Perplexity prompt volatility tracking
perplexityprompt_volatilityAnonymous
Perplexity is the least volatile engine we track, which is exactly why we use it to sanity-check new sampling logic before rolling it to noisier surfaces.
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
Low volatility can lull teams into single-reading reporting. It is lower, not zero, and the same prompt still returns different sources between runs.
What the API returns
A single collection call against Perplexity for the prompt volatility field, sampled n times and returned as one object per prompt.
{
"engine": "perplexity",
"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"
}
}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. |
Perplexity prompt volatility FAQ
- Is Perplexity stable enough for n=1?
- No. It is the most stable engine we cover and it still disagrees with itself between samples.
- What n do you recommend?
- Twenty is usually sufficient. The recommended_n field confirms it per prompt.
Calibrate your sampling on the quietest engine first
We'll run a volatility baseline on Perplexity before you scale to the noisy ones.
Other Perplexity data
Across every engine
Compare how prompt volatility behaves on every engine we track.
Prompt volatility across all engines →