> 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/models/rfm-model.md).

# RFM model

Segmentation of one-off buyers and donors by recency, frequency and monetary value

## What it predicts

Not a probability but a **segment**: every customer with orders in the RFM period is placed in one of eight segments (Champion, Loyalist, Potential Loyalist, New, Need Attention, At Risk, Hibernating, Lost) based on how recently, how often and how much they bought. The segment is the churn signal for businesses without a contract to cancel.

## Data it needs

Orders (or one-off donations) with a customer id, date and value. Nothing else is required, which is why RFM is available from day one.

## How it works

Within the configured **RFM period** (typically 12 to 24 months, set per organisation) every customer is scored 1 to 5 on recency, frequency and monetary value relative to the rest of the base. The three digits form an RFM code and a fixed mapping turns the code into a segment. The full code table is on [RFM model and one-off metrics](/churned-docs/metrics-and-definitions/metrics/rfm-model-and-one-off-metrics.md#rfm-segments).

## How often it runs

Daily. The previous segment of every customer is stored, so movements such as Loyalist to Need Attention are visible and can trigger workflows.

## How to read the output

The home dashboard shows each segment's share of orders, value and customers. Movement between segments matters more than the segment itself: a customer sliding from Loyalist to At Risk is the moment to act. The KPIs computed alongside (retention rate, AOV, AOF, expected ACV, cohorts) are defined on the same reference page.

## Where it surfaces

The One-offs tab of subscription dashboards, the E-com Dashboard, the Non-structural tab for NGOs, the Segments and Customers pages, and audience conditions.