Marketers have been cranking away with technology to address how and when they reach out to their customers. And they have been successful, sometimes too much.  As these processes have been created then automated, it’s become too easy to flood customers with messages coming from all different directions. Meanwhile, customers are concerned about getting too many messages they do not want. It’s time to take the customer conversation to the next level.

The paramount importance of “what” to say to customers in your communications, however or whenever delivered, has too often been overlooked. This is because while the how and when are relatively easy to automate, figuring out exactly what to say to each individual customer — the right message or offer that will improve that specific customer’s experience and loyalty — is really hard. It’s a lot easier to send everyone, or everyone with a certain profile, or everyone with the same recent activity, a similar communication and call it a personalized experience.

Research with our clients shows that sending the right message motivates customers to buy far more effectively than how you reach them or when. That’s backed up by real campaigning. Fewer but more relevant messages consistently beat more frequent and general messaging. The message is everything. It’s just a matter of getting it right. The how and when of getting it to them is important, but it’s more important not to squander with an irrelevant message those increasingly rare moments when you might have the customer’s attention.

While profiling and segmenting customers are a step in the right direction, determining the best possible message or recommendation at the individual customer level requires a deep analysis of each customer over time. A customer’s actions at the present moment can provide indicators, but they are often misleading or unrepresentative of their true or deeper loyalty and interests.

Figuring out the best “what” to say is the next frontier in marketing technology. But to make it practical at scale, we have to do what was done for the “how” and “when.” We have to automate it.

It will not be as easy. Working with data is always hard. It has to be sourced, cleansed, integrated, normalized, and governed, and that’s all before it can be analyzed, mined, and modeled for predicting what each customer might want. Fortunately, there is a lot of innovation happening in this area.

One of the most interesting new approaches is the idea of packaging up specific analytic problems as easy to consume services. By using promising new data science that allows for limited source data and by constraining analytic objectives, such as focusing on product recommendations for each customer, the process can be automated. The idea is, send a well-defined set of data to a cloud service, for example, and get back customer specific answers on what incentives or products to offer. In this way the incredible power of truly personalized product recommendations and offers may soon become as efficient and affordable as sending messages has already become.

Peter Moloney is CEO of Loyalty Builders, whose Marketing Lift Service offers a simple, cloud-based predictive analytics service enabling marketers to get revenue lift from more relevant communications to their customers.

Tags :