Segmenting by Time... any insights out there?

julies4559538 25-01-2018

Here is the use case:

People launch an app numerous times within a day but they may not use the app every day.

I want to look at people that use the app for 5 or more days within a quarter ​and classify them as my "loyal" app users.

Now, I don't have the people metric implemented so I will settle for unique visitors.

But is there a way to classify time as 5 or more days within a segment?

All of my "day" dimensions ask for a specific date...

Any thoughts or suggestions around this?  I'm all ears.


Accepted Solutions (1)

Accepted Solutions (1)


Would this work?

The 'page views' exists is repeated 5 times. I tried this in my report suite and it appeared to be a solid segment definition for me.

Answers (9)

Answers (9)


Dear Juile,

Not sure whether we can take the results now.

But if you are planning to do the implementation for future use, please see Visitor Lifetime Value.

Based on your logic, you can increase the Life Time Value of the customer to bucket accordingly! For your case, Life Time Value should increase if a customer open the App in different days.

Thank You!


julies4559538 08-02-2018

I got it!  Maybe because I logged into analytics and went right to components to create the new segment...?

But I went to a report and then created the segment and it seems to work!  Thank you for your help.


Can you double check that's the right report suite? "0 of 0" means that there's no data in that report suite for the last 90 days. If it was an issue with the segment definition, it would say "0 of 1000" or "0 of 43,000,000" or however big your data set is.

julies4559538 08-02-2018

Dear Gigazelle,

Love the logic, however, I can't even save this as a segment.  My 90 day sample box also yields 0.  I tried adding it into containers, taking it out of containers, etc. but no luck.  Any thoughts as to why?  Thanks!Segment.PNG


i‘m not sure you can do that „the normal way“ since the question is a) at visitor level and b) need the information looking back...

i can think of the following idea: why don‘t you create  a virtual report suite with custom settings for „context-aware sessions“? i would create one with visit definition of say 12 hours, trying to capture all visits of one day (this settings would capture alle hits within 12 hours as 1 visit). doing that, you have an approxiamte count of „days“ where users used the app, and can divide “visits“ it by total days to get an average frequency. not quite what you looked for but best what i can think. maybe play around with the time frame until it matches best for your case.

more information about Context-Aware Sessions

pabloc82923542 25-01-2018

What about unique visitor metric? It wont allow you to see one specific user but it would allow you to see the sum of unique visitors (some duplication as users on multiple devices are counted multiple times then)

Bear in mind Adobe analytics is a aggregate level type of data collector. The "if a user used an app on day 2,5,17,23 of a given month" would then require the specific user to be identified(tagged) on each visit. Do you have user IDs that are assigned to users in a logged in state? if so then you could do a report by each user or pull a daily report by user ids. This will not help you for non logged in users though. There is a marketing ID in the latest DTM implementation but it is a cookie level identifier of all traffic visitors. it doesnt de-duplicate users who access via multiple devices as an example. Not sure if its available in iOS/android environment.

good luck

julies4559538 25-01-2018

That just gives intervals between sessions and it's at the visit level. 

I want to see by day; if you used the app one day in January, 2 days in February, and then another day in August.  I don't care how many times within that day you've launched or used the app.  Just trying to get overall usage over time at at the visitor level. 

Does that make sense?

pabloc82923542 25-01-2018

Have you looked at the return frequency metric? It allows you to establish user visits by:


less than 1 day
1 to 3 days
3 to 7 days
7 to 14 days
14 days to 1 month
longer than 1 month

This can then be used to segment your reports and other key KPIs.