A colleague and I want to compare Grok’s recommendations for the same laptop brief. We plan to use identical wording, but our accounts have very different conversation histories: I mostly ask about software development, while they ask about photography.

Is starting a new chat a meaningful clean slate? I don’t want to label a difference as a reasoning problem if account preferences or other personalization could explain it.

For a fair comparison, what should we record or disable beyond the visible conversation history?

I’m looking for a practical way to test Grok account personalization, not a guarantee that repeated answers will be identical. Would matching the model, mode, language, and search settings be enough, or are there other account-level controls to check in the version we’re using?

If you’ve tried cross-account AI testing, how did you distinguish persistent preference effects from ordinary variation between responses?