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

# Analytics Metrics Glossary

> Plain definitions of tinyanalytics traffic, acquisition, conversion, AI, and revenue metrics so you know exactly what each number counts across reports.

This glossary defines the metrics you see most often in tinyanalytics. The same definition follows the metric across reports unless a page states a narrower scope.

## Core metrics

| Metric                       | Definition                                                                                                                                                                                                                                                                                                                              |
| ---------------------------- | --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| **Sessions**                 | The number of distinct sessions in the date range. A session is one visitor's activity within a 30-minute sliding window of inactivity.                                                                                                                                                                                                 |
| **Users**                    | The number of distinct visitors. A visitor is counted by the ID you supplied with [identify](/docs/user-identification-analytics), when present, otherwise by their [cookieless ID](/docs/resources/cookieless-analytics-identity). One signed-in person on a phone and a laptop counts as one user when both devices use the same identified ID. |
| **Pageviews**                | The total number of pages viewed — every page load and, in a single-page app, every route change.                                                                                                                                                                                                                                       |
| **Events**                   | All accepted events in the range, including pageviews, custom events, autocaptured interactions, engagement, errors, and performance events. A report can narrow this to one event type.                                                                                                                                                |
| **Pages per session**        | The mean number of pageviews per session.                                                                                                                                                                                                                                                                                               |
| **Events per session**       | The number of accepted events divided by sessions.                                                                                                                                                                                                                                                                                      |
| **Bounce rate**              | The percentage of sessions with exactly one pageview.                                                                                                                                                                                                                                                                                   |
| **Session duration**         | The average time between a session's first and last event, in seconds.                                                                                                                                                                                                                                                                  |
| **AI traffic (AI sessions)** | The number of sessions that *entered* from an [AI assistant](/docs/ai-referral-traffic-analytics) — a visitor who clicked through from ChatGPT, Perplexity, Gemini, Claude, Copilot, and similar. Each session is counted once, by the assistant it arrived from.                                                                            |

## Acquisition metrics

| Metric                         | Definition                                                                                                                                                 |
| ------------------------------ | ---------------------------------------------------------------------------------------------------------------------------------------------------------- |
| **New users**                  | Users whose first-ever event falls inside the selected date range.                                                                                         |
| **Returning users**            | Users active in the selected date range whose first-ever event happened before the range started.                                                          |
| **New-user share**             | New users divided by all users in the row or report, expressed as a percentage.                                                                            |
| **First-touch channel/source** | The acquisition channel or source from a user's first-ever session. Later sessions do not rewrite it. See [User acquisition](/docs/user-acquisition-analytics). |
| **Session channel/source**     | The entry acquisition context of the session being measured. The same user can have sessions attributed to different sources over time.                    |

## Conversion and revenue metrics

| Metric                   | Definition                                                                                              |
| ------------------------ | ------------------------------------------------------------------------------------------------------- |
| **Goal conversions**     | Sessions that matched a goal's page, custom-event, or autocapture rule inside the selected range.       |
| **Goal conversion rate** | Converting sessions divided by all sessions in the same selected date range and filter scope.           |
| **Total revenue**        | Gross positive revenue in the selected range, converted to the site's reporting currency.               |
| **Net revenue**          | Total revenue minus Stripe refunds and lost disputes. Client-event revenue has no negative adjustments. |
| **Refunds**              | The amount subtracted by Stripe refunds and lost disputes.                                              |
| **Average order value**  | Total revenue divided by the number of orders.                                                          |
| **Paying sessions**      | Distinct sessions that produced at least one revenue event.                                             |
| **Orders**               | Completed revenue-bearing purchases reported by the selected revenue source.                            |

## What counts as a session?

A **session** is one visitor's activity within a **30-minute sliding window of inactivity**. Every event the visitor sends pushes the window forward; the session ends only after 30 minutes of silence, and their next visit starts a new one. Because the window slides, a visitor who's active across the half-hour boundary stays in a single session rather than being split in two.

## How is a bounce counted?

A **bounce** is a session with exactly one pageview — the visitor saw one page and didn't view another. Bounce rate is the percentage of sessions that bounced. Custom events don't change the count: a single-pageview session is a bounce even if the visitor clicked a button or submitted a form on that page.

## Why can user counts change after identify?

[`identify()`](/docs/user-identification-analytics) can merge recent anonymous activity into the stable ID you provide. That can turn two anonymous cookieless IDs into one identified user and improve cross-device totals. Historical pageviews and events remain in place; only the effective user used by people-level reports changes.

## Related

<Columns cols={2}>
  <Card title="Data dictionary" icon="book" href="/docs/analytics-data-dictionary">
    The events, properties, and traits catalog.
  </Card>

  <Card title="Dashboard overview" icon="gauge" href="/docs/web-analytics-dashboard">
    See these metrics for your site.
  </Card>

  <Card title="How cookieless identity works" icon="fingerprint" href="/docs/resources/cookieless-analytics-identity">
    How a visitor becomes a "user."
  </Card>

  <Card title="How traffic is classified" icon="signs-post" href="/docs/resources/analytics-channel-attribution">
    How referrers become channels.
  </Card>

  <Card title="User acquisition" icon="user-plus" href="/docs/user-acquisition-analytics">
    Apply the new and returning definitions by first-touch source.
  </Card>

  <Card title="Revenue analytics" icon="dollar-sign" href="/docs/revenue-analytics">
    Understand the source and currency rules behind revenue metrics.
  </Card>
</Columns>


## Related topics

- [Analytics Alerts](/docs/analytics-alerts.md)
- [Organization Analytics Overview](/docs/organization-analytics-overview.md)
- [Core Web Vitals Analytics](/docs/core-web-vitals-analytics.md)
