Methodology

1. What we measure

Each scan runs a battery of real buyer-intent questions — the same kind of question a real prospect would actually ask an AI assistant, not a synthetic "does X exist" prompt — against ChatGPT, Claude, Gemini and Perplexity. We record whether your brand is named, where it ranks relative to competitors, and how it's framed (sentiment, objections, pricing perception, and more).

2. Sequential sampling, not a fixed sample size

We sample prompts in rounds across every engine and stop as soon as your brand's mention-rate estimate reaches a target precision (a 95% Wilson confidence interval of a set half-width) — or the prompt pool is exhausted. A clear-cut brand (named almost always, or almost never) converges quickly; a contested one keeps sampling toward the cap. This means the precision behind your headline number is a guarantee, not a side effect of however many prompts happened to run.

3. Confidence intervals, not point estimates

Every score ships with a real 95% confidence interval, built by combining three independently measured uncertainty sources in quadrature — not stacked naively, and not invented:

• Rerun variance — how much the exact same question, asked twice, moves just from model stochasticity.
• Per-engine extraction reliability — measured against a gold-labeled dataset of real AI answers, so we know how often our deterministic parser correctly detects a mention, a rank, or a sentiment.
• Category-phrasing sensitivity — a near-synonym category can ask AI a genuinely different question. We measured this directly: the same real brand scored differently under two near-synonym category phrasings, with non-overlapping confidence intervals. That's real AI behavior, not our measurement noise, and the interval accounts for it.

4. Deterministic scoring

Given the same answers, the scoring engine always produces the same score — there's no hidden randomness in how we grade an AI's response once it's been collected. The uncertainty in your interval comes entirely from the three real sources above, not from an unstable scoring formula.

5. What this doesn't do

A score is a point-in-time snapshot. AI answers shift as the underlying models and their sources update, so a re-scan next month can move for reasons that have nothing to do with anything you did — that's why trends across scans matter more than any single number. We're also honest about the limits of automation: our extraction reliability is measured, not assumed to be perfect, which is exactly why it's folded into your confidence interval instead of ignored.

6. Data sources

Every scan queries ChatGPT, Claude, Gemini and Perplexity directly — real API calls to real models, not scraped or cached responses standing in for a live answer. Growth-tier plans and above add real sentiment measured from actual Reddit and X posts, kept as a separate, clearly labeled signal rather than blended into the AI-engine numbers.