0
$\begingroup$

I am working with a dataset with about 10,000 customers. About 3,000 engaged with dozens of marketing campaigns over the years.

I am trying to create a model to find which marketing campaign to use on a given customer. I was thinking of creating affinity scores based on conversions from these campaigns and do some sort of random forest/logistic regression. It gets tricky because only 3,000 have engaged and I am not sure what to recommend for the other 7,000. Any suggestions?

$\endgroup$

1 Answer 1

0
$\begingroup$

Use an unsupervised method such as clustering to group users, then assign marketing campaigns that have been used by others within the same cluster.

$\endgroup$
2
  • $\begingroup$ thanks, but let's say in cluster one, there are dozens of campaigns that were successful for those customers. which campaign do you then choose? $\endgroup$
    – Alex
    Dec 2, 2022 at 22:45
  • $\begingroup$ The most common/successful ones $\endgroup$ Dec 3, 2022 at 23:32

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.