Jim answers questions from fellow Drillers

Topic Overview

Hi again folks, Jim Novo here.

So, how to you actually set up the measurment for a retention program? Since you will ultimately want to conclusively prove that what you are doing is working, the 1st thing you have to is measure what the current retention rate is, hopefully by segment so you have a basis for creating successful program features. But how do you do that? Read on, fellow Driller!

---

Q: I got the job of servicing the top users on our network.  I must confess I dazzled the interviewers with statistics from your Recency, Frequency, and Monetary model !!!

A:  Congratulations!  Well done.

Q:  I know I should be paying you for this but I would like to communicate with you once in a while to tell you what’s going on..?  If you are not happy with this I really understand.

A:  I’m fine with it, as long as it doesn’t become a full time job!

Q:  Here’s my plan:

1. I would like to start by segmenting high users by usage, (usage bands)

2. Next I would like to profile demographics for each usage band

3. Next I will keep a Monthly tab on their account credit refills i.e. Frequency of refills in a month and Monetary value of refills

4. A special helpdesk will be set up specifically for them (top users) to log complaints hence we will have a log of their complaints.   All their complaints are to be resolved in four hours (if it is not network related)

5. We will communicate with them by calling twice a year (they don’t like being called too often) also we are thinking of sending birthday text messages (it is a form of communication)

That is all I can think of for now.  If you have any tips I would really appreciate it.  Thank You for helping me get this job.

A:  You’re welcome, glad it worked out!

As far as financial success of your program goes, I’m not sure how helpful the demographics will be, at least in the beginning. That would typically be a topic for “Phase 2”.  Ultimately, you will be judged on the increased retention and related profitability of these top users; you are going to have to prove what you are doing is working.  So it is very important to first establish exactly where you are with retention.

The usage bands idea is a good one, but that only gives you the segments, you need to then determine defection rates for the segments, since a drop in this defection by usage band metric is probably the best proof you will have that your program is working. And once you have the defection rate and the value of a customer in the band, you can get to program profits. If possible, it would also be great to have source data – sign-up campaign media source, offer, content of campaign etc., so you can sub-segment defection source within the band.

For example, let’s say you find out there are 1,000 customers in the “best” usage band, they are worth $100 in profits annually, and their 1 year defection rate is 30% – a year after these 1,000 customers start, there will only be 700 of them left.  You implement your programs, and the annual defection rate drops to 20%, meaning there are 800 customers left at year end.  

This means your programs are responsible for retaining 100 customers with an annual value of $100 – you generated an additional 100 x $100 = $10,000 to the gross profit line before the cost of your programs.  If your programs for these 1,000 customers cost $2,000, you increased cash flow by ($10,000 – $2000 = $8,000 or $8 per customer ($8,000 / the original 1,000).  This is the kind of “proof” you are looking to have – the kind of proof that will make management really sit up and notice!

So, go back 2 years or as far back as you can, and look at all the customers who joined, and segment by usage bands / recharges.  Then ask this question: what percent of the customers in each band are still customers after 1 year, and what percent are still customers today?  These are your baseline 1 year and 2 year (if you can go back that far) retention rates for each band.  Subsegment if possible by campaign / source that created the new customer. Your mission is to improve these rates, and in so doing, you will increase cash flow / profits.

Then perhaps you can look at demographics to help design your programs, but I would do something else first, if possible.  Interview defectors, and find out why they are leaving.  You can interview Recent defectors from each band, you don’t need to talk to the ones from 2 years ago.  What they tell you will be more important to designing your program than any demographics (save geography, which will highlight potential service problems).

You should be really most interested in what people do and why (behavioral models), rather than who they are (demographics), because behavior predicts behavior, demographics usually do not predict behavior, but do potentially help with fine-tuning campaign messaging.

Get the book at Booklocker.com

Find Out Specifically What is in the Book

Learn Customer Marketing Concepts and Metrics (site article list)

Download the first 9 chapters of the Drilling Down book: PDF