> For the complete documentation index, see [llms.txt](https://docs.churned.io/churned-docs/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.churned.io/churned-docs/metrics-and-definitions/metrics/segment-analysis-and-cohorts.md).

# Segment analysis and cohorts

Metrics shown when you compare segments on the technical dashboard, plus cohort plots

On the [Segments page](/churned-docs/metrics-and-definitions/segments-page.md) you create custom segments; on the [Technical dashboard](/churned-docs/metrics-and-definitions/technical-dashboard.md) you compare them. All metrics below are calculated daily over the **active** customers in the selected segment.

## Engaged customers

*The total number of engaged customers of the active customers in the selected segment.*\
*<mark style="color:blue;">How engaged this segment is compared to the rest of the active customer base.</mark>*

Customers in the segment with an [engagement score](/churned-docs/metrics-and-definitions/metrics/health-risk-and-engagement-scores.md#engagement-score) of 7 or higher.

## Non-engaged customers

*The total number of non-engaged customers of the active customers in the selected segment.*\
*<mark style="color:blue;">How non-engaged this segment is compared to the rest of the active customer base.</mark>*

Customers in the segment with an [engagement score](/churned-docs/metrics-and-definitions/metrics/health-risk-and-engagement-scores.md#engagement-score) of 4 or lower.

## High-value customers

*The total number of valuable customers of the active customers in the selected segment.*\
*<mark style="color:blue;">How valuable this segment is compared to the rest of the active customer base.</mark>*

The number of customers within the segment with an ARR above the median ARR of the whole active customer base.

## Low-value customers

*The total number of less valuable customers of the active customers in the selected segment.*\
*<mark style="color:blue;">How valuable this segment is compared to the rest of the active customer base.</mark>*

The number of customers within the segment with an ARR below the median ARR of the whole active customer base.

## Top three risk drivers per segment

*What are the biggest drivers of churn risk in this segment.*\
*<mark style="color:blue;">The top three drivers of churn risk for the selected segment.</mark>*

Calculated by counting how often each [risk driver](/churned-docs/metrics-and-definitions/metrics/health-risk-and-engagement-scores.md#top-3-risk-drivers) occurs in the top drivers of the active customers within the segment. Drivers are split into those increasing and those decreasing risk.

## Per risk level

*How (un)healthy the selected segment really is.*\
*<mark style="color:blue;">The distribution of the active customers in the selected segment over the three risk levels.</mark>*

The share of the segment's active customers in each [risk level](/churned-docs/metrics-and-definitions/metrics/health-risk-and-engagement-scores.md#risk-levels). Clicking a bar opens the list of customers in that level.

{% hint style="info" %}
**Example**

Consider the situation where you have selected 2 segments:

1. Customers in onboarding
2. Loyal customers

Looking at the distribution in the graph we now see that 4.8% of the active customers in the segment *onboarding* is at High risk while this percentage for loyal customers is only 0.8%.

<img src="https://246234927-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FkSyCQUyE0u7ya9ULaprd%2Fuploads%2FYVgB9ZxS7pEa9pHXx2vu%2FScreenshot%202022-12-12%20at%2012.03.50.png?alt=media&#x26;token=74eba835-a36e-47ed-9644-a252ccbec4c8" alt="" data-size="original">
{% endhint %}

## Trend over time

The technical dashboard also plots churn rate, ARR and other KPIs over time, optionally with one line per segment or sub-segment. Hovering a month shows the value and the top 3 risk drivers of that month.

***

## Cohort analysis (subscriptions)

*The percentage of customers per start month that is still active.*\
*<mark style="color:blue;">How well do you retain each generation of customers.</mark>*

Customers are grouped by the month in which their first subscription started. For every cohort the plot shows the number of new customers and, per month of tenure, the percentage that still has an active subscription. Reading down a column compares cohorts at the same tenure, which reveals whether onboarding changes, pricing changes or recent campaigns retain better or worse than before.

<figure><img src="https://246234927-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FkSyCQUyE0u7ya9ULaprd%2Fuploads%2Fj7KZ9dgGLLaGByusj7cF%2FScreenshot%202025-03-18%20at%2010.47.52.png?alt=media&#x26;token=cf563cb0-7790-4110-99e6-82d72a605862" alt=""><figcaption><p>Example of the subscription cohort plot</p></figcaption></figure>

**Avg. retention**\
*The average retention rate for the complete column (a specific month of tenure).*

**Avg. dropoff**\
*The average percentage of customers that churns moving from month X to month Y of their subscription.*

Cohorts for one-off buyers and donors (per order and per month) are defined on the [RFM page](/churned-docs/metrics-and-definitions/metrics/rfm-model-and-one-off-metrics.md#cohort-analysis-one-off).