> For the complete documentation index, see [llms.txt](https://docs.airops.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.airops.com/insights/prompts/your-prompts.md).

# Your prompts

Manage tracked prompts and analyze prompt-level AI search visibility.

The **Prompts** tab contains the prompts AirOps tracks for your brand. Use it to organize prompts by topic or tag, compare performance, and inspect the answers returned by each AI platform.

Each prompt runs daily across your connected AI search platforms.

## View tracked prompts

The prompts table contains the following columns:

| Column             | Description                                                                                                                                                                                                                        |
| ------------------ | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| **Prompt**         | The question or prompt tracked across AI platforms                                                                                                                                                                                 |
| **Topic**          | The topic category assigned to the prompt, such as Content Refresh                                                                                                                                                                 |
| **Tags**           | Custom tags used to organize and filter prompts                                                                                                                                                                                    |
| **Intent**         | The classified stage: Awareness, Consideration, Evaluation, or Purchase                                                                                                                                                            |
| **Prompt Type**    | Category Related or Brand Related, based on whether the prompt names your brand                                                                                                                                                    |
| **Query Fanouts**  | Observed search queries associated with the prompt. Hover over the count to preview queries; the platform filter applies. See [Query Fanouts](https://docs.airops.com/insights/prompts/query-fanouts.md) for the query-level view. |
| **Prompt Volume**  | The estimated monthly volume for the prompt, shown as a number. See [Prompt volume](#prompt-volume) for how it's calculated.                                                                                                       |
| **Mention Rate**   | The percentage of AI responses that mention your brand, including the change over time                                                                                                                                             |
| **Citation Rate**  | The percentage of AI responses that cite your content as a source, including the change over time                                                                                                                                  |
| **Share of Voice** | Your mentions divided by mentions of your brand or tracked competitors, shown as a percentage with its change over time                                                                                                            |

<figure><img src="https://3762890407-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FX2n5yPRPynbnWuO4SH0M%2Fuploads%2Fgit-blob-125e142e75ba9c796a560154926db9f54a64e746%2Fprompts-01-table.png?alt=media" alt="Prompts table listing tracked SpaceOps prompts with topic, tags, query fanouts, prompt type, intent, volume, mention rate, citation rate, and share of voice"><figcaption><p>Each tracked prompt runs daily across your AI search platforms.</p></figcaption></figure>

### Filter and organize prompts

Use the controls above the table to change the data shown:

* **Date Range:** Select the period to analyze.
* **Region:** Filter answers by geographic region.
* **Filter:** Narrow the table by **Prompt Intent** or additional criteria. You can also filter by **Share of Voice** and its trend.
* **Topics:** Open the [Topics tab](https://docs.airops.com/insights/prompts/topics.md) for category-level performance and coverage.
* **Search:** Find prompts by keyword.

### Manage table columns

Open **Manage columns** in the table toolbar to show, hide, or reorder configurable columns. Select **Save changes** to apply the layout. Closing the dialog without saving discards your edits. Required columns stay visible and in their fixed positions.

Preferences apply to the current Brand Kit and table, including grouped and ungrouped views. They stay in this browser and reset if you clear browser storage.

Drag a resizable column's header edge to widen it. Expanding a column beyond the available table width adds horizontal scrolling.

## How to configure tracked prompts

Open **Insights → Settings → Configuration** to add prompts and manage assignments. The Prompts tab remains the place to analyze performance and export results.

### Upload prompts with a CSV file

In Configuration, open **Add Prompts** and choose the CSV import option. You can drag a CSV file into the Upload Prompts modal or click to browse.

* Upload CSV files up to **4 MB**.
* Click **Download template** to get the required format.
* Include a topic for each prompt. AirOps creates missing topics during the upload.

Topics group related prompts for filtering and analysis.

### Edit topics

In Configuration, click **Edit topics** to manage topic categories. From the modal, you can:

* View topics and their color indicators.
* Edit a topic's name and description. Clear the description to remove it.
* Delete a topic only when it has no prompts. Editing topics can affect measurement and performance.
* Click **+ Add Topic** to create a topic.

