I am looking to build a model to enhance CTR. Below is the business description.

  • We have a coupon based website.
  • For each retailer, we have a retailer page.
  • Each retailer page will have up-to 100 coupons

What I want to do is to build a sort order of coupons based on a model so that it increases the CTR from our existing CTR.

We have click data. What it means is that every time someone clicks on the coupon we create an event in the database.

Below are some of the features we store around that

  • coupon_id
  • user_id
  • time
  • coupon position
  • coupon type
  • coupon text

I need some help on what kind of model should I build. I was first thinking of using logistic regression but then every time some one loads a page but doesnt click on a coupon, I cannot use any of the features related to coupon so the only 2 features that remain are user_id and time. Also since its a coupon site, there is very less loyalty so people don't really come back that often.

If I build a model by taking the sum of clicks on a particular coupon as a dependent variable, then I can use all the coupon related features as independent variables. However, I cannot factor in time and user id in that case.

Any suggestions on what would be best approach to build a model like this? Has anyone built a CTR model for a page like this in the past?


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.