> 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/rfm-model-and-one-off-metrics.md). # RFM model and one-off metrics Segmentation and KPIs for customers without a subscription, such as one-off buyers, users and donors Customers without a contract cannot cancel, so churn has to be inferred from their behaviour. Churned does this with an **RFM** model over a configurable **RFM period**. The same engine is used for one-off purchases (the *One-offs* tab of a subscription dashboard and the *E-com Dashboard*), for one-off donations (the *Non-structural* tab of NGO dashboards) and, as **RFV**, for product usage of SaaS users. {% hint style="info" %} The **RFM period** is configured per organisation to match the buying rhythm of your customers, typically 12 to 24 months. All metrics on this page look at the window between today and *RFM period* months ago. *Vs last month* compares that window with the window ending 30 days ago. {% endhint %} ## RFM and RFV RFM stands for **Recency**, **Frequency** and **Monetary value**: * **Recency**: how recently the customer last ordered, based on the latest order date. * **Frequency**: how many orders the customer placed within the RFM period. * **Monetary value**: the total amount the customer spent within the RFM period. For SaaS product usage the third dimension is **Volume** instead of Monetary value (RFV): the depth of usage, such as the number of features used, sessions or transactions. Everything else works identically. For each dimension every customer receives a score from 1 (weakest) to 5 (strongest), relative to the other customers in your base. Together they form a three-digit **RFM code** (recency, frequency, monetary), and each code maps to one of eight **segments**. Segments are recalculated daily and the previous segment of every customer is stored, so movements between segments are visible.

Intuitive meaning of the RFM segments

## RFM segments The segment names below are the labels used on the dashboard. | Segment | RFM codes | Criteria | Meaning | | ---------------------- | ------------------------------------------------------------------------------------------------------------------------------------- | ----------------------------------------------- | -------------------------------------------------------------- | | **Champion** | 555, 554, 545, 544, 455, 454, 445 | Highest recency, frequency and order value | Your best customers. Reward and recognise them. | | **Loyalist** | 553, 552, 551, 543, 542, 541, 444, 435, 355, 354, 345, 344 | High recency and high frequency | Consistently engaged. Keep delivering value. | | **Potential Loyalist** | 535, 534, 533, 532, 531, 525, 524, 523, 522, 521, 453, 452, 451, 442, 441, 433, 432, 431, 425, 424, 423, 422, 421, 353, 352, 351 | High recency, medium frequency | Almost loyal customers that should improve frequency. | | **New** | 515, 514, 513, 512, 511, 415, 414, 413, 412, 411, 313 | High recency, low frequency | Only one recent order at any value. Focus on the second order. | | **Need Attention** | 443, 434, 343, 342, 341, 335, 334, 333, 325, 324, 323, 315, 314 | Medium recency and frequency, often high value | Good customers that are slipping; re-engage before they lapse. | | **At Risk** | 255, 254, 253, 252, 245, 244, 243, 242, 235, 234, 225, 224, 215, 214, 155, 154, 153, 152, 145, 144, 143, 142, 135, 134, 133, 125, 124 | Low recency, previously high frequency or value | Formerly good customers not seen for a long time. Win-back. | | **Hibernating** | 332, 331, 322, 321, 312, 311, 251, 241, 233, 232, 231, 223, 222, 221, 213, 212, 211, 132, 123, 122 | Low recency, frequency and value | Sporadic customers not seen in a while. | | **Lost** | 151, 141, 131, 121, 115, 114, 113, 112, 111 | Very low recency and frequency | Not expected to return without intervention. | On the home dashboard each segment shows its share of total orders, total value and total customers; clicking a segment opens the customer page filtered on it. Segment playbooks per industry are under [RFV-Model Segment - Best Practices](/churned-docs/metrics-and-definitions/rfv-model-segment-best-practices.md). *** ## Customers *Customers that made a purchase in the last RFM period.*\ *Active customer base.* A customer is active whenever they have at least one order within the RFM period. On NGO dashboards this tile is called *Donors*. ## Retention rate *Percentage of returning customers.*\ *How loyal are your customers.* $$ Retention \ Rate = 1 - \frac{\textnormal{Churned}}{\textnormal{Active}} $$ The percentage of active customers that ordered more than once within the RFM period. A customer counts as churned when their only order in the RFM period is about to fall out of the window without a follow-up order. New customers, whose first order is too recent to expect a second one, are not taken into account. ## Average Order Value (AOV) *The average amount of money spent per order.*\ *The willingness to pay of each customer.* Total order value divided by the number of orders (n) in the RFM period. On the customer page AOV is also shown per customer; for a customer with one order it equals that order's value. On NGO dashboards this is the **ADV**, average donation value. $$ AOV = \frac{\sum\_{i=1}^{n} \textnormal{Order value}\_i}{n} $$ ## Average Order Frequency (AOF) *The average number of orders per customer over the RFM period.*\ *How frequently are your customers buying.* Number of orders (n) divided by the number of active customers (c) in the RFM period. On NGO dashboards this is the **ADF**, average donation frequency. $$ AOF = \frac{n}{c} $$ ## Expected Average Customer Value (Expected ACV) *The expected revenue value of an active customer over an RFM period.*\ *What are your customers worth.* A weighted sum of the current average customer value (AOV times AOF) and the previous expected value, so that a single strong or weak month does not swing the number. On NGO dashboards this is the **Expected ADV**, expected average donor value. $$ Expected \ ACV\_t = w\_1 \times AOV\_t \times AOF\_t + w\_2 \times Expected \ ACV\_{t-1} $$ ## Cohort analysis (one-off) *The percentage of returning customers per cohort.*\ *How loyal is your current customer base.* A cohort groups the **unique new customers** whose first order fell in a specific year-month. The *Cohort Analysis* page has two tabs: * **Per Order**: the columns show which percentage of the cohort came back for a 2nd, 3rd, 4th … order, together with the time since the previous order. This measures how well you convert a first order into a second. * **Per Month**: the columns show which percentage of the cohort ordered again in each following month. This measures how quickly customers return. In the example below 3,534 new customers ordered in December 2020. Of those, 80% returned for a third order and 65% placed 5 orders or more.

Example of a cohort plot with the number of orders in the columns