Why does AI Overview volatility matter for budget?

It tells you where to spend samples. Stable prompts can run at n=10; volatile ones need n=40 to say anything.

How is volatility calculated?

Mean pairwise Jaccard overlap of citation sets plus variance in brand presence, banded into stable, moderate and volatile.

Google AI Overviews prompt volatility tracking

google_ai_overviewsprompt_volatilityAnonymous

Volatility tells you how much an AI Overview disagrees with itself. We return citation-set overlap across samples and the n required to reach a ±10 point interval on that prompt.

How it renders

Derived across the sample set.

Sampling

Volatility is what tells you the right n for every other field. We recommend using it to allocate credits: sample noisy prompts deeply, stable prompts lightly.

Gotcha

Volatility is not constant across a prompt set. Head terms are usually stable and long-tail commercial prompts are wildly unstable, so a flat n across your whole list wastes credits at one end and produces junk at the other.

What the API returns

A single collection call against Google AI Overviews for the prompt volatility field, sampled n times and returned as one object per prompt.

{
  "engine": "google_ai_overviews",
  "field": "prompt_volatility",
  "prompt": "best ai visibility tracking tool",
  "country": "GB",
  "language": "en",
  "n": 30,
  "volatility": {
    "jaccard": 0.42,
    "mention_variance": 0.11,
    "recommended_n": 75,
    "band": "moderate"
  }
}

GET /v1/collect?engine=google_ai_overviews&field=prompt_volatility

Response fields

Key	Type	Notes
jaccard	float	Mean pairwise citation-set overlap across samples.
mention_variance	float	Variance in brand presence across samples.
recommended_n	integer	Samples needed for a ±10 point interval on this prompt.
band	string	stable \| moderate \| volatile.

What it costs

Derived from samples you already pay for. No separate charge.

See credit pricing →

Google AI Overviews prompt volatility FAQ

Why does AI Overview volatility matter for budget?: It tells you where to spend samples. Stable prompts can run at n=10; volatile ones need n=40 to say anything.
How is volatility calculated?: Mean pairwise Jaccard overlap of citation sets plus variance in brand presence, banded into stable, moderate and volatile.

Spend your credits where the noise is

We'll run a volatility pass over your prompt set and hand back a per-prompt sampling plan.

Other Google AI Overviews data

Citations Brand mentions Sources Answer text Share of voice Sentiment Position Locale variance

Across every engine

Compare how prompt volatility behaves on every engine we track.

Prompt volatility across all engines →