You have a model, and you use the model to predict the probability of an individual customer visiting your website in the next thirty days, right? RIGHT?
There are two metrics worth modeling - scoring your database on a daily / hourly basis.
- Probability of Customer Visiting Your Website In The Next 30 Days (re-visit rate).
- Number of Customer Visits - Should Customer Re-Visit Your Website.
If you know that a customer has an 85% chance of visiting your website next month and will visit seven times, well, then you know that this customer is there every four days ... and you better not bore the living daylights out of this customer, right?
If you know that a customer has a 15% chance of visiting your website next month and will visit twice in the next month if the customer visits, well, now you have to craft a clever marketing story to get this customer to put her fingers on the smart phone, right?
Most important - you're going to need multiple stories for multiple customer segments, right?
But you can't get started unless you know how likely each customer is to visit your digital presence monthly. And worse, if your customer interacts with you on Instagram and not on your website, then your website is a glorified order form, and you have to get intelligence about what the customer is doing elsewhere.
So please, perform the analysis. Score your file on a daily / hourly basis.