For dimension 1, I’d start with the classic N-day/week/month retention and investigate the rest later. With so many measures, which one do we use and where do we start? Should we define “use” as visiting the product page, staying for a certain amount of time, conducting certain actions, or purchasing a product? We call the actions we use to calculate retention “key actions”. When we say users use the product, we didn’t define what we mean by “use”. However, in the meantime, the new user retention could go up, suggesting that this feature change could be an improvement in the long run. In the scenario of feature changes, we might see active user retention goes down because users have already gotten used to the product. Active user retention measures the proportion of active users who stay active. For example, new user retention measures the proportion of new users who stay active. It is often important to calculate retention for different user statuses. Active user: active users who are not new and not reactive.Reactive user: active after churned/inactive.Here is one way user status can be defined: There are many ways to define status, and different companies/products often have their own ways to define user status. First, let’s take a look at how to define user status. In addition to time, user status is often another important dimension to consider. For example, Pinterest measures “ the percentage of new signups that are still doing key actions during a one-week time window of 28–35 days after signup”. You can define whichever timeframe you are interested in. rolling retention) measures among users who first used the product at day 0, what proportion of them are still active on and after day N.īracket retention is more flexible. For example, for gaming products with high stickiness, it’s typical to measure N-day retention on a daily basis. Whether to use daily, weekly, or monthly retention depends on your product and how often users use your product. N-day retention is the most classic way to calculate retention, which measures among users who first used the product at day 0, what proportion of them are still active at day N. Understanding different measures of retention and comparing them will help us find the appropriate retention measure for your product. There are three dimensions to consider in terms of measuring retention. Retention measures how many users return to your product over some specified time. Thus, the AARRR framework was later reprioritized to the RARRR (Retention, Activation, Referral, Revenue, Acquisition) with Retention as the No.1 priority.ĪARRR reprioritized to RARRA (image made by author) How do we define and measure retention? There are many reasons why retention is more important: acquisition strategies (e.g., ads) are expensive and it is often cheaper to retain a user than get a new user retention is the foundation of growth user retention is more directly connected to revenue than acquisition. However, many people critique that AARRR focuses too much on Acquisition even though Retention is more important, especially for many internet products that struggle to retain people. One of the model popular business growth frameworks is AARRR (Acquisition, Activation, Retention, Revenue, and Referral). Why is it important to do retention analysis? This article will cover three dimensions of measuring retention (time, user status, action), analytical frameworks, how to find the “aha moment” and the “habit moment” through analysis, and investigate why users leave and stay. Are you a data scientist seeking to understand users of your product or website? This article will walk you through a user retention analysis framework from a data science perspective.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |