limitations

what this project does not show.

—it does not prove that any recommended fix improves visibility. the one before/after data point on the findings page is observational, not experimental.
—scores are a snapshot of three engines (chatgpt, gemini, claude) at a single point in time. providers update their models continuously; a score from last month may not hold today.
—presence, position, sentiment, and share of voice are the four dimensions this project chose to measure. they are not validated proxies for real-world outcomes like sales, awareness, or trust.
—this is coursework-adjacent student research, not a peer-reviewed study. no claim here has been through academic review.
—do not use this tool’s output as the basis for a business, financial, or legal decision.