Every CMO I talk to wants to know the same thing: how big is this AI search thing, really? Is it hype, or is it already taking share from Google? Should I be building a budget line for AEO, or waiting a year?
Here are the 2026 numbers I trust — pulled from Semrush, Similarweb, Gartner, Forrester, and first-party platform disclosures, then cross-checked where possible. This is an annual update; we'll refresh it quarterly as new data lands.
User adoption
The raw usage numbers. These tell you the audience size on each platform — which is where citation impressions ultimately come from.
900M+
ChatGPT weekly active users
Source: OpenAI, Q1 2026
~17%
ChatGPT share of Google query volume equivalent
Source: Semrush analysis
800M+
Google AI Overviews monthly impressions
Source: Google I/O 2026
~50M
Perplexity monthly active users
Source: Perplexity, late 2025
~25M
Claude.ai weekly active users (estimated)
Source: Third-party estimates, Q1 2026
~30M
Deepseek monthly active users (rising)
Source: Public download data, 2026
Reading the numbers:ChatGPT is still the largest AI chat audience by an order of magnitude. But Google AI Overviews reach a bigger surface area because they're injected into existing Google searches — the billions of searches per day don't go away, they just get an AI summary on top. Perplexity, Claude, and Deepseek are all in the tens of millions and climbing.
Referral traffic
Usage is one thing. Traffic that actually lands on your site is another. This is where the real business signal lives.
+206%
YoY growth in ChatGPT referral traffic
Source: Semrush, 2025-2026
+145%
YoY growth in Perplexity referral traffic
Source: Similarweb, 2025-2026
+80%
YoY growth in Claude referral traffic (from a smaller base)
Source: Similarweb, 2025-2026
30-45%
Share of sessions from AI referrals now outperforming paid ads on conversion rate
Source: Multiple e-commerce aggregators, 2026
The high-conversion surprise:across multiple e-commerce and B2B aggregators, users who arrive via AI-chat citations convert 30-45% higher than paid ads. The hypothesis: they've already done their research and the AI has pre-qualified you. The click is much deeper in the funnel than a cold paid impression.
Search behavior shift
How users actually search is changing, not just which platform they use.
25%
Projected decline in traditional search by 2026
Source: Gartner
~60%
Commercial-intent queries now showing AI Overviews
Source: Rank-tracking tool aggregates, Q1 2026
40%+
Average CTR reduction on organic listings when AI Overview is present
Source: Seer Interactive, 2025
3.5x
Longer average query length in AI search vs. traditional Google
Source: Perplexity internal data, 2026
55%
Of Gen Z users report starting product research in AI chat rather than Google
Source: Edelman Trust Barometer 2026
Query length matters:AI search rewards specificity. Users ask complete questions in natural language — “which boutique consultancy in San Diego handles both website redesign and SEO for enterprise?” — rather than stripped keyword phrases. Pages written for these longer intent-rich queries get cited. Keyword-stuffed pages don't.
Business adoption
Where marketing and procurement budgets are moving.
72%
Of B2B buyers have used AI to research vendors in the past 90 days
Source: Forrester, 2026
~35%
Of enterprise marketing teams have AEO/GEO line items in 2026 budgets
Source: Gartner marketing survey, 2026
~$2-6B
Estimated size of AEO/GEO services market by end of 2026
Source: Industry aggregator estimates
100+
Dedicated AEO/GEO agencies with public service pages
Source: First Page Sage directory + manual audit
Enterprise momentum:when Gartner surveys show a third of enterprise marketing teams with explicit AEO line items, the category has crossed from “emerging” to “table stakes.” If your 2026 plan doesn't include AI visibility work, your competitors' plans do.
What the numbers actually mean for your business
Five takeaways that should drive 2026 planning:
- 1.AI search is no longer a future bet. Traffic exists today. It's meaningful. It converts at premium rates. Ignoring it = leaving money on the table every month.
- 2.Google still drives the volume, but the surface is changing. AI Overviews appear on most commercial queries. Your Google SEO plan must account for AI Overview citations, not just top-10 rankings.
- 3.Multi-platform matters. Optimizing only for ChatGPT is like optimizing only for Bing in 2015. Different platforms, different citation behaviors. Test on all five.
- 4.Long-form, specific content wins. Users ask longer questions; models reward pages that actually answer them with depth and specifics. Thin keyword-stuffed pages are obsolete.
- 5.The AEO services market is real. With $2-6B in projected 2026 spend, expect pricing and positioning volatility. Sort signal from hype carefully when hiring.
Methodology and caveats
Some of these numbers are well-disclosed (ChatGPT WAU, Google AI Overview coverage). Others are estimates from third-party analytics (Similarweb, Semrush). A few are industry rollups from aggregators. Where possible, I've noted sources. Where estimates conflict, I've used the midpoint of reputable sources.
Directional accuracy matters more than point precision here. The exact WAU of Claude doesn't change your 2026 plan. The order of magnitude does.
We'll refresh this post quarterly as new data lands and annually as a full rewrite. Current version: April 2026.
Building an AEO strategy from these numbers?
We redesign websites to capture AI search traffic — across ChatGPT, Claude, Gemini, Perplexity, and Deepseek. Strategy, design, SEO, AEO, and citation monitoring in one engagement.