Churn Survey

Ed Sander
Member

Posted 19-Feb-2006 11:26 PM
I noticed that while our company is trying to fight churn with promotional activities we hardly know why people walk away. Sure, we've got certain gut feelings but there's no cold hard facts. And all of the promotional activities don't even seem to counter the suspected issues. I'm planning to make a turn-around in this attitude by actually asking some ex-customers who bought with us in 2004 and didn't buy in 2005 what the reason for this was. In other words, a churn survey. I'm planning to do this by e-mail and donate 1 euro to charity for every fully filled-out questionnaire, in order to stimulate response.

I would be very interested to hear any suggestions and see actual examples of churn surveys. Your help is therefore highly appreciated.

Cheers,

Ed

Ed Sander

Business Development Manager
Viking Direct/Office Depot

Database Marketing Specialist
Failsafe Database Marketing

George R
Member

Posted 21-Feb-2006 11:47 AM
Sounds like a good idea. I like the donation part.

I have had a number of clients attempt such surveys with limited to no success. Keep us posted on results and I will pass this along to them.

George Reavis
ThankingCustomers.com

Max
Member
Picture of Max

Posted 28-Feb-2006 10:21 AM
Hi Ed (Sander)!
Your problem is the problem of the world, so do not worry.
Now, in a way you are smart. You do not know why your customer churn and ,you do not, like many, make the mistake to assume (even decide) why they churn and act accordingly (although your promotions tend to do this a bit)and...you guessed, this is almost garanteed disaster.

Have you done behavior models?

It is far from complex.
They show you what happens in the «non financial $» life of the customer which you do not see if you use the CLTV as the sole measurement of satisfaction. Better, be careful as your best customers might churn from a day to another, this is a scenario that has happened frequently.

Your survey idea is a good one, but surveys are surveys especially on ex customers. They might not tell you the truth.

A good way to start, would be to clone the data of your ex customers on to current ones (within the same segment) and see if there are similar ones and survey them while they are still with you and compare with the ex customer survey results.
Also, you will discover the breaking points and be able to act before your customers walk away.
Then you will have a model which will evolve as you add variable data to it with time.
The thinking behind it is that while I am a customer, I want to improve my relation and the way you treat me.
Once I have left I have less motivation to help you. Sorry, but that's human nature. If you need to know more, let me know. That should help you.
Look for signals: how many have phoned before churning, when did they phone? when did they churn? After receiving a bill? After returning some merchandise?
Also, «paying» for the survey risks to bias the results as some people will tell you.
I hope this can help you

Ed Sander
Member

Posted 28-Feb-2006 11:42 PM

Hi George,

We decided to try running the survey without the donation part on part of the total sample group first and do a second batch if that wouldn't work. Guess what, the response so far on the first batch has been marvellous ! Within one day I'm already hitting a 15% response rate ! The results are quite interesting and besides confirming some general 'gut feelings' there's also some stunning stuff which we'd never really considered in the answers.

And what's more .... the questionnaire has even triggered some customers to place another order. Quite a cool side effect, don't you think? ;-)

Ed Sander

Business Development Manager
Viking Direct/Office Depot

Database Marketing Specialist
Failsafe Database Marketing

Ed Sander
Member

Posted 28-Feb-2006 11:48 PM
Hi Max,

Yep, we're running various behavior models, some more succesful than others. Problem is that these could predict *if* and *when* customers are churning, but not *why*.

Very true, and previous experiences with market research have taught me to be careful with the results, especially when asking for intentions. Then again, even information we have to take with a grain of salt is much better than no information at all. It will at least tell us where the problem areas might be.

Good idea. I will keep this in mind as a follow-up activity.

Thanks for your feedback !

Ed Sander

Business Development Manager
Viking Direct/Office Depot

Database Marketing Specialist
Failsafe Database Marketing