How to use Retention Dataset (ThoughtSpot)
Retention Dataset
This dataset is geared towards executives, publishers, marketing managers, or anyone responsible and/or interested understanding how well they are doing with retaining customers. It can be used to analyze all retention events ie anything that retains the customers between term. Noteworthy features include the ability to analyze retention and attrition rates, understand the timeframe of retention and attrition events and use these rates to help project future retention/renewal revenue. Attrition is used to analyze how customers are not being retained. There are two different ways to look at retention and attrition, at the term or lifecycle level. We’ll explain the distinction in a later section but please note for general use cases, use lifecycle fields. Term fields should be reserved for advanced use cases.
Provides:
An analytical look into how well subscribers were retained.
Insight into when and why subscribers are falling off of your file.
Use Cases for this data set:
Compare Front-end vs Backend retention
Identify how many new subscribers have I added by month and year
Identify how are the subscribers retain by key factors like channel, offer, creative and campaign
Identify what was the event or effort that caused their attrition (customer cancellation, non-renewal or expiration or system cancellation for non-payment)
Identify what was the event or effort that retained the subscribers (auto renew, recaptured via upfront promotion, responded to a special renewal or renewal series effort) and at what point in time did that event occur
Identify how are they retaining over time (at 3 months, 6 months, 1 year, 18 months, 2 years, 3 years and 4 years)
Dataset Details:
The source of this dataset is Advantage.
Data is refreshed daily.
Table of contents
Table of Contents | ||
---|---|---|
|
Using this Dataset
Key Considerations:
Lifecycle vs. Term Fields:
This dataset will include ‘Lifecycle’ fields and ‘Term’ fields. Example: Lifecycle Promo Code vs. Term Promo Code.
These cannot be used interchangeably.
Lifecycle fields will reflect the attributes of the initial subscription that began the lifecycle, and will cohort all subsequent terms of the lifecycle back to the initial purchase (similar to original retention).
Term fields will reflect the attributes of the individual terms within lifecycles.
Example:
If you filter to lifecycle promo code ABC, you will be able to see how the subscribers who began their subscription on this promo code were retained over time, regardless of the promo codes tied to their subsequent term purchases.
If you filter on term promo code ABC, you will only see the subscription details within this term. So, if customers began on this promo code BUT were retained at one year on promo code XYZ (with no lapse), you will not see the retention event since you are only looking at terms tied to that promo code.
Note: For general use cases (and to mimic original retention), use lifecycle fields. Term fields should be reserved for advanced use cases.
Excluding folds:
Original retention excluded folds on default. To replicate in this dataset, exclude channel = V.
To view paid subscribers, select
Is lifecycle comped? No
Is Lifecycle Unpaid? False
12 month vs 14 month retention:
Timeframe is an important consideration. If you would like to look at retention over a year, you should use 14 month to allow time for any refunds and chargebacks to occur
There are a few ways to analyze active subscribers by using different fields and flags
Is Lifecycle currently active: True
Lifecycle status: Active
Key Terms/Measures:
Initial Batch Subs:
Using paid and comp flags to get to ‘paid’ subscribers like original retention
Is lifecycle unpaid: false
Is lifecycle comp: false
Count of customers who subscribed to a subscription.
Caveat: No flag for ‘non-pay’ customers currently (so bit of a misnomer) but there are plans to add it.
Original retention only counted subscribers who were fully paid (i.e. excluded installment, comped or trail subscribers).
Lifecycle:
Continuous time active on a subscription. A lifecycle will include all the terms within a subscription as long as there is no lapse.
Lifecycle Date – Start:
Start of the subscription lifecycle. Reflects the order date vs. the date of the first issue (original retention).
Lifecycle Date – End:
End of the subscription lifecycle. A lifecycle will end if there are no active terms tied to a subscription.
Anything currently active has a null end date.
No grace period; if a subscription is cancelled/expired and a subscriber re-purchases, it will start a new lifecycle*.
Exception: If someone gets suspended and then re-activated, we will consider them active that entire time.
Note that original retention had a 3 month grace period, so if a subscription ended and was restarted within 3 months of the end date, it was shown as one continuous lifecycle.
Cancel subs use event date of the cancellation for end date.
For non-pay cancels who were in bill suspend, their end date will be the bill suspend date vs. the true cancel date.
Term:
The terms within a subscription/lifecycle.
‘Term’ fields provide the ability to analyze subscriptions at the term level and see how many terms are included in lifecycle.
Any term that never starts (future term that gets cancelled) will have a null term start date
Term Length Descriptions (Term and Lifecycle):
Monthly = .07 - .1 (~ 1 Month (.08))
Quarterly = .23 - .28 (~3 Months (.25))
Semi-Annual = .46 -. 55 (~6 Months (.51))
Annual = .92 - 1.1 (~11 - 13 Months (.93 -1.1))
Two-Year = 1.84 - 2.18 (~22 to 25 Months (1.85 - 2.19)
Greater Than Two Year = >2.18 (>= 26 Months (2.19))
Days btw term order data and lifecycle start date:
Helpful for showing the timeframe of the retention event.
Days btw lifecycle start and end dates:
Helpful for showing length of lifecycle.
Retention Count vs. Flag vs. Rate Fields:
Retention Count fields: Number of initial paid subs who made it to given milestone.
Retention Rate fields: Divides initial paid subs who made it to given milestone by the initial paid subs of the cohort.
Retention Flag fields: Applies a value of 1 to initial paid subscribers who made it to the milestone and a value of 0 those that did not.
Good for filtering to only include/evaluate initial paid sub who made it to that milestone.
Starter Board
Retention Starter Board
Tip |
---|
Wrap upHopefully this article helped walk you through the purpose and intricies of the Retention dataset in ThoughtSpot. A guided analytics app in Qlik Sense will be developed as well in the future. |
Info |
---|
Still need help? We know this can be frustrating. To get further help please open a Support ticket. |
Related FAQS
Expand | ||
---|---|---|
| ||
Expand | ||
---|---|---|
| ||
Related Content
Level up with a Masterclass
Evolve course title
Evolve course title
Note |
---|
Cannot find the article you’re looking for? Suggest a new article here! |