Question tracking is the repeated execution of a defined set of saved questions on selected AI platforms. Each question-and-platform pair creates a separate tracking run, allowing answers and derived metrics to be compared over time.
In plain terms
A monitoring question is sent as stored; GetCited.me does not inject the website, brand, competitors, or positioning into the provider request.
After the provider returns an answer, a separate analysis step applies the monitored brand context to validate identity, detect competitors, classify sentiment, estimate share of voice, and normalize citations.
Repeated results are still samples: model updates, web-search results, location, and normal response variation can change an answer between runs.
Why it matters
Trends, not snapshots, show whether visibility work is moving the monitored answers — essential when responses fluctuate run to run.
Scheduled tracking can catch losses, such as a competitor overtaking you on a key question, early enough to respond.