Peek - Churn type of analysis

[stevegimbrollmt]

2015-03	Product1	652811	0.28	Recurring
2015-02	Product2	652811	0.05	New
2015-02	Product1	652811	8.547149	New
2015-04	Product2	652811	3.2	Lapse
2015-04	Product2	652811	2.2	Reactivated

Where New is (played product 1)

Recurring is (played on both months)

Lapse (played last month but not this month)

Reactivated (did not play last month but played this month)

I am so confused on what to do, I am using peek to work it out but im finding an issue identifiing New and Lapse or Reactivate,

Do I need to add a null value in between?

If confusing pls let me know

[Gysbert_Wassenaar]

Perhaps this discussion helps: how to count null records

[#3]

Hi,

What you want to achieve please clarify more..

[stevegimbrollmt]

I did actually see that but it is not dynamic as I wish,

I want to achieve

*Lapse customers (played last month not this month)

*First time players (never played before)

*Reactivated players (did not play last month, played this month)

In terms of the definitions above can the data by product,

for lets say 2 years, so I do not want to stay selecting manually the selection like in that null record example

Thanks, is it a bit clear?

[#5 · Accepted solution]

Hi,

PFA,

Hope it will match with your case.

you can follow same approach for other flags.

Thanks

[stevegimbrollmt]

Hi, Thank You for your help

After going through your sample, and tweeked it a bit to use a date field, I managed to get close to what I want,  I am sure I can fine tune the code but it looks like it did the job

Thanks