### Configure mention extraction

AirOps detects whether each answer mentions your brand and competitors. Configure custom regex mention extraction in Prompt settings when a name causes false positives or requires precise matching.

Use regex mention extraction when:

* Your brand or competitor name is also a common word.
* You need to match specific spellings, products, or branded phrases.
* LLM-based extraction counts mentions that do not refer to the tracked company.

Generate a regex pattern with AI, then test it against sample answer text before saving. Testing confirms that the pattern matches the mentions you expect and ignores unrelated text.

### Manage tags

Tags organize prompts with groupings beyond topics. You can use them to group prompts by campaign, season, business unit, or product line.

Tags are optional. Add the tags your team needs to filter and report on tracked prompts.

### Apply bulk actions

Manage assignments in **Settings → Configuration**. Select prompts, open **Reassign**, and choose regions, platforms, or personas. Review the change badges, then choose **Save changes** to apply the draft or **Discard** to abandon it.

Use the available topic and tag controls in Configuration to organize the selected prompts. Changing the tracked set can change the metrics you see for a topic.

## Export prompts and answers

On **Insights → Prompts → Prompts**, open **... → Export CSV** and choose an export:

* **Export Prompts** downloads a CSV with one row per matching prompt, up to **10,000 rows**. It follows the page's filters, search, and sort; the date range scopes calculated metrics. Columns: Prompt, Topic, Tags, Query Fanouts, Prompt Type, Prompt Intent, Prompt Volume, Mention Rate, Positive Answer Rate, and Citation Rate.
* **Export Answers** emails a CSV with one row per answer. It includes date, platform, prompt, topic, tags, persona, region, full answer text, Brand Kit name, mentioned brands, whether your brand was mentioned, and citations. It uses selected dates and platforms, but ignores the table's other filters.

For smaller answer exports, select a shorter date range.

## Analyze a prompt

Click a prompt to open its detailed analysis.

### Prompt visibility chart

The time-series chart shows your brand's visibility percentage for the selected date range. Use it to identify changes in visibility over time.

### Leaderboard

The leaderboard compares your brand with competitors for the selected prompt:

| Column         | Description                                            |
| -------------- | ------------------------------------------------------ |
| **Topic**      | The brand or competitor name                           |
| **Mentions**   | The number of tracked responses that mention the brand |
| **Visibility** | The percentage of responses in which the brand appears |

### Mention rate by platform

Compare your mention rate across ChatGPT, Gemini, Perplexity, and Google AI Mode. This breakdown shows where your brand appears and where it has visibility gaps.

### Answer history

The Answer History table contains every tracked response for the prompt:

| Column             | Description                                         |
| ------------------ | --------------------------------------------------- |
| **Date**           | When AirOps captured the response                   |
| **Persona**        | The persona used to run the prompt, such as Default |
| **Platform**       | The AI platform that generated the response         |
| **Answer Preview** | A preview of the generated answer                   |
| **Mentioned**      | Whether the response mentions your brand            |
| **Competitors**    | The competitors mentioned in the response           |

Click an answer row to open the full response.

<figure><img src="https://3762890407-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FX2n5yPRPynbnWuO4SH0M%2Fuploads%2Fgit-blob-229efee53fa97c62b84e4ba328fe3f1922a3f61f%2Fprompts-02-prompt-analysis.png?alt=media" alt="Prompt detail for best satellite fleet monitoring software with a Mention Rate chart, Mention Rate by Competitor, and Mention Rate by Platform"><figcaption><p>Open a prompt to compare visibility by competitor and platform.</p></figcaption></figure>

## Review an answer

The answer detail view shows the response from an AI platform and how it references brands and sources.

### Compare platforms

Use the platform tabs to compare responses from ChatGPT, Google AI Mode, Google AI Overview, Gemini, and Perplexity.

### Check brand status

Brand status shows whether the response mentions your brand and cites your content.

### Review mentions and citations

The Mentions section lists the brands and competitors named in the response. The Citations section lists the sources the platform referenced and links to the cited content.

