Microsoft Copilot sentiment tracking
Copilot's answers are terse, so sentiment tends to be either a clear recommendation or nothing at all — a cleaner binary than the hedged Google surfaces.
How it renders
Derived from the answer text with the evidence sentence retained.
Sampling
Classifications are reported as distributions across n, because the same engine will frame the same brand differently between samples.
Gotcha
What the API returns
A single collection call against Microsoft Copilot for the sentiment field, sampled n times and returned as one object per prompt.
{
"engine": "copilot",
"field": "sentiment",
"prompt": "best ai visibility tracking tool",
"country": "GB",
"language": "en",
"n": 30,
"sentiment": [
{
"brand": "Nostimates",
"label": "recommended",
"confidence": 0.86,
"distribution": {
"recommended": 0.6,
"neutral": 0.33,
"hedged": 0.07
}
}
]
}Response fields
| Key | Type | Notes |
|---|---|---|
| brand | string | Brand the classification applies to. |
| label | string | recommended | neutral | hedged | negative. |
| confidence | float | Classifier confidence for this sample. |
| evidence | string | The sentence the label was drawn from. |
| distribution | object | Label shares across the n samples. |
What it costs
Add-on field. Priced per collection at a small multiple of the base credit.
See credit pricing →Microsoft Copilot sentiment FAQ
- Is Copilot sentiment reliable?
- Per sample, less so than on longer-form engines. Across n, the distribution is sound.
- Do you return confidence?
- Yes, per label, so low-confidence classifications can be filtered.
Add Copilot sentiment to your coverage
Send a prompt set and we'll return the distribution with evidence.
Other Microsoft Copilot data
Across every engine
Compare how sentiment behaves on every engine we track.
Sentiment across all engines →