JT on EDM

Graham Hill wrote a piece on Evidence-based CRM that focused on evidence-based CRM programs and it made me think about evidence-based CRM processes.

To me, evidence-based CRM means customer relationships, and thus customer treatments, that are based on evidence (data) and not judgment, hope, guesswork etc. It means

• making offers that you have evidence this customer will want
• testing things to make sure they work before rolling them out to everyone
• ensuring that different agents will take the same, evidence-based, action with a customer
• treating customers as risks (fraud risks, collections risks, retention risks) based on historical data
• formulating policies, segmentation, and customer treatment approaches based on data
• ensuring that these evidence-based actions are taken consistently across channels

Making CRM evidence-based is, I think, another way of saying that CRM should be decision-centric. If you externalize and manage the decisions that impact your customers – that contribute to the customer experience and so build the customer relationship – then you can drive those decisions with evidence. If you don’t then you can’t. Evidence-based decisions would:

• use predictive models to turn uncertainty about the future or a customer’s preferences or risk into probabilities
• use data mining to find the treatment rules or segmentation that have worked in the past
• group customers into segments that are both alike and statistically significant so they can be treated similarly
• be consistent across agents and channels because they are automated
• use adaptive control – champion/challenger or A/B testing – to try new approaches, new models on a subset of the population before rolling them out
• use simulation to see what the impact of a change to a decision might be (or would have been) before trying it at all

In other words, Enterprise Decision Management, EDM, is the best way to deliver this kind of evidence-based CRM.

While on this topic you might want to check out a couple of articles in DM Review this month. First is this one on Next-best offer and the use of predictive analytic – well worth a read for a nice overview. The second is Six Best Practices for Delivering a Successful Customer Experience. In this Ray talks about a customer-focused strategy and gives a great example:

For example, when designing solutions to implement a consistent, channel-independent customer experience, a specialized offer service can be created to provide access to offer-generating algorithms regardless of the channel.

Well of course I completely agree – an offer service is after all a decision service by another name. Ray also makes the point that organizational transformation and continuous measurement are critical for improving the customer experience just as they are for improving the decision making that drives them.