RFM Analysis

Being a part of CleverTap's 'Discovery platform',RFM is available for Enterprise customers.

Overview

CleverTap's RFM Analysis feature helps you analyze the health of your user base, and run engagement campaigns to target specific user segments that need improvement.

RFM Analysis is a user segmentation model that segments your users based on how recently and frequently they performed a specific event. The output of RFM Analysis is a segmentation of your users into ten RFM user types, which range from Champion users who are your best customers to Hibernating users who are likely to churn.

For example, let's say you have an e-commerce app. You can run a RFM Analysis to understand the segmentation of user base based on how recently and frequently they purchased a product. If you discover you have a lot of Hibernating users, you can run a campaign to re-engage these users and encourage them to make a purchase by sending them a discount code.

As part of our RFM Analysis feature, we provide two tools:

  • RFM Grid: Visualization to show the number of users in each RFM segment, their average monetary value, and their reachability on different marketing channels.
  • RFM Transition: Visualization to show the flow of users from one RFM segment to another.

RFM Grid

RFM Grid is a visualization tool available in the CleverTap dashboard. This tool presents the results of the RFM Analysis for your user base in a simple chart highlighting the number of users in each RFM segment, their average monetary value, and their reachability on different marketing channels.

RFM Grid Guide

To access RFM Grid, login to the CleverTap dashboard, and click on the RFM button under the Segment tab.

Select the date range and event for your RFM Analysis. For your first analysis, App Launches over the past 30 days is a good choice. Click the Calculate button to run the analysis.

The RFM Grid will be shown on the bottom of the same page. You can click on any segment to get more information or save it for later analysis. You can also create a campaign directly from the RFM Grid by clicking on the Create message button.

RFM Transitions

RFM Transition is a visualization tool available in the CleverTap dashboard. This tool helps you understand the flow of your users from one RFM segment to another, such as how many New Users became Champions.

For example, let's say you see a lot of your Hibernating users are coming from the New Users segment. Unfortunately, now you know that most of your new users churn, but with this information you can improve the situation by creating a better onboarding experience.

Hibernating vs Inactive

The difference between a Hibernating user and an Inactive user is a Hibernating user has performed the selected event at least once in the defined time range.

RFM Transition Guide

To access RFM Transition, login to the CleverTap dashboard, and then click on the RFM button under the Segment tab.

Select the date range and event for your analysis. For your first analysis, App Launches over the past 30 days is a good starting place. Click the Calculate button to run the analysis.

Click on the RFM Transitions button.

Once the RFM Transitions tool loads, you can click on any segment to understand the flow of users into that segment. Click on the Champions segment to find out the sources of its users.

You will see a visualization on the left and table with data on the right. Both of these show the sources of users into the Champions segment.

From the table on the right, you can click the three vertical dots to save a user segment for later analysis or to create a campaign targeting that user segment.

RFM User Segments

User Segment
Description

Champions

These users are your most active users. They have the highest recency and frequency scores.

Loyal Users

These users have the highest frequency of use with strong recency scores.

Potential Loyalists

These users have visited your app very recently and have the potential to become loyalists or champions.

New Users

These users are your most recent users with low frequency scores. Strong candidates to encourage repeat use.

Promising

These users have high recency scores with the potential to become high frequency users.

Needing Attention

These users have above average recency and frequency scores.

About to Sleep

These users have below average recency and frequency scores. May slip away if not engaged with.

At Risk

These users have above average frequency but low recency scores. Strong candidates to re-engage.

Cannot Lose Them

These users were active at one point in your app, but haven’t been back recently. Strong candidates to re-engage.

Hibernating

These users have the lowest recency and frequency scores. May be lost.

How RFM Analysis Works

For every user who has performed the selected event, our RFM Analysis will segment users based on:

  • How many times a user performed the event.
  • The last time a user performed the event.

Calculating Recency Score
Once we’ve determine how recently each user in your analysis performed the selected event, we rank them in order of percentile (person who has performed it most recently would constitute the 100th percentile), then we rank the users on a score of 1 to 5 based on their percentile, with 5 being the highest.

Calculating Frequency Score
Once we’ve figured out how frequently each user in your analysis performed the event, we rank them in order of percentile (person who has performed it most frequently would constitute the 100th percentile), then we rank the users on a score of 1 to 5 based on their percentile, with 5 being the highest.

Note

Currently, in RFM analysis, we sample your data to load it faster. We sample the results only when the query exceeds 50,000 events. The sample size is 10%

You may see a minor difference in counts when you compare these numbers with the numbers in your dashboard. However, the numbers are directionally correct.

Updated 4 months ago

RFM Analysis


Suggested Edits are limited on API Reference Pages

You can only suggest edits to Markdown body content, but not to the API spec.