70K Responses Over 28 Days: Measuring How Volatile ChatGPT and AI Overviews Really Are
gaganghotra_ · x · 2026-09-26
Glen Allsopp of Detailed.com ran a 28-day study tracking 1,300+ prompts daily across ChatGPT and Google AI Overviews — over 70,000 total responses checked.
- Prompts were mostly commercial, converted from keywords with traditional search volume into conversational questions (best binoculars for stargazing, best digital marketing agency in Sydney, best 3D design software, etc.).
- The goal: see what stays consistent when AI recommendations and cited links shift between repeated asks. A February pilot found exact matches were rare, but prominently named brands appeared frequently.
- Allsopp acknowledges limitations: tracked prompts don't fully mirror real usage, where requests often arrive as follow-ups; brands should track prompts specific to their own business.
Directly relevant for anyone doing AI visibility tracking (GEO/AEO) and measuring brand placement in AI answers.
More from Models
- Opus 5.5 does worse and costs more at Max thinking — Medium beats Max on hard benchmark — davidyin44 · 2026-09-26
- LibertAI ships open-weight Deem 9B on Qwen3.5, trailing Jev 68.9% vs 74.1% — Pokenhagen · 2026-09-26
- Peter Steiberger says he codes with Codex plus an OC harness, not Claude — steipete · 2026-09-26
- Why Anthropic lags OpenAI on math: compute constraints and 400k GPUs coming online — haider1 · 2026-09-26
- Hytale WorldGen V2 face-off: Claude Opus 4.6 vs Opus 5.5 compared — Angaisb_ · 2026-09-26
- OpenAI's rumored persistent agent "o" may tie to old "rebranding to O" report — Dullydude · 2026-09-26