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

# Retention

> Measure how often users return after their first action with day, week, and month cohorts in Klaritics. Pin retention widgets to dashboards and set alerts.

Retention in Klaritics shows how many users return after performing an initial event. You can group users into cohorts by the day, week, or month they first appeared, then track how each cohort engages over time. This helps you understand whether your product is sticky and if recent changes are improving long-term engagement.

## When to use Retention

Use Retention when you want to:

* Measure product stickiness after a key action
* Compare retention across weekly or monthly cohorts
* Evaluate whether a launch improved long-term user behavior

## Build a retention analysis

<Steps>
  <Step title="Pick the starting event">
    Choose the event that defines when a user enters a cohort, such as sign-up or first purchase.
  </Step>

  <Step title="Pick the returning event">
    Choose the event that counts as a return visit. This can be the same as the starting event or a different one, such as any active session.
  </Step>

  <Step title="Select the cohort granularity">
    Group users by day, week, or month. Weekly and monthly views are useful for smoothing out daily noise.
  </Step>

  <Step title="Compare cohorts">
    Review the retention table to see how each cohort performs over time. Look for trends that show improving or declining return rates.
  </Step>
</Steps>

## Save and share

Save the retention analysis and pin it to a dashboard as a widget. Set alerts on the widget to be notified when retention for a recent cohort drops below an expected level.

<Tip>
  Pair retention with [Cohorts](/analysis/cohorts) to define custom user groups and compare their return behavior.
</Tip>

## Next steps

* [Cohorts](/analysis/cohorts) to build custom user groups
* [Engagement Matrix](/analysis/engagement-matrix) to segment users by depth of engagement