Do Backlinks Still Matter in the Age of AI Search?

Every few months someone declares that links are dead. AI search is the newest reason. If buyers ask ChatGPT or read a Google AI Overview instead of clicking ten blue links, why chase backlinks at all?

The short answer: AI answers are built from the same web that links describe. The sources these systems trust, retrieve and cite are mostly the sources that already earned authority the old-fashioned way. Links haven't stopped mattering. What has changed is which links matter most.

How AI answers choose their sources

Most AI search experiences work in two steps. First they retrieve a set of documents that look relevant to the question. Then a language model writes an answer from those documents, often with citations.

The retrieval step leans heavily on classic search signals. Google's AI Overviews sit on top of Google's index and ranking systems. Tools like Perplexity and ChatGPT search pull from web search results as well. If your page doesn't rank, or isn't on a site the retrieval system trusts, it rarely makes it into the answer.

That gives links two jobs in AI search:

1. Direct: links still help your pages rank, and ranking pages get retrieved and cited more often.
2. Indirect: links and mentions on trusted publications teach models which brands belong to which topics.

Mentions are the new anchor text

For years, SEOs obsessed over anchor text: the clickable words in a link. In AI search, the sentence around the link matters just as much.

When an industry publication writes "Acme is one of the few SOC 2 tools built for startups", that sentence becomes evidence. Repeat it across enough credible sources and AI systems start associating your brand with that use case, even when nobody clicks a link.

This changes what a good placement looks like:

• Context beats raw authority. A DR 55 niche publication that describes exactly what you do is often worth more than a DR 85 general news site that mentions you in passing.
• Consistency matters. Describe your product the same way across placements so the association is clear.
• Category language wins. Get mentioned alongside the category terms buyers actually use, such as "link building agency for SaaS" rather than "growth partner".

What to change in your link strategy

You don't need a separate "AI link building" program. You need to tighten the one you have.

1. Prioritize sources that AI systems already cite

Ask ChatGPT, Perplexity and Google a dozen questions your buyers ask. Note which publications and sites appear as sources again and again. Those domains are your highest-value targets, because they're already in the retrieval set for your topic.

2. Earn links to pages that answer questions directly

AI answers quote pages that state facts clearly. Comparison pages, "what is" explainers, pricing breakdowns and original data perform well. These are often the same SaaS pages that deserve links first. A link to a page with a clear, quotable answer is worth more than a link to a generic homepage.

3. Publish data others will reference

Original statistics are the most cited content type in AI answers because models need a source for numbers. A yearly benchmark report or survey can earn links and mentions for years.

4. Keep quality ruthless

AI search makes spam more dangerous, not less. Sites built to sell links rarely end up in retrieval sets, so links from them do nothing for AI visibility. They can still hurt your normal rankings, and buying them breaks Google's spam policies. Real publications with real readers win on both fronts.

How to measure progress

AI visibility is harder to track than rankings, but it isn't impossible:

• Manual prompt checks: run the same 20 buyer questions monthly across ChatGPT, Perplexity, Gemini and Google. Record whether your brand appears and which sources get cited.
• Referral traffic: watch analytics for visits from chatgpt.com, perplexity.ai and similar referrers.
• Branded search: rising branded search volume often follows increased AI mentions.
• Classic rankings: pages climbing for informational queries usually show up more in AI answers too.

The bottom line

AI search rewards the same thing good link building always did: being referenced by sources people trust. The difference is that context, consistency and topical relevance now carry even more weight.

If your link strategy is "get as many high-DR links as possible", AI search will expose it. If it's "get described accurately on the publications our buyers read", you're already building for what comes next.