Google still handles the majority of web queries in 2026, but AI search is taking an estimated 10–15% of informational intent — and share is shifting faster than forecasters predicted. Here's the current breakdown across the six engines that matter for SaaS visibility, with sources and what it means for your AEO strategy.
[next-mdx-remote-client] error compiling MDX: Unexpected character `1` (U+0031) before name, expected a character that can start a name, such as a letter, `$`, or `_` More information: https://mdxjs.com/docs/troubleshooting-mdx
Run a free AI Recommendation Audit across 6 engines. See your biggest visibility gaps and what to fix first.
Perplexity is the AI search engine most aligned with B2B SaaS buying intent. Users who reach Perplexity are disproportionately technical, disproportionately research-intent, and disproportionately pre-purchase. Perplexity disclosed 780 million queries in May 2025 and has been growing quarter over quarter. For SaaS visibility, Perplexity per-query is higher-value than any other engine despite its smaller total share.
This is separate from Gemini-powered AI Overviews inside Google Search. The standalone Gemini app (gemini.google.com) has traction mostly inside Google Workspace accounts and Android deep integrations. Estimated share is small but growing. For SaaS founders, Gemini answers draw on similar data sources as Google search but with different ranking — worth monitoring separately.
DeepSeek reached the top of US App Store downloads in January 2025 and has been a major force globally. Inside mainland China it holds a meaningful share of AI search usage alongside Kimi, Doubao, and Qwen. For any SaaS serving Asian markets, DeepSeek visibility is non-optional.
Anthropic's Claude is used more for conversation, coding, and long-form work than for search. Its search share is small. But Claude's reasoning quality means it is the model many advanced users turn to for deep research queries — and when cited inside a Claude response, the conversion rate to a paid SaaS is higher than any other engine in our internal audits.
Mistral's Le Chat, You.com, Brave Leo, Kagi Assistant, and Andi collectively hold single-digit share but over-index on privacy-aware and European audiences. If your SaaS has significant EU revenue, these engines deserve tracking.
If you run a SaaS and want to be recommended when buyers ask an AI "best tool for X," here is what the share numbers translate into operationally:
If you want to know your current share of voice inside AI search, the practical path is:
That comparison — citation count and accuracy vs. competitors — is your current share of voice baseline. Improve it by fixing the underlying signals (schema, freshness, answer-first content, Bing indexing).
Share numbers in this post are triangulated from vendor disclosures (OpenAI blog, Perplexity updates, Anthropic, Google Cloud blog), third-party traffic panels (Similarweb, SEMrush, Datos.live Q1 2026 reports), and Ahrefs + Semrush studies on AI Overview citation patterns. Numbers are best-effort estimates for Q1 2026 and will shift as vendors release new data. If you want a current audit of your own share of voice across these six engines, EurekaNav runs one on demand.
Sign in to leave a comment
No comments yet. Be the first to share your thoughts!
2026/04/23
Run a free AI Recommendation Audit across 6 engines. See your biggest visibility gaps and what to fix first.
