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

# Prompts

> Find the prompts where you are missing, then open one to read every engine's answer, the pages it cited and the searches it ran.

> [!NOTE]
>
> **Plans:** All plans · **In the app:** Sidebar → Analytics → Prompts

The Prompts page answers: **for which questions do engines include me, and what do they say instead when they do not?** It is where a metric turns into an answer you can read.

The searches engines run behind your prompts have their own tab: see [Query fan-outs](https://docs.usequotable.com/analyze/query-fan-outs.md). Older links to the fan-outs view of this page open that tab.

## Tracked prompts

One row per prompt, 25 per page. Two columns are always shown:

* **Visibility**: the share of the selected engines whose latest answer to this prompt names you.
* **Prompt**: the text, with its location.

Add more with **Columns**:

| Column | What it shows |
| - | - |
| **Location** | Where the prompt runs. Change it here |
| **Type** | Who the prompt names (Branded, Competitor, Organic) and what the asker wants (Commercial, Informational). See [Prompts setup](https://docs.usequotable.com/setup/prompts.md) |
| **Sentiment** | 0 to 100, averaged over answers that include you. Above 70 is positive, below 40 negative. See [Sentiment](https://docs.usequotable.com/metrics/sentiment.md) |
| **Brands** | The tracked competitors that appear in the prompt's latest answers |
| **Shopping** | How often ChatGPT showed a product carousel for the prompt in the last 30 days. See [Shopping](https://docs.usequotable.com/analyze/shopping.md) |
| **Comparison** | How often ChatGPT answered the prompt with a comparison table in the last 30 days |

Use the **Models** filter to see one engine at a time, and **Active** or **Inactive** to switch between running and paused prompts. The **Type** and **Intent** cards at the top show the mix of your prompt set.

> [!TIP]
>
> Sort by **Visibility** and start at the bottom. A commercial, organic prompt where you have no visibility and competitors do is the highest-value gap you have.

## A prompt's page

Open a prompt to see everything about it. The header shows its location, whether it is **Active** or **Paused**, and its **Last run** ("Not run yet" until the first run, which is queued as soon as you add the prompt).

Five figures follow: **Visibility**, **Citation share**, **Brand rank** (your place among the brands its answers name), **Net sentiment** and **Avg. position**, each against the previous period where there is one. Below them, a trend chart and a ranking of the brands the answers name.

| Tab | What it shows |
| - | - |
| **Responses** | Each engine's answer. Open one to read it in full, with its citations and the pages it read but did not cite |
| **Citations** | The pages the answers cited |
| **Searches** | The searches the engines ran to answer it. See [Query fan-outs](https://docs.usequotable.com/analyze/query-fan-outs.md) |

**Avg. position** here is **source position**: where your site ranks among the sources of the answers that cited you. See [Position](https://docs.usequotable.com/metrics/position.md). Share of voice is not shown for a single prompt, because one prompt is too small a sample to split.

Naming you and linking you are separate facts: an answer that links your page without writing your name has cited you, not recommended you.

## What to do with it

1. Find organic, commercial prompts with low visibility.
2. Open the prompt and read the responses. Note which brands are recommended and which sources are cited.
3. Check the **Citations** tab. If a cited page is a list or comparison you are not in, that is your outreach target. If it is your own page, it is not doing enough to name you.
4. Turn the gap into work with [Opportunities](https://docs.usequotable.com/actions/opportunities.md) or [Quotable Agent](https://docs.usequotable.com/actions/quotable-agent.md).

## Next steps

> [card] [Write better prompts](https://docs.usequotable.com/setup/prompts.md)
>
> Build a prompt set that measures what matters.

> [card] [Citations](https://docs.usequotable.com/analyze/citations.md)
>
> The sources behind the answers.


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.