> ## Documentation Index
> Fetch the complete documentation index at: https://docs.trymeridian.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Analyze prompts

> See platform breakdown, brand visibility trends, and response-level results.

## What this is

A deep-dive report for one or multiple prompts, showing how you performed across platforms and what responses contained.

## Why it matters

* You can’t improve rank without seeing **position** and **who beat you**.

* Platform breakdown shows where you’re strong/weak.

* Response tables show the exact responses behind the metrics.

## Where to find it

**Analytics → Prompt Analytics → Prompts → Analyze**

## How it works

* Meridian aggregates results for selected prompts.

* It shows:

  * **Platform Breakdown** (visibility by platform)

  * **Brand Visibility** trend lines over time

  * Response table with **Mentioned**, **Position**, **All mentions**, **Platform**, **Region**

<img src="https://mintlify.s3.us-west-1.amazonaws.com/meridian-25fd9b88/analytics/prompt-analytics/SCREENSHOT_ANALYZE_VIEW" alt="Analyze view with platform breakdown, brand visibility, and response table" />

Caption: Analyze reveals platform-level performance and the responses behind the score.

## How to use it

1. From **Prompts**, click **Analyze** on a prompt (or select multiple prompts).

2. Review **Platform Breakdown** to see which platforms are performing best/worst.

3. Review **Brand Visibility** trend to spot changes over time.

4. Scroll to the response table:

   * Find rows where you are **Not mentioned**

   * Find rows where you’re mentioned but low **Position**

5. Click into a row / response to open **Mention details** (modal).

## How to interpret results

* If Gemini is high but ChatGPT is low → optimize toward ChatGPT’s citations and formats (often different sources).

* If the trend drops on a specific date → look for citation shifts or competitor changes around that time.

* If you’re “Mentioned = Yes” but Position is #4+ → you’re included but not preferred → improve credibility + comparisons.

* If “All mentions” shows many competitors and you’re absent → you’re out of the consideration set → build targeted content + off-page mentions.

* If results differ by region → localize content (regulatory, pricing, availability) and add region-specific proof.

## Common questions / troubleshooting

* “Why does it say ‘Analyzing 1 prompt’?” → you might have only one selected prompt; select more to compare.

* “Why do I see ‘No mentions’?” → the model didn’t mention you in that run; check citations and competitor list.

## Related docs

* [Mention details](/analytics/prompt-analytics/mention-details)

* [Citations](/analytics/prompt-analytics/citations)

* [Competitors](/analytics/prompt-analytics/competitors)

* [Opportunities](/action/opportunities)