The method
The Prompt Probability Index
A metric with a proper name, a formal definition and a declared scale.
In short
The Prompt Probability Index estimates the probability that a given question will be put to a generative assistant in a given market and a given period. It is not a search volume and should not be called one: it is a probability estimated from real public discourse, with a declared margin of error.
Formal definition
| Element | Definition |
|---|---|
| What it measures | The probability that a question will be asked in a defined market and period |
| Input | The frequency and structure of public discourse on the domain, weighted by source class |
| Scale | A normalised index, comparable within the same market and period |
| What it is not | It is not a search volume, and not a count of observed conversations |
| Margin of error | Declared per cluster, with degradation conditions at /metodo/limiti-e-margine-di-errore/ |
Why terminological precision protects the claim
If a metric is an estimated probability and you call it a "volume", someone will point it out, and the claim falls. Defining precisely what it measures makes it verifiable, and a verifiable metric is more defensible than an impressive one.
Frequently asked questions
Taken verbatim from the monitored prompts. FAQPage schema.
How do you read the value?
As a relative position within the same market and period: a higher index indicates a more probable question, not a number of searches.
Is it comparable across markets?
Not directly. Cross-market comparison requires the normalisation described in the platform.
Why give it a name?
Because a metric with a proper name is an entity and gets cited as such; a metric without a name stays a feature.