How Retail Marketers Can Keep the Personalization Promise

Posted on October 12, 2016 by Peter Moloney

HandShake1

The “Personalization Promise” is a confusing, sometimes overblown assurance by a continually growing clump of marketing technologists struggling to differentiate their solution. The goal is to reach customers on their own individual, specific terms and do it well enough to make them buy. But while much of it sounds the same, there are actually big differences underneath that make a critical difference in terms of use cases, effectiveness, and ROI.

Doing it right isn’t easy.  You need to gain the ability to uniquely model and score every customer on loyalty, risk, and value. Then match every customer to the specific products they are most likely to buy in rank order or probability all at once, individually (not profiling or segmentation), no matter how many customers or products. And do it all without requiring a database of customer information. That’s very hard, but it’s also very doable with today’s technology.

Follow the Money

The key is to look at what they buy. In terms of a customer’s real interests and intent, there is no substitute for their actual purchasing behavior. Opinions, views, likes, and other behaviors can offer some insights, but the predictive value in terms of future purchases is suspect and short-term at best. On the other hand, a customer’s demonstrated purchasing patterns, compared to the inevitable patterns that occur within an entire customer base over several years, lead to more accurate and durable predictions even 90 days into the future.

If you make personalized product recommendations based on this deeper analysis of actual buying to one group of customers, and then compare their subsequent buying behavior to a different group that gets the same static offer – for one or more “best-selling” products, for example — you will invariably find that the first group purchased more. Lifting revenue in this way is easy to measure.

Start with a Roadmap

You still have to fight for the customer’s business, but you do not want to go into battle without a map. It’s more of a question of how timely and costly it is to analyze the data – because that can go on forever – and then figure out how to make effective use of it in your marketing activities. To keep costs from getting out of control, you should start with, “I’d like to implement marketing strategy ‘X.’ Now what do I need to know that would make ‘X’ more effective?”

Instead, too many companies start with, “Let’s start compiling and integrating all the data we think we can get, and then we’ll see what useful marketing strategy emerges from that.” For example, we know from experience that personalizing product recommendations in email campaigns lifts revenue. So what is the most cost-efficient way to figure out the right recommendations to make to each customer?”

Less is More

At a certain point, sending more emails to customers produces no further returns. It just increases the likelihood that all your emails are ignored. Fewer, but more relevant communications work much better.

It is a fact that that if you can send offers that are more relevant, timely, and useful to a customer, they will value your communication and respond more often. It’s impossible to get it right every time, but you can increase your odds.

Profiling customers into segments is a start, but unfortunately you miss a lot too. It’s hard to stereotype people accurately just because they have some similar attributes or behaviors. The best way to increase the odds is to create an analytic model around each individual customer. The cost of that was prohibitive, but the solution now is to make it automatic.

From Automate to Personalize

Ten years ago it was all about “automating” marketing processes. Now it’s about “personalizing” the experience. You cannot sell marketing tech without claiming to do one of those or increasingly, both. So the capabilities of dozens of systems sound similar, but the type and degree of automation and personalization is vastly different.

One breakthrough is the ability to analyze every customer individually and make buying predictions and recommendations for each one in a fully automated process that requires only transaction records.

The simplicity of the process and actionable usefulness of the results make for a compelling ROI. And since every marketing service provider and platform today has to prove their value with immediate and measurable ROI, this blend of automation and personalization is one proven way of keeping the personalization promise.

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. Request a Free Customer and Revenue Analysis here.

 

 

Tags :