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

# Understand your results

> Read AI Visibility, Share of Voice, Mentions, and Citations together.

Start with **AI Visibility** to see how often your Brand appeared across the response opportunities in an Analysis. Use **Share of Voice** to compare your Brand with the Competitors you track. Then inspect the underlying **Mentions** and **Citations** before deciding what to do.

| View | Question it answers | What it counts |
| - | - | - |
| [AI Visibility](/metrics/ai-visibility-score) | How often did an AI response mention our Brand? | Mentioning Responses against planned response opportunities, averaged equally across AI Platforms. |
| [Share of Voice](/metrics/share-of-voice) | What share of competitive mentions was ours? | Your Mentioning Response Count against yours plus tracked Competitors' counts. |
| Mentions | What did the response actually say about us? | Brand references in response text. Expand a Prompt to open a recorded answer. |
| Citations | Which source links did the response use? | Links to sources. Browse by Prompt, Competitor, source, or page to reach the answer. |

The [Technical Audit](/features/geo-audit) is a separate examination of website issues and fixes. Its Audit Runs do not represent another AI Visibility measurement.

## If a result surprises you

| What you see | First check | Next place to look |
| - | - | - |
| AI Visibility increases but Share of Voice decreases | Whether tracked Competitors gained mentions faster than your Brand. | [Share of Voice](/metrics/share-of-voice) and the responses in Mentions. |
| Your Brand appears in a response but your Website has no Citation | Whether the response links to another source or provides no source link. | Citations and the response evidence; a Mention alone does not imply a Citation. |
| A trend changes after you edit Prompts or Competitors | Whether the measurement scope changed between Analyses. | [Prompt Library](/how-to/manage-prompts) and [Analysis History](/how-to/run-your-first-audit). |
| A Technical Audit Finding disappears after a fix | Whether a later Audit Run checked the deployed Website. | [Technical Audit](/features/geo-audit); use a later Analysis to observe AI response changes separately. |

These checks help narrow an investigation. A metric change alone does not establish what caused it.

## A practical review sequence

1. In **Dashboard → Overview**, choose a date range and note which completed Analyses fall within it. Check **Analysis History** if the latest Analysis is still running or failed. Use the AI Visibility Trend to compare your Brand Line with tracked Competitor Lines.
2. Open **Dashboard → Analytics → Share of Voice**. Keep the **Mentions** share basis selected when interpreting Share of Voice; the **Citations** basis, when available, shows Citation Share instead. Select a Brand or Competitor cell in the AI Platform matrix to see the answers behind that cell.
3. Open **Dashboard → Analytics → Mentions**. Find a relevant Prompt under **Mentions by Prompt**, expand its row, then select a Mention to read the stored AI Response. Check which AI Platform produced it, what the response actually said, and which other tracked Brands it named. The **Mentions over time** chart helps locate dates to investigate.
4. Open **Dashboard → Analytics → Citations**. In **Citations Explorer**, use **By prompt** to find the question, **By competitor** to find a rival's cited Prompts, or **By page** to find your cited URL. Open a row to inspect the recorded response and its source links. **Gaps** lists Prompts where a tracked Competitor was cited and your Website was not.
5. Before comparing periods, match the date range and **Models** filter across views. Check whether the active Prompt set, tracked Competitors, or response Coverage changed. A chart point is an observation from an Analysis; a blank, failed, or unmeasured point is not a measured zero.
6. Use Technical Audit findings to prioritize Website fixes, then confirm a deployed fix with a later Audit Run. Watch subsequent Analyses to observe any AI Visibility change without assuming the fix caused it.

The selected date range can combine multiple completed Analyses. A view's summary may aggregate that window, while a chart plots separate Analysis points. Open the response and its Analysis date when you need evidence for one observation.


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