Use these sections to determine:

* Whether an AI platform cites your content.
* Which competitor or third-party sources influence its answer.
* How the answer positions your brand relative to competitors.

### Review ChatGPT shopping results

When ChatGPT includes product listings in a tracked answer, the answer detail shows those shopping results below the response. Use them to inspect the products and merchants shown for the prompt.

<figure><img src="https://3762890407-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FX2n5yPRPynbnWuO4SH0M%2Fuploads%2Fgit-blob-980231e466cf5d76c1fc975fa656e60be278889d%2Fprompts-04-shopping-results.png?alt=media" alt="ChatGPT answer detail for the prompt best cubesat kits for university teams, with the answer text above a Shopping results section listing four products with ratings and merchant offers"><figcaption><p>Shopping results show the products and merchant offers ChatGPT returned with the answer.</p></figcaption></figure>

In the Prompts table, **Shopping results load** measures the percentage of collected answers from the selected product-collecting platforms that include at least one product listing. Filter to **ChatGPT** to analyze its shopping results. In Answer History, **Shopping results** shows the number of products in an answer. A dash means product data was not collected; it is different from a collected answer with zero products.

<figure><img src="https://3762890407-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FX2n5yPRPynbnWuO4SH0M%2Fuploads%2Fgit-blob-8b747dbd384f9787986c58f02cdd43d8c0d65deb%2Fprompts-05-shopping-results-load.png?alt=media" alt="Prompts table filtered to ChatGPT and sorted by Shopping results load, with three Hardware &#x26; Kits prompts at 80.0%, 73.3%, and 60.0% and the other prompts at 0.0%"><figcaption><p>Filter to ChatGPT to see how often answers to each prompt include shopping results.</p></figcaption></figure>

### Read the full answer

The Full Answer section contains the complete response. Read the surrounding context to understand how the platform describes your brand and where your content could improve.

<figure><img src="https://3762890407-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FX2n5yPRPynbnWuO4SH0M%2Fuploads%2Fgit-blob-8148348699d421846bf2a479136ae478e6f5e9e5%2Fprompts-03-answer-detail.png?alt=media" alt="Answer detail for a ChatGPT response that mentions but does not cite SpaceOps, with the mentioned brands, cited sources, and full answer"><figcaption><p>Check whether an answer mentions and cites your brand.</p></figcaption></figure>

## Understand prompt measurements

### Prompt generation

AirOps generates tracked prompts using:

* Traditional SEO keywords from your keyword strategy.
* Panel data that reflects AI search behavior and trends.
* Data science extraction that surfaces questions and themes in your category.

### Prompt volume

Prompt Volume is the estimated monthly volume for a prompt, shown as a number in the **Prompt Volume** column.

AirOps calculates Prompt Volume from a search-demand baseline and three adjustments:

* **Search-demand baseline:** AirOps maps each prompt to related keywords and their search demand, comparing similar keywords semantically so overlapping demand isn't counted multiple times. The result is a deduplicated estimate of the demand behind the prompt's topic.
* **AI platform market:** A scaling factor reflects how much demand has shifted from traditional search to AI assistants like ChatGPT and Perplexity.
* **Audience specificity:** A prompt aimed at a specific persona, industry, company size, or geography represents only part of the broader audience, so AirOps reduces the baseline to match.
* **AI suitability:** Some intents are more likely than others to be handled by AI. These include comparison, recommendation, explanation, writing, and context-heavy questions.

### Regional tracking

Add regions in your Brand Kit's Insight Settings to track prompt visibility from different geographic contexts. Regional tracking helps you compare brand mentions, competitive positioning, and content opportunities by market.

### Persona tracking

Configure personas in your Brand Kit's Insight Settings with a title and description. AirOps runs the same prompt for each persona so you can compare answers by audience segment or funnel stage.

Each persona adds another run for every tracked prompt. For example, tracking 200 prompts with three personas produces 600 answers.


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

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

Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters:

```
GET https://docs.airops.com/insights/prompts/your-prompts.md?ask=<question>&goal=<user_goal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case.

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.
