Last year I posed the question “Who owns the customer experience in retail?” It was the kind of question that generated a lot of debate – arguments that are still happening inside retailers today.
But in my travels during this year’s spring conference season, it occurred to me that there is an even more fundamental question on the table: who owns the customer profile? Because it seems to me that the system side of this question could be just as important to answer as the organizational side.
It reminds me a lot of the days when retailers were trying to figure out distributed order management. This was in the 2005-2007 timeframe. Prior to that, distributed order management was a term more familiar to the supply chain people found in electronics manufacturing, where multiple items had to be sourced and assembled into a finished product. But as retailers came to understand that they needed customer order management, and they started to realize the inadequacies of the order management systems they had in house, they finally came to the conclusion that they needed distributed order management – but with much more of a customer focus than DOMs of the past.
However, to get to that conclusion was a painful journey. During that time, I worked through a lot of workshops with retailers where we absolutely had to go through these three questions: Why not merchandising order management? (not designed for customer orders) Why not POS order management? (um, that doesn’t really exist) Why not eCommerce order management? (because eCom can take the order, but then it doesn’t know what to do with it). Finally, they were ready to come to the conclusion that customer orders needed something else – something more – than what they already had.
I don’t know that retailers will come to the same conclusion with customers as they did with orders – that a new class of software was needed to fill their gap. But it’s clear that they are in the midst of looking around in house at what they have when it comes to customer data, and seeing that there are gaps.
Besides your traditional CRM solutions, there are a lot of applications out there that hold customer data of one kind or another – I covered some of the differences late in 2013, particularly between CRM and digital marketing platforms, or DMP’s. But there are even more solutions out there that hold, not just customer data, but customer profile information.
What is the difference? Customer data is just that – data you hold about customers. A customer profile, on the other hand, holds that data in a way where some insights are already being made, and those insights are as much a part of the information about customers as the data itself. Some of that may be semantics to data scientists, but when it comes to retailers’ goal of having one view of the customer, the semantics become important.
For example, every retailer has customer order history somewhere. That is customer data. But some retailers analyze that data to understand recency, frequency, and monetary value, commonly referred to as RFM. RFM is customer profile information. And retailers could take that even farther to understand their customers in terms of the minimum recency or frequency value to put a customer at risk of lapsing. If a customer hasn’t purchased from you in a month, are they at risk of lapsing? Six months? The answer depends on the retailer, but the degree of “at risk” that any consumer is – that should be part of their profile.
In addition to DMP’s and CRM’s, there seems to be a sudden resurgence in personalization software. Solution providers tend to come at it from one of two sides. On the one hand, from the side of email campaign management, where email content is personalized based on, you guessed it, a customer profile put together by the solution.
On the other hand, there are the search/recommendation solutions, which come at the problem from website behavior. They too build a customer profile, based on shopping behavior, and begin to personalize the site based on what cohorts of customers who also have exhibited that behavior did next. If you have looked at reviews of the last three products, then you are exhibiting a review-driven shopping approach, and the reviews might get prioritized higher on the page for the next product you view. Or, if you only shop the women’s section of the site, the product recommendations will filter out all the men’s results, because clearly you’re not interested in those.
In both cases, the personalization software builds an ever-evolving profile of the customer. It can hold on to customer profiles whether they are known or unknown, and can detect trends about unknown customers (anonymous browsers) that can be applied to other unknown customers. And when the customer is known, then even more interesting things can happen, because instead of subtle things like evolving the product recommendations a customer is shown, retailers can engage in specific conversations (dare I say “one-to-one”?).
The problem is, now there are an awful lot of solution types out there that provide customer profiles. There are CRM solutions, there are DMP solutions, there are ESP providers (email service providers), there are search/recommendation providers, which may or may not call themselves site personalization.
And there are a lot of retailers who are beginning to realize that they don’t need a data warehouse. They need a customer profile solution – one with a lot of data sources going in, and a lot of ways to reach out and connect those profiles to action systems like digital marketing platforms or email campaign tools. Or location-aware marketing through the retailer app on a customer’s phone. Or something else that hasn’t been invented yet.
It seems like retailers are in the middle of asking themselves all those why-not questions – Why can’t my ESP be my customer profile? Why can’t my site personalization solution be my customer profile? Why can’t I use my CRM system?
Unfortunately, unlike with DOM, the answer isn’t a simple, “Because” for each one of these things. The answer is more like “Well, yeah it could – mostly.” Unfortunately, there are a lot of wide gaps that live in that “mostly” space – and none of those gaps overlap well solution to solution.
So is a new solution needed in order to provide a single customer profile for retailers? I don’t know. And unfortunately, retailers are going to have to slog through a lot of choices before they’ll be able to figure it out. Bottom line: there’s still a lot of disruption ahead for marketing in retail, whether organizationally, or on the systems side.