I am attempting to created a weighted ratio/score for customers based on the number of support tickets they have entered in a time period against the number of units they have in service with us.

Example :

Small Customer - 5 Support Tickets / 20 Total Units Supported = .4

Large Customer - 50 Support Tickets / 350 Total Units Supported = .14

I would like to have this ratio be weighted based on the number of total units supported. In this scenario the larger customer has a better ratio, but in reality are a much larger risk for a loss and ideally prioritized above the smaller customer. The units counts of the customer base are extremely varied from a few to a thousand so its difficult sift through the noise of smaller customers.

My initial thought would be to apply some kind of multiplier based on unit count. In the scenario below even if the larger customer was amplified X3 it would put them in the same level of urgency.

Are there any other ways of quantifying this ratio based on the value (units in service) of the customer?

  • $\begingroup$ What is the score attended to represent? $\endgroup$ – whuber May 8 '19 at 18:26
  • $\begingroup$ @whuber the score is intended to flag those customers who may need extra assistance due to the amount of technical issues they are submitting. If a customer has one support ticket for every unit in service this is a scenario where we would want to reach out. The issue I am having is how to create a score that prioritizes/weighs larger customers as they represent more value and a greater loss risk if we don't intervene and support. In the scenario above ideally the larger customer is prioritized in score. $\endgroup$ – Josh May 8 '19 at 18:33
  • 1
    $\begingroup$ It sounds like you need to supply additional information about "value" and "loss risk," as well as data about intervention and support experiences, if you are to have any reasonable chance of concocting a useful score. $\endgroup$ – whuber May 8 '19 at 18:35
  • $\begingroup$ @whuber What type of information would you expect on those data points? What in your view would make this equation complete? The "value" is simply being defined by the number of units the customer has. Obviously two customers with the same issue:unit ratios could be viewed as having a similar level of "risk" since the same percentage of thier unit base has experienced issues. The larger customer would pose a greater priority due to the number of units supported (What we considering the value of the customer ) $\endgroup$ – Josh May 8 '19 at 18:59

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.