The V-Score is a metric for AI search visibility. It measures how — and how prominently — your brand appears when AI models answer questions relevant to your business. It's a number from 0 to 100, where 0 means you're not mentioned at all and 100 means you're the first recommendation, with a URL cited, consistently across models.
The problem it was built to solve
SEO gave us Google rankings, impressions, and click-through rates. These metrics are useful because they're concrete, trackable, and tied to search behaviour that matters.
AI search has no equivalent native metric. You can ask ChatGPT whether your brand appears in an answer, but:
- Responses vary between sessions
- One query isn't representative of how you'd perform across the queries your customers actually ask
- There's no structured way to compare your result to a competitor
- There's nothing to track over time
The V-Score is a response to that measurement gap. It standardises how AI visibility is measured across multiple queries and multiple models, and produces a score that means the same thing each time you run it.
How it's calculated
The audit runs a set of queries representative of what buyers in your category actually ask — "what's the best [tool] for [use case]?", "recommend a [service] in [city]?", and so on. For each query, across each model (ChatGPT, Claude, Perplexity, Gemini), the response is checked for your brand and scored:
- Not mentioned → 0 points
- Mentioned late in the response → 40 points
- Mentioned mid-response → 60 points
- Mentioned first → 85 points
- Mentioned first with URL cited → 100 points
These per-query scores are weighted by query intent and rolled up into a per-model V-Score and an aggregate V-Score across all models.
What a typical score looks like
Most brands score in the 10–40 range when they first run an audit. This reflects the reality that AI visibility, like search visibility, has a long tail: a small number of brands dominate the recommendations for any given category, and most brands sit at various points below them.
A score of 0 doesn't mean the brand doesn't exist — it means AI models aren't confidently recommending it for the specific queries tested. A score above 60 indicates consistent, prominent mentions. A score above 80 is strong: the brand is reliably first or near-first across most queries.
What the V-Score doesn't tell you
The V-Score is a visibility metric, not a business outcome metric. A higher score means you appear more prominently in AI responses to relevant queries. It doesn't directly measure revenue, conversions, or the quality of traffic from AI mention.
It also doesn't tell you which queries to target, or what specific changes to make to improve it. Those questions require looking at the query-level breakdown and understanding the gap between your score and your competitors' scores on the same queries.
What moves a V-Score
The factors that improve AI visibility — and by extension the V-Score — are the same as the factors that determine whether AI models know who you are and trust the recommendation:
- Third-party editorial coverage, especially in publications AI models retrieve from
- Review platform presence (volume and recency)
- Category association strength across multiple independent sources
- Content that earns citations from other writers and publications
A V-Score trend over time is more useful than a single snapshot. Measuring the same queries month over month tells you whether the work you're doing is translating into improved AI visibility — or not.
Run your free audit and get your V-Score across ChatGPT, Claude, Perplexity, and Gemini.