Research

How big is the AEO problem, and why is it growing so fast?

Short answer: nobody has a precise number yet — this space is too new. Longer answer: here's what the actual research says, with sources, instead of a made-up statistic.

Published 2026-08 · Citely Research

Here's the current state of the field, as of mid-2026 — sourced, not guessed.

Search is evolving, not collapsing — and that's the more important story

As of mid-2026, Google still holds roughly 90%+ of search market share. Search didn't collapse. But ChatGPT alone reached an estimated 883 million monthly active users, and AI assistants are now a routine first stop for the kind of research/recommendation queries that used to open with "search for X near me." Google's own response has been folding AI Overviews directly into search results rather than losing that traffic to a standalone chatbot. The honest framing isn't "search is dying," it's "the answer is increasingly assembled by a model instead of just linked to" — even when the click still technically starts on a Google-owned surface.

The crawler-access problem comes before the content problem

Before any of that matters, there's a more basic gate: can the assistant's crawler even read the page? Cloudflare's popular "block AI bots" dashboard toggle, for instance, defaults to disallowing GPTbot, Google-Extended, and similar crawlers in robots.txt — meaning a business can rank perfectly well on Google and still be structurally invisible to an AI assistant, not because of anything on the page, but because the page was never fetched at all.

We checked this directly against a small sample of real local-business websites rather than assuming it — see our companion post for the actual numbers.

The gap is biggest for exactly the businesses we serve

The most directly relevant data point we found is a 2026 empirical study. Ranqo — an AI-visibility measurement company — published Generative Engine Optimization at Scale, analyzing 100,000+ real prompt responses across 100+ brands between March and May 2026. Its central finding is a visibility gap by brand size:

Global household names: ~73% visibility across sampled prompts. Mid-market/regional brands: ~44%. Niche and small brands: ~11%. — Kumar, P., Generative Engine Optimization at Scale: Measuring Brand Visibility Across AI Search Engines, arXiv:2606.20065 (2026)

That's roughly a 7x visibility gap between a household name and a small/local brand — measured this year, on live commercial engines, not modeled. The same paper found 78% of citations trace back to the brand's own corporate website (not third-party reviews or social), and that a listing's sentiment (positive vs. negative framing) flips about 6.7x more often between runs than whether the brand is mentioned at all — visibility is comparatively sticky, tone is not.

What actually moves the needle, per the latest research

A 2026 critical survey of the field's evidence base draws a clear line: topical relevance and whether a page is retrieved into a model's context at all are the biggest, most reproducible levers currently documented — no technique yet shows a stable, cross-platform effect on organic discoverability by itself.

"Topical relevance and context position are the most reproducible levers... query-document relevance and rank within context are major determinants." — Optimizing Visibility in Generative Engines: A Critical Survey of Generative Engine Optimization, arXiv:2607.14035 (2026)

In plain terms: get retrieved first — crawler access, topical match — and content quality helps you get quoted once you're actually in the room. That ordering is why our audits check crawlability before anything else.

Why "growing so fast" is the right framing

Three things are true at once: assistant-mediated answers are a real and rapidly growing share of how people research local businesses; the visibility gap between big and small brands is large and currently measured, not projected; and almost no small business has checked either one. That gap — real trend, current data, low awareness, a small set of tools addressing it — is the actual size of the opportunity.