Catching Shifts Before They Hit You · 4.11
Watching For AI Platforms Changing How They Answer
Re-running the same set of AI questions periodically and noting when the answer itself changes — a platform updating how it decides who to recommend, with no notice to anyone.
“OpenAI, Google, and Perplexity can each shift how they decide who to recommend overnight. You can't stop the goalposts moving, but you can notice the moment they do, instead of finding out three months later from a quiet enquiry slump.”
Why this matters
The customer’s side
Not directly customer-facing — this is a monitoring habit that protects everything the GEO guide already helped you build, rather than something a customer ever sees.
The search engine’s side
Signal — outside your control, but not outside your awareness
This is the one thing genuinely outside anyone's control — see our Guarantees page for why we don't pretend otherwise. What's within your control is noticing the change fast enough to adapt, rather than being the last to find out.
How to check yours
- 1
Keep the same fixed set of 3–4 realistic questions from earlier in this course, and re-ask them to ChatGPT, Perplexity, and Google AI Overviews.
- 2
Compare the new answers to what you recorded last time — a different structure, different names, or a different tone is worth noting even if you're still named.
How to fix what you find
If your visibility has genuinely dropped since your last check
revisit your GEO fundamentals — trust signals and schema can need re-verifying after a platform update, even if nothing on your own site changed.
Keeping on top of it
Monthly, alongside your AI mentions comparison — the two checks pair naturally since they use the same questions.
How Reyse handles this
Done With You
we help you build the habit of re-testing consistently, so a platform shift is never a surprise.
Done For You
we monitor for exactly this every month and re-verify your GEO fundamentals whenever a platform's behaviour genuinely shifts.