn post arguing that google doesn’t need links anymore now that it has AI overviews and years of user behavior data to lean on. I’ve been running SEO tool reviews and a handful of small affiliate sites out of Singapore since before Penguin rolled out in April 2012, and no, they’re not dead. What’s changed is which links actually move anything, and how much junk you can get away with before google notices.

This is for anyone who’s never had to think about this before: what a backlink is, why google still cares, and the wrong ideas that cost people the most time and money chasing them.

what it is

A backlink is just a hyperlink on someone else’s website that points to yours. If TechCrunch writes about your product and links to your homepage, that’s a backlink. If your cousin’s food blog links to your recipe post, that’s also a backlink, just a much weaker one.

Every backlink carries two kinds of signal to a search engine. It says “this page exists and is reachable,” and it says “someone thought it was worth pointing at.” The second part is the one people argue about, because not every link is cast with the same intent. A link from a government transportation department to a logistics company’s freight calculator is a strong, hard-to-fake vote. A link bought cheap on Fiverr and dropped into a random blog’s footer is not, and google’s spam systems have gotten a lot better at telling the two apart since google’s link spam policies were folded into the Search Essentials docs.

how it works

Google’s ranking systems still use links as one signal among many for figuring out which pages to trust, alongside content relevance, page experience, and how well a page actually answers the query. The mechanism traces back to PageRank, the algorithm Larry Page and Sergey Brin published while at Stanford, which treated each link as a vote and weighted that vote by how authoritative the linking page itself was.

That basic idea hasn’t gone away. It’s just one input now, not the whole model. In practice, three things determine whether a given link actually helps:

• where it sits: a link inside the body of a genuinely relevant article carries more weight than one buried in a sitewide footer or sidebar
• who’s linking: a page with its own traffic, its own backlinks, and topical relevance to your niche passes more value than a low-traffic directory nobody visits
• how it’s marked up: google treats rel=”sponsored” and rel=”ugc” as hints, not hard directives, but whether a link is disclosed as paid or dumped in as nofollow still shapes how much trust flows through it

None of this happens instantly, either. A new link usually needs to get recrawled and then sit for a few weeks before you see any movement in rankings, so anyone promising a ranking jump inside a week is either lying or something else changed at the same time and they’re taking credit for it.

why it matters

A few concrete reasons this still isn’t optional if you’re trying to rank for anything competitive.

• discovery: google still finds a lot of new pages by crawling links from pages it already knows about. A page with zero inbound links from anywhere is much likelier to sit unindexed or barely crawled at all
• competitive gap: for any keyword with real commercial intent, the top handful of results are almost always sites with a meaningfully larger, more relevant backlink profile than the ones sitting below them. I’ve watched well-written pages sit on page three for a year because literally nothing linked to them
• trust on YMYL topics: for anything touching money, health, or safety (what google’s own rater guidelines call YMYL, short for “your money or your life”), google leans harder on external validation because it can’t just take the page’s own claims about itself at face value
• the AI overviews problem: with more zero-click results, a real backlink profile is part of what keeps a domain showing up as a cited source instead of getting summarized and skipped entirely. If you’re building genuine topical depth, pairing that with a real internal linking structure does more than either tactic alone

common misconceptions

The authority score your tool shows you is not a google metric. DR from Ahrefs, DA from Moz, and every other third-party score are proprietary estimates vendors built to approximate how google might treat a domain. Google has said repeatedly it doesn’t use any of them internally. Fine for tracking your own site’s trend over time, useless as a target you chase for its own sake.

More links is not automatically better. I’d rather have twelve links from sites that actually cover my niche than two hundred from a PBN or a link farm. Quantity without relevance is often what gets a site flagged, not what fixes a ranking problem. I learned this one the expensive way: bought a batch of links for an early site of mine, got hit not long after, and spent close to six months clawing traffic back. The J.C. Penney case is the bigger, more public version of the same lesson: the New York Times uncovered a paid link scheme that had the retailer ranking first for an enormous range of unrelated household product searches, before google manually penalized the site in February 2011.

Buying links isn’t automatically a death sentence, but it’s against google’s guidelines and it’s a real risk, not a theoretical one. This isn’t legal advice, and I’m not going to pretend there’s a fully safe way to do something google explicitly prohibits, just an honest description of the tradeoff people actually make when they go this route.

Nofollow links are worth more than zero. They still get crawled, they still send referral traffic, and google has treated nofollow as a hint rather than a hard rule since that 2019 change, meaning some of them likely pass signal in aggregate even if you can’t count on any single one.

where to go from here

If you’re new to this, the more useful next questions are about execution, not theory. I’d read how to earn backlinks without buying them next, since most of the tactics that hold up long-term don’t involve paying anyone. After that, how to build backlinks that still work in 2026 gets into the outreach and asset-based approaches I still run today. If a past ranking drop is what brought you here in the first place, how to recover from a google core update is the more relevant read.

Worth knowing too: a lot of the outreach grunt work behind link building gets faster with the right tooling, and if you’re already using AI to draft pitch emails or triage prospect lists, aitoolgazette.com/blog covers which of those tools are actually worth paying for versus which ones just add noise. And before you commit to tracking any of this, it’s worth comparing platforms. I wrote up ahrefs vs semrush for 2026 with backlink monitoring specifically in mind. For everything else we’ve published, the blog index has the full list.

Written by Xavier Fok

disclosure: this article may contain affiliate links. if you buy through them we may earn a commission at no extra cost to you. verdicts are independent of payouts. last reviewed by Xavier Fok on 2026-09-11.