Glossary

A–E

AI Score — The 0–100 grade for one probe, and the average of those grades on the panel. For discovery prompts it is position + sentiment + frequency bonus; for branded prompts it is derived from the verdict. → AI Score

Adjustable (region signal) — An engine whose retrieval genuinely runs from the target country. The only class used for cross-region comparison. → Engines & regions

Authority Media — Citation bucket for news, magazines and third-party review sites. → Citations & sources

Branded — The bucket containing comparison and trust prompts: prompts whose text already names your brand. → Prompt intents

Category — A discovery intent. Broad head-term questions that name no brand (“best massage gun 2026”).

Citation — A source URL an engine cited in its answer. Stored per probe and aggregated by domain on the Sources tab.

Comparison — A branded intent. Your brand placed against a named rival; scored win / tie / lose.

Cross-region average — The baseline each region’s delta is measured against. The mean of the regions’ scores, weighting regions equally rather than probes. → Averages & denominators

Discovery — The bucket containing category and scenario prompts: prompts that never name your brand, so a mention is earned visibility.

Engine — One AI assistant (ChatGPT, Gemini, Claude, Perplexity, Google AI Overview, Google AI Mode, Grok, DeepSeek, Doubao).

Entity resolution — Folding competitor surface forms (“Timberland PRO”, “timberland”) into one canonical parent brand before counting. → Competitors

F–M

Final Score — The headline number: 0.8 × average AI Score + 0.2 × SEO Score, with the SEO term dropped when no audit exists. → Final Score

Fixed (region signal) — An engine whose retrieval is inherently locked to one market. Grouped under that market rather than the primary one.

Frequency bonus — 0–8 points for being referenced repeatedly or across several sections of one answer. Discovery prompts only.

Global engines — The scope containing engines with no location signal.

Intent — What a prompt measures: category, scenario, comparison or trust. → Prompt intents

Mention Rate — Share of discovery conversations that mention your brand. Branded prompts are excluded and reported separately.

Mentioned — The analysis model found your brand in the answer, by name, alias, URL or a pronoun clearly referring to it.

N–R

Not analyzed — A probe whose grading failed after retries. Its answer and citations are kept, but it is excluded from every metric. → Averages & denominators

Owned Assets — Citation bucket for your own site and domains.

Position Score — 0–85 for how prominently your brand sits in an answer: by rank in an ordered list, or by role in prose. Discovery prompts only.

Primary market — The first region in a target’s list. Region-agnostic engines run there.

Probe — The atom of measurement: one prompt × one engine × one region × one execution, plus the graded answer it produced.

Probe group / run — Every probe launched together, plus the statistics and interpretation computed over them. One point on the score history chart.

Prompt — A question a customer would ask an assistant, stored in a target’s library.

Region — A market, as a two-letter country code. A property of retrieval, not of the model.

Run — See probe group.

S–Z

Scenario — A discovery intent. Long-tail questions naming a persona, use case, budget or constraint, with no brand named.

Scorable — A probe with a usable model judgement. The base set for every average.

SEO Score — Mean of four Lighthouse categories from a PageSpeed Insights run of your homepage. Site-level, so identical in every region. → Site Audit

Sentiment Score — 0–15 for tone toward your brand, averaged over answers that mention you and displayed rescaled to 0–100.

Share of voice — Your share of discovery conversations against competitors. → Competitors

Target — One brand you monitor: name, URL, markets and brand context.

Trust — A branded intent. Legitimacy and reputation questions naming your brand; scored legit / mixed / scam.

User Consensus — Citation bucket for Reddit, YouTube, LinkedIn and other social/UGC platforms.

Verdict — The discrete outcome recorded for a branded prompt, and the basis of its AI Score.

Visibility % — In the competitor leaderboard, the share of conversations in an engine’s bucket that named a given brand.