The measured state of AI search · No. 02 · June 2026
We asked 4 AI engines for the best observability tool. They disagreed. - and what it means for your AI visibility
Here’s what’s new from our desk this month, the AI-search shift worth your attention, and one fix you can ship.
The one thing
Across AI engines, roughly 95% of the citations behind answers come from third-party pages, not a brand’s own site (Otterly, State of AI Search) — so being measured and cited where AI looks matters more than another homepage tweak.
Teardown of the week
❯ best observability tool for a Series A startup
Ask the same buyer question across engines and the named leaders differ — no single vendor leads on all five. The takeaway isn’t which tool an engine prefers; it’s whether your share of model moved, and on which engines, for your real buyer prompts.
One fix you can ship
Baseline one buyer prompt before you change anything.
Pick one prompt in your category, run it 10+ times each across the engines your buyers use, and record which sources get cited. That number tells you where you stand.
From our desk this month
We asked 4 AI engines for the best observability tool. They disagreed.
Original data: share-of-model across ChatGPT, Perplexity, Gemini, Google AI.
Read it →See where the engines put you
Measured across all five, free.
Get a free teardown