Automate your list cleaning process

Hiya!

It depends what you mean by list cleaning . . . 😊

We use several filters and segments to help ensure that we are sending to only folks who have a valid email address, are active, and want to receive messages from us.

1. If you haven’t deleted it, you should still have an example segment called Invalid Email Address that was created when you initiated your workspace. It looks for People without an RFC 5322 compliant email address, including blank email addresses. If there’s no phone number, ID, and/or activity for these profiles either they might be candidates for deletion. If you removed the segment, it’s easy to recreate. 

   

    

2. An email engagement filter (segment) helps to make sure that we are only sending emails to those who have interacted with our emails in some way over the last 120 days. It’s useful for any campaigns that are NOT onboarding campaigns. This helps a lot to keep deliverability high.

   

3. Creating a segment that defines a Sunset policy is also super useful. The example in the docs is specifically filtering on email and should definitely be implemented, but you can add other conditions for things like the existence of a device, Push message delivery, phone number/SMS delivery, product activity (if you’re sending that in), etc. These are folks that you’d want to exclude from all or certain types of campaigns/broadcasts and consider for removal.

   

    

4. If you’re looking to get a bit more sophisticated on whether or not emails are still valid/in use, you can also investigate email verification tools like Kickbox, ZeroBounce, Briteverify, etc. Customer.io provides a recipe for integrating with Kickbox on the fly. 

All of the above segments can be added as filters for campaigns and broadcasts. Invalid emails and Sunset segment profiles also might be considered for removal after a certain amount of time if their history is not important to you. Since all that history is forever lost when the profiles are deleted it can be a tough call. 

Hope that helps!