What exactly are Retention tables?

Retention tables show a group of people that visited your site or used your app for the first time during a certain time frame. They also display the progressive 'drop-off' or decline in activity over time for that particular group (a cohort). 

What does a Retention table look like?

How to use Retention tables in Smartlook

On the Events tab, select an event:

Retention tables rely on data displayed in the event charts. These are the two (2) settings that affect Retention tables:

(1) You can choose a date range for the Retention data: The date range determines how long Smartlook can identify the same user as returning.

(2) You can choose the precision by which you want to analyze your data: You can switch between days, weeks or months. Different views will be useful in different situations.

When you’re happy with the settings, scroll below the event chart and statistics. On the Retention table click "Generate". You can generate up to four retention tables from four events at the same time.

When you click on it, you’ll see a Retention table similar to this one.

(This Retention table displays 7 days of activity in cohorts segmented by day

What exactly are cohorts?

A cohort in Smartlook is a group of people that visited your site or used your app for the first time during a certain time frame. It could be the same day, week or month. It depends on your needs and on your Smartlook settings.

Following a cohort:

Below you can take a look at a new cohort in a Retention table to see how it progresses:

(1) The newest users
Here is how a cohort appears on Monday, October 19. In the dark blue, 376 users completed an action. The column group here is named "Day 0"

(2) The next day:
On the following day in a Retention Table, we can see that only some (76) of the original cohort (376) performed the same action the next day (in column "Day 1").  

(3) 5 days later:
Smartlook has collected data about the cohort starting on Monday 19th for six days.

The above image displays the retention of the group starting Monday 23. From left to right, each column represents how many of the original group completed the same action on those following days.   

The same image can be expressed in percentages (%)

How to understand the data:

Cohort size:

The value on the left, the "Cohort size" (389) is the total number of new visitors first seen on Monday 19th, and triggered the event at any time within range of the Retention table. You can think of this group as the amount of 'enrollments' on that date. 

The number on the right in column "Day 0" (376) is the number of unique users that triggered an event on the same day they were first seen in Smartlook. You can think of this group as the amount of 'new enrollees' who showed up for the first class.

Cell colors

Colors are an important part of any Retention table. Events that are triggered more often have darker cells. This lets you see trends in the way events are triggered. 

What are the columns?

The 0 "Zero" column is the original cohort size which completed an event. It can be grouped in days, weeks or months. Here is an example showing cohorts with weekly values. 'Week 1' represents the following 7 day period. Each following column represents the following chronological period.

Display values

The entire Retention table can display values as numbers or percentages. Simply use the selector in the top left:

Additional details

When you hover the mouse above any frame in the Retention table, a detailed explanation of the cohort is given. 

So why should you use Retention Tables? 

Visitors who come to your site during a specific time often behave in a similar way.

It may be due to events that are happening outside of your company, but it can also be due to the actions you take. Retention Tables help you to look past growth and focus on engagement. If you only look at absolute numbers, it’s easy to get a false sense of success. Oftentimes, a lot of users might try some function during the first couple of days only to never come back to it.

New users are usually more engaged

New users are usually more engaged so when their numbers grow, the averages might be going up even if your retention stays the same. This makes it easy to overlook areas where your efforts could be improved.

And identifying these problems is one of the things Retention tables excel at.

Retention Tables are useful for:

  • UX designers & Product managers - Following cohorts give you detailed information about the adoption and retention of any function.

  • Marketers can use Retention tables to analyze the quality of users brought by a marketing campaign. And compare it to other sources of traffic.

Retention tables are an extremely powerful feature that takes almost no time to use. They are a vital tool when it comes to improving retention. The data is easily understood thanks to its visual nature and provide a great overview of how good you are at teaching users to adopt a new function.

Did this answer your question?