1 minute readReference note
What should be recorded in a search and AI reputation baseline?
In brief
A search and AI reputation baseline records the information important audiences can encounter under defined conditions. It needs more than a screenshot or headline score. Useful dimensions include inclusion, factual accuracy, narrative context, visible source support, and the significance of any gap for the decision being studied.
The scope should identify entities, audiences, markets, languages, devices or products where relevant, named peers, and a stable question set. The record preserves observation dates and exact outputs so later reviewers can distinguish a new finding from a changed collection method. A competitive benchmark adds equivalent tests for alternatives rather than simply gathering favorable examples about the brand.
The baseline also needs a reference for what is true now and what was true historically. Without that, accuracy labels become opinions about preferred wording. Changes to the question set can improve coverage, but a separately reported fixed comparison set preserves trend meaning. The baseline's conclusion should remain proportional to its scope: a defined sample can reveal actionable problems without proving how every person or AI interaction represents the company.