# Sentiment

The Sentiment tab provides deep analysis of how AI search platforms talk about your brand. Rather than just tracking whether you're mentioned, Sentiment reveals *what* AI engines say about you and whether it's positive, negative, or neutral — broken down by automatically extracted themes.

## Core Metric

### Sentiment Score

Positive mentions plus half of neutral mentions, divided by total brand mentions (0-100). Higher values indicate more favorable sentiment in how AI platforms portray your brand.

Sentiment scores are color-coded throughout the tab:

| Score Range | Label    | Color |
| ----------- | -------- | ----- |
| 60-100      | Positive | Green |
| 40-59       | Neutral  | Gray  |
| 0-39        | Negative | Red   |

## How Themes Work

AirOps automatically extracts recurring themes from AI-generated answers that mention your brand. Themes are short descriptors (2-3 words) representing topics that AI engines consistently associate with your brand — such as "customer support", "pricing flexibility", or "product reliability".

Themes are extracted using a multi-stage AI analysis of your answer data and are refined over time as new data comes in.

## Sentiment Score Time Series

Track how your sentiment score trends over time with four views:

### Overall

View your aggregate sentiment score trend across all platforms, topics, and themes. Track whether sentiment is improving, stable, or declining.

### By Topic

Break down sentiment score by topic cluster. Identify which topics drive the most positive or negative sentiment and track shifts over time.

### By Platform

Compare sentiment scores across AI answer engines (ChatGPT, Gemini, Perplexity, Google AI Mode, Google AI Overviews). Understand which platforms portray your brand most favorably.

### By Theme

View sentiment score trends for individual themes over time. Track whether specific aspects of your brand perception are improving or declining.

## Sentiment Treemap

The treemap provides a visual overview of all your sentiment themes at a glance:

* **Tile size** represents how frequently the theme appears (occurrences)
* **Tile color** ranges from pink (negative sentiment) to teal (positive sentiment)

Hover over any tile to see the theme name, sentiment score, and a breakdown of positive, negative, and neutral counts.

Use the sentiment filter to focus on **All Sentiments**, **Positive**, or **Negative** themes.

## Sentiment Themes Table

The themes table provides a detailed breakdown of each extracted theme with the following columns:

| Column          | Description                                                         |
| --------------- | ------------------------------------------------------------------- |
| Theme           | The name of the extracted theme                                     |
| Sentiment Score | The theme's sentiment score (0-100), color-coded by sentiment level |
| Volume          | Percentage of total theme occurrences this theme represents         |
| Occurrences     | Total number of times this theme appears across answers             |

### Filtering by Sentiment

Use the sentiment dropdown to filter themes by **All Sentiment**, **Positive** (score >= 60), or **Negative** (score < 40).

### Drill-Down to Answers

Expand any theme row to see the individual AI-generated answers that contributed to that theme's sentiment. Each answer row shows:

* **Provider** — which AI platform generated the answer
* **Date** — when the answer was captured
* **Answer text** — a snippet of the AI-generated response

Click on any answer to open a detail panel with the full answer content.

## Filters

All charts and metrics on this page can be filtered by:

* **Date Range**: Select a time period for your analysis
* **Region**: Filter by geographic region
* **Prompt Type**: Category-related (general prompts) vs Brand-related (prompts mentioning your brand directly)
* **Topic**: Filter by topic clusters
* **Platform**: Filter by specific AI answer engine
* **Persona**: Filter by audience persona


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://docs.airops.com/insights/insights/sentiment.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.