I am wanting to predict how long (if ever) a customer will transact with us before churning in R using survival analysis

Currently my data is at the transaction level since a customer can return after churning. Churn is defined as not transacting for 3 months

Here is an example of the data that I was looking to analyse:

enter image description here

Should I be aggregating the data at the customer level? I'm unsure about this since I was to be able to have a column for each transaction that gives the probability for that customer to churn, 3 months from the transaction date

What would be the best approach here? Thanks a lot

  • $\begingroup$ As an experiment i'm using transaction level data, but at transaction, looking back for that customers to aggregate some metrics. I think this seems to be like a sensible compromise? $\endgroup$ – SuperSecretAndHiddenFromWork Jul 16 '18 at 15:10

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.