Recency and Frequency Matrix

Recency and Frequency Matrix

The Recency and Frequency Matrix automatically divides your audience into segments based on how recently, and frequently they have performed an event that you have selected. The segments range from “Champions”, who are your best customers, to “Hibernating” users who have not been active recently and are prone to churning.

How the Recency and Frequency Matrix is built

Choose an event that signifies activity for your application, and set the date range you want to analyze. An e-commerce app might track purchases, while a media app might monitor content viewed or rated as their activity event.

For every user who has performed the defined event, the RF Grid will segment users based on:
– How many times a user has performed the event
– The last time a user performed the event

Calculating Recency Score
Once we’ve figured out how recently each user in your analysis has performed the 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 has 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

Reading the Recency and Frequency Matrix

Your users are automatically segmented based on their Recency and frequency score which is calculated as explained above. Each segment has its own score range and you can see who are the users that belong to each segment, and further use this information for analytics or engagement by clicking on any segment in the Recency and Frequency Grid.

Rankings

  • Champions: Most active users. Highest recency and frequency scores.

  • Loyal Users: Highest frequency of use with strong recency

  • Potential Loyalists: Very recent users with the potential to become loyalists or champions

  • New Users: Most recent users with low frequency scores. Strong candidates to encourage repeat use.

  • Promising: High recency scores with the potential to become high frequency users

  • Needing Attention: Above average recency and frequency

  • About to Sleep: Below average recency and frequency. May slip away if not engaged

  • At Risk: Above average frequency but low recency. Strong candidates to re-engage.

  • Can’t lose them: Were some of the most active users at one time, but haven’t been back recently. Strong candidates to re-engage.

  • Hibernating: Lowest recency and frequency scores. May be lost