The answer shows no web search: how to report what evidence remains
A saved AI answer with no search on record can mean three different things. Learn to tell an observed no-search from missing telemetry, with reporting language for each.
15 min read
Blog
AI Engines
All DiscoveredBy blog posts on AI Engines, newest first. Our blog covers AI search visibility and generative engine optimization.
Does adding “for a small team” change the question you are measuring? A prompt-edit decision tree
Adding “for a small team” to a tracked prompt can quietly turn it into a different question. A decision tree separates harmless rewording from a new constraint or audience.
15 min read
Source links, brand links and shopping cards in AI answers: how to code each one
A link in an AI answer can be a source, a brand's own site, or a product card. Code each one separately, so your report says what the answer actually showed.
14 min read
How to keep a measurement change log for AI engine changes
A copyable change log for the moments an AI engine, its collection method or your own setup changes, so you know which before-and-after comparisons need a note.
17 min read
Gemini app vs Gemini API answers: how to investigate a disagreement
Gemini (app) and Gemini (API) can answer the same prompt differently. Align the question, date and collection context first, then compare answer evidence with a worksheet.
15 min read
Why your manual ChatGPT check differs from a monitoring result: a discrepancy log
You asked ChatGPT and saw your brand. Your monitoring report says otherwise. Record seven variables, then compare like with like using this discrepancy log.
14 min read
Google AI Overviews, AI Mode and Gemini: plan separate measurements
Google AI Overviews, Google AI Mode, Gemini (app) and Gemini (API) are four different things to measure. A copyable measurement matrix shows what each can and cannot tell you.
14 min read