Solutions
Customer Retention
and Next Best Action
With the same data integration and just a few predictive queries, deeply understand each of your customers and ensure you do what it takes to prevent churn, encourage higher levels of spending over time, and find new business opportunities.
Related case studies
Powering better targeted next best actions for…
A leading US online personal finance & banking company wanted to improve revenue by doing…
Read case studyEnabling rapid model prototyping for a global…
In an effort to help reduce customer churn and retain revenue, this payment company had…
Read case studyWithout any ML expertise required
-
Flexibly predict any aspect of future customer activity in a single sitting:
- Changes in aggregate views/purchases/etc.
- High impact events like membership cancellation/app uninstall, etc.
- …and more, across timeframes most relevant to you taking business actions (ranging from days to quarters)
- Pinpoint exact events/changes in a customer’s history that indicate retention risk or, on the flipside, that they are ready for spending expansion
- Uncover key drivers behind both customer churn and expansion, at both a macro level and an individual customer lev
- Easily add 3rd party holidays, events, macroeconomic/demographic data to enrich your predictive accuracy
- Make not only short-term predictions, but also long-horizon predictions for each of your customers that factor in seasonal trends
Downstream workflows turbocharged
-
Optimization of go to market campaigns of all kinds to build customer loyalty:
- who to target
- which action to take (emailed promotion, sales outreach, etc.)
- what content/incentive levels to include
- Financial planning and analysis
- Targeting your market outreach and making sure the right discounts/promotions and other content are sent to the right customers at the right time
- Fixing your products, services, and customer experiences to further prevent churn and drive growth
- …and more!
Representative workflow
Define the ML Problem
Structure your problem using a simple query to predict affinity to categories, brands, and products.
Predict 90-day churn across 3-month active users:
PREDICT Count(Sessions.ID, 0, 90) = 0
FOR EACH Users.ID WHERE
COUNT(Sessions.ID, -90, 0) > 0
FOR EACH Users.ID
Predict 90-day churn across 3-month active users, assuming they get a coupon this week:
PREDICT Count(Sessions.ID, 0, 90) = 0
FOR EACH Users.ID WHERE
COUNT(Sessions.ID, -90, 0) > 0
ASSUMING COUNT(Coupons, 0, 7) > 0
Predict 2-day notification open rate:
PREDICT EXIST(NotificationOpen, 0, 2)
ASSUMING EXIST(NotificationSent, 0, 1)
FOR EACH Notifications.ID