AI search competitive intelligence

Track your brand's share of voice across ChatGPT, Perplexity, Claude, Gemini, and Google AI Overviews

See which brands AI engines recommend for the questions buyers ask. Foglift keeps the prompt, answer, competitor mentions, cited URLs, sentiment, and measurement window behind every score so your team can see what changed and choose the next action.

ChatGPTPerplexityClaudeGeminiGoogle AI Overviews

Start monitoring free Explore AI monitoring

A score you can audit

The denominator matters as much as the percentage

The basic formula is brand mentions divided by all tracked brand mentions, multiplied by 100. That number is useful only when every comparison uses the same prompt set, competitors, engines, and measurement window. Adding easier prompts or removing a strong competitor can raise the percentage without improving real buyer visibility.

Foglift stores the evidence at prompt level. Report every engine separately before looking at a combined view, because ChatGPT, Perplexity, Claude, Gemini, and Google AI Overviews can retrieve and cite different sources for the same buyer question.

AI share of voice formula

Your brand mentions

All tracked brand mentions

× 100 for the declared comparison window

Zero-mention answers remain in the evidence set, while the share calculation uses the brands actually mentioned. Pair the percentage with answer coverage so silence is not hidden.

Measure the evidence behind each win

A useful share-of-voice tool connects the headline percentage to the answer, source, and competitor evidence that explains it. That makes the metric actionable for content, product marketing, and digital PR.

Signal	What Foglift keeps	Decision it supports
Brand mention	Whether your brand appears in each answer	Shows where buyers can discover you and where you are absent.
Competitor mention	Every named competitor in the same answer set	Keeps the share-of-voice denominator visible and comparable.
Cited source	The exact URL and domain cited by the engine	Reveals which owned or third-party pages support the winning answer.
Answer sentiment	Positive, neutral, or negative brand framing	Separates a useful recommendation from a mention that harms trust.
Prompt history	Prompt, engine, answer, and run date	Makes changes auditable instead of treating one volatile answer as a trend.

From score to action

A repeatable AI share-of-voice workflow

Consistency turns volatile generated answers into a useful trend. Keep inputs fixed, retain the raw evidence, and review weak commercial prompts before changing the measurement design.

Step 1

Define the market

Choose the commercial questions buyers ask and the competitors they are likely to compare. Keep this prompt set stable between reporting windows so movement means something.

Step 2

Run the same prompts

Monitor those questions across the engines your buyers use. Foglift keeps the prompt and engine attached to every answer instead of blending away the underlying evidence.

Step 3

Count mentions consistently

For each answer, record your brand, every tracked competitor, cited sources, position, and sentiment. The score should be reproducible from those rows.

Step 4

Fix the losing prompts

Use cited pages and competitor patterns to decide whether the next step is a clearer product page, stronger comparison evidence, technical cleanup, or third-party coverage.

Find prompt gaps

Separate category discovery, comparison, use-case, and branded prompts. A healthy overall score can still hide absence on the few questions closest to a purchase decision.

Trace winning sources

See whether a competitor wins through its product page, documentation, original research, customer proof, or a third-party comparison. The cited source points to the evidence gap worth fixing.

Watch durable movement

Compare stable windows rather than reacting to one answer. Prompt history distinguishes a persistent gain from normal engine variation and gives teams an audit trail for releases and campaigns.

Keep the methodology beside the product

Need the calculation details, worked examples, and evaluation checklist first? Read the methodology guide, then use recurring monitoring to measure the same prompt and competitor set across supported engines.

Read the AI share-of-voice methodology

Frequently asked questions

What is share of voice in AI search?

Share of voice in AI search is the percentage of tracked brand mentions that belong to your brand across a declared prompt set, competitor set, engine set, and measurement window. It measures presence inside AI-generated answers, not traditional search traffic.

How do you calculate AI share of voice?

Divide your brand mentions by all tracked brand mentions for the same prompt set, competitors, engines, and measurement window, then multiply by 100. Keep the denominator and answer history available so a team can reproduce the score.

What is a good AI share of voice score?

There is no universal evidence-backed target. Compare your score against the same competitors and commercially important prompts over time. A rising score is useful only when the denominator, engine mix, and sampling cadence remain stable.

Why should each AI engine be reported separately?

AI engines retrieve and cite different source sets, so a blended total can hide a weak platform. Report every engine separately first, then use a combined view only when its weighting is declared and supported by your audience evidence.

How is AI share of voice different from SEO share of voice?

SEO share of voice typically estimates visibility from ranked search results and keyword volume. AI share of voice counts brand presence inside generated answers. An engine can recommend a brand or cite a third-party page even when the brand does not hold the equivalent classic ranking position.

See where your brand wins and disappears

Start with weekly Perplexity monitoring on Free, or compare all five supported engines on a paid plan.

Start monitoring free