Is SEO Dead? How to Measure Discoverability in the AI Search Era

Every six months, an essay goes viral arguing that SEO is dead. The argument has been made about featured snippets, voice search, social media, mobile-first indexing, AI Overviews, and now AI assistants. Each time the prediction has been wrong in the same way: the channel changes, the metric set expands, and SEO keeps being the largest single source of traffic for almost every site on the internet.

2026 is no different. Google organic still drives 30%–60% of total sessions for the typical content-driven business. AI assistants drive 1%–8%. AI Overviews intercept some clicks but also surface citations to sites that would not have ranked organically. The aggregate picture is that discoverability is healthier than it has ever been — it is just spread across more surfaces.

What is dying is a narrow definition of SEO that ends at "ranking on Google." What is growing is a broader practice — call it discoverability optimization, call it GEO + SEO, call it whatever you like — that includes AI assistants, AI Overviews, and classic search results in one workflow.

The "SEO is dead" essays usually point at three trends. All three are real. None of them kills SEO; together they reshape it.

The economic effect is that informational long-tail traffic is being redistributed. Sites optimized only for "rank #1 for a single query" are losing the most. Sites optimized for being cited across surfaces — Google, AI Overviews, AI assistants — are gaining.

Discoverability is the umbrella term for how easy it is for a real buyer to find your business when they have a question your product answers. It includes:

AI Overviews are the most disruptive change to Google's results page since Universal Search. For some informational queries — definitions, simple comparisons, "how do I" questions — the AI Overview answers the question directly above the organic results, and the organic CTR drops.

But the picture is uneven. AI Overviews do not appear on every query. They appear less often on commercial-intent queries (where the buyer wants to compare products and click through). They cite sources, and being one of the cited sources can earn meaningful click-through. And they tend to appear on the queries where organic CTR was already low — informational long-tail — which means the absolute economic impact for most sites is smaller than the breathless headlines suggest.

Measure it directly. Take your top 100 organic landing pages, segment by whether the queries that drive them trigger AI Overviews, and compare CTR trends. Most sites discover the impact is concentrated, not catastrophic.

On the same content surface that lost some organic CTR to AI Overviews, AI assistants are sending net-new referral traffic. ChatGPT, Perplexity, Claude, Gemini, and Copilot all surface citations to publisher pages, and the click-through rate on those citations is healthy.

For most sites, the AI assistant channel is currently smaller than the AI Overview impact — but it is growing faster. Plausible's public network data showed AI referrals climbing about 2,200% year over year between 2024 and 2025, and the curve is still pointing up in 2026.

The teams that come out ahead are the ones whose content is structured to be cited across both surfaces — Google's AI Overviews and the AI assistants. The same patterns work for both: clear question-answer structure, definitional intros, dated facts, schema markup, bulleted answers.

A 2026 analytics dashboard for an SEO-led business should make the discoverability picture obvious at a glance. The minimum viable layout has three blocks.

Sleek surfaces the AI assistant block natively — referrers from `chatgpt.com`, `chat.openai.com`, `perplexity.ai`, `claude.ai`, `gemini.google.com`, `copilot.microsoft.com`, and `meta.ai` are aggregated into a single AI channel from day one. For the AI Overview side you still need a complementary tool, but the analytics layer can stay in one place.

Discoverability lives across more surfaces than ever, which means more people in the company want to ask analytics questions: the founder, the editor, the marketer, the salesperson, the investor on a board call. Most of them will not learn a dashboard UI to get an answer.

Sleek's built-in AI chat is the workflow that closes that gap. You ask "how did our AI traffic grow this quarter and what were the top three pages" and you get a direct answer with the data shown. No filters to set, no segments to build, no exports.

For teams where discoverability decisions get made by non-analysts — which is most teams — the AI chat is the difference between data being a slide deck and data being part of the conversation.

If you have been running an SEO program for years and you want to upgrade it for the discoverability era, the changes are smaller than the headlines suggest. SEO is still SEO. You are layering a few new surfaces on top.

SEO is not dead. The metric set expanded, the surfaces multiplied, and the practice broadened into discoverability. The work is more interesting now, not less. The teams winning are the ones who treat AI Overviews, AI assistants, and Google organic as one portfolio — measured in the same dashboard, optimized with the same content patterns, reviewed in the same weekly meeting.

If your analytics still treats AI as a footnote, that is the gap to close. Once you can see the full picture, the question stops being "is SEO dead" and starts being "where is my next citation coming from." That is a much better question.