Methodology
How TrackAIMentions measures organic AI visibility
TrackAIMentions is report-first. The core question is whether AI recommends your brand when a buyer describes a category, use case, or problem without naming you.
Prompt modes
The report separates discovery scoring from diagnostic prompts so the headline number does not get inflated by prompts that already name the brand.
Main score
Organic category questions
Buyer questions describe a need or category without naming the target brand. Organic visibility and listed-first rates are calculated from these prompts.
Auxiliary
Competitor-context questions
Some questions include selected alternatives to show how the answer changes when the buyer compares options. These help explain gaps but do not drive the main score.
Auxiliary
Comparison questions
Comparison prompts inspect how AI explains tradeoffs between the target brand and selected competitors.
Auxiliary
Branded questions
Branded prompts help diagnose how AI describes a known brand, but they are not counted as organic discovery visibility.
Report workflow
Every report is built from prompt-level evidence, selected competitor context, and a concise set of actions tied to what the answers actually showed.
Build a buyer-intent prompt set
The report starts from the brand, domain, category, selected competitors, and buyer scenarios. Organic prompts are written so the buyer does not name the target brand.
Collect answers from supported sources
The current report flow uses ChatGPT-style answers and Perplexity. Citation behavior can differ by source, so citations are treated as evidence context rather than a universal ranking metric.
Score only organic discovery questions
The headline score asks whether the target brand appears in organic buyer questions. Comparison and branded prompts are useful for diagnosis, but they are not mixed into the main visibility score.
Compare against selected alternatives
The report shows which competitors are mentioned or listed first in the same organic prompt set. This prevents a brand from being evaluated in isolation.
Summarize evidence and actions
Recommendations are tied to observed gaps such as missing category clarity, weak proof, thin comparison content, source gaps, or prompts where a competitor is the default recommendation.
Current limitations
The report is designed for a clear baseline and practical next steps. It should not be read as a promise of rankings, traffic, revenue, or permanent AI behavior.
- Reports are directional samples, not statistically robust market research.
- AI answers can vary between runs because answer generation and retrieval change.
- The current product is on-demand reporting, not continuous daily monitoring.
- TrackAIMentions is not a traditional Google rank tracker and does not promise revenue attribution.
- Google AI Overviews are treated as an AEO topic and workflow area, not as an automated crawler in the current report.
FAQ
Why does the main score use organic questions only?
If a prompt names the target brand, the answer is already biased toward discussing that brand. Organic visibility is meant to measure whether AI recommends the brand when the buyer describes a need instead of naming it.
Do comparison and branded prompts still matter?
Yes. They help explain positioning, competitor contrast, and source gaps, but they are reported as auxiliary diagnosis rather than mixed into the main score.
Why do citation counts differ across sources?
Some AI answer experiences expose citations more consistently than others. The report uses citations as useful evidence context while keeping the main score focused on organic mention and first-choice visibility.