Model breakdown
Model breakdown compares the answers captured from each model in your segment. Use it to see whether a project-wide result is shared across models or concentrated in one.
Read the breakdown
The analytics detail view shows each model's name, score, and mention rate for a brand. The per-model score uses the percentage of captured responses that mention the brand. The overall visibility score also includes position weighting, so the two scores measure different things.
Use the model names shown in your project. Availability depends on your plan and the segment's selection.
Investigate a difference
- Select the same date range and segment for the comparison.
- Open Answers -> Prompts and compare the same question across models.
- Check the capture times and whether each model returned responses.
- Read the answer text and cited sources to understand the difference.
A stronger result on one model does not tell you what was in its training data. Differences may come from response variation, model behaviour, available grounding, or the questions sampled.
Track changes
If one model drops while others stay stable, inspect its recent answers before changing your content strategy. A model update is one possible cause, but the chart alone cannot establish it.
Prioritise models relevant to your audience and questions where your brand is a suitable answer. Do not assume that a content change will improve every model together.