How to Prepare a Successful Integrated Digital Marketing Program, Part 2

In this three-part series, we explore what an integrated digital marketing program is, why having one is important and how to make it successful using four foundational tools. In part one, we discussed the basics around an integrated digital marketing program. In today's installment, we discuss the first two foundational tools required for a successful integrated digital marketing program: the ability to recognize customers across channels and smart data. Part three, which will appear in the July 31 edition of eM+C Daily, will discuss the ability to personalize communications as well as how to correctly do attribution, the last two foundational tools.

Customer Recognition
There are many ways to recognize a customer: their email address, phone number, physical address, credit card number, name, Twitter handle, even a cookie. Since companies generally treat their marketing channels as separate entities, however, they're not gaining a full understanding of the consumer’s overall behavior and motivation, as well as what marketing messages they’re likely to relate and respond to. Channeled marketing yields channeled results. With no way of linking these results to individual consumers, they provide a one-dimensional view of the consumer at best.

This is more than just a question of a technological solution, however. It’s a core data issue. It’s not enough to integrate your systems or implement a central hub for your data. According to Acxiom, at least 2 percent of your customer records are made irrelevant each month. That’s nearly a fourth of the records, which you rely on to recognize customers, changing each year. Bringing in a partner like Acxiom or Epsilon is a good way to start laying your foundation.

Smart Data
By now, marketers have gotten the message that better data makes for more impactful communications. However, "smart data" doesn't define itself. What makes some data valuable to marketers and others less so? Here are some guidelines to follow:

Less is sometimes more. In theory, a good analytics practitioner can take a huge, undefined mass of consumer data and find the salient points. However, the "data by the ton" approach has several immediate pitfalls. For one, processing large amounts of data takes valuable time. Secondly, sometimes data remains unavailable due to technology, security or, more often, process barriers. Lastly, the larger a batch of data is, the greater the chance that inaccurate data is hiding in it.

Focus instead on relatively small, relevant data sets. Part of the danger of large data sets lies in the presence of irrelevant data. For retailers, for example, purchase data generally represents the most valuable data because it represents what people actually bought. Similarly, publishers use browsing behavior to determine what content may interest visitors.

Unfortunately, not all marketers have purchase or browsing behavior data at their disposal. What then? Simply put, you need to ask yourself, "which data can best predict behavior or interest?"

If a clothing retailer didn't have purchase data, for example, it could use gender to determine which items to feature on its website when a cookied user visits. On the other hand, a hardware retailer might not have much use for gender data, but would benefit from knowing whether the visitor lived in a house or an apartment.

Always test results. While data conveys a sense of mathematical certainty, it's only that — a sense. Test various data sets against one another and against a control. From personal experience, we can think of several smart marketers who learned that all the data they had couldn't do any better than the static offers in their emails.

Just because you can collect a piece of information doesn’t mean you should. As discussed, not all data is actionable; that which isn’t is nothing more than overhead. Data can be labeled “smart” if it enables you to build strategies around customers and segments, and if it gives you insight into what a person might — or might not — be interested in at a particular point in time. It’s data that makes engaging with your company more relevant and satisfying for your customers.

Chris Marriott is an experienced digital marketing executive, speaker, consultant and contributing writer. He also serves as an independent director on the board of eDataSource, a marketing services company that collects, analyzes, organizes and archives marketing messages, providing competitive intelligence and analytics to the email marketing community. Chris can be reached at csmarriott1@aol.com or @CSMarriott. Ben Rothfeld is the founder of Translinear Marketing Strategy, a marketing consultancy that focuses on driving prospects and customers from awareness to repeat purchase and advocacy. Ben can be reached at plannerben@gmail.com.