Nostimates

Gemini app share of voice tracking

geminishare_of_voiceAccount required

Gemini share of voice is the share of samples naming a brand against its competitor set, reported with a 95% interval and split by grounded and ungrounded runs because the two behave differently.

How it renders

Derived across the sample set from mentions and citations combined.

Sampling

This is the field where n matters most. At n=1 the metric is noise; at n=30 the interval is typically tight enough to defend in a client report.

Gotcha

Mixing grounded and ungrounded runs into one share figure hides the mechanism. A brand can dominate the model's memory and lose every grounded retrieval, or the reverse.

What the API returns

A single collection call against Gemini app for the share of voice field, sampled n times and returned as one object per prompt.

{
  "engine": "gemini",
  "field": "share_of_voice",
  "prompt": "best ai visibility tracking tool",
  "country": "GB",
  "language": "en",
  "n": 30,
  "share_of_voice": [
    {
      "brand": "Nostimates",
      "share": 0.34,
      "ci95": {
        "low": 0.2,
        "high": 0.52
      },
      "n": 30
    },
    {
      "brand": "Competitor",
      "share": 0.21,
      "ci95": {
        "low": 0.1,
        "high": 0.38
      },
      "n": 30
    }
  ]
}
GET /v1/collect?engine=gemini&field=share_of_voice

Response fields

KeyTypeNotes
brandstringBrand from the configured competitor set.
sharefloatPoint estimate across the sample set.
ci95object95% Wilson interval as { low, high }.
nintegerSamples behind the estimate.
delta_priorfloat | nullChange against the previous run of the same prompt set.

What it costs

Derived at no extra credit cost, but it scales with n — the interval is bought with samples.

See credit pricing →

Gemini app share of voice FAQ

Why split grounded and ungrounded share?
They are different systems. One reflects training data, the other reflects retrieval. Averaging them tells you nothing actionable.
How many samples does Gemini need?
More than the anonymous engines. We usually recommend n=40 because grounding variance adds a second source of noise.

Report Gemini share with the caveats built in

We will size the sample your prompt set needs and quote the credits against it.

No newsletter. One reply from a human.

Across every engine

Compare how share of voice behaves on every engine we track.

Share of voice across all engines →