Raising Database Marketing's ROI
The database marketing audit is a step-by-step procedure that will ensure a profitable return on your investment in database marketing. It is based on years of observing how corporate cultures and the behavior of individual stakeholders can affect the implementation of database marketing strategies and tactics. The objectives of a database marketing audit are to make the process of implementation as painless as possible, to shorten the time required for implementation and, above all, to achieve the desired result with the smallest possible dollar investment.
Identify stakeholders. A necessary first step is to identify the people in the enterprise who stand to gain the most from the implementation of database marketing technologies and programs. This will include product managers, business analysts and channel managers. If you work for a company with a field sales organization, it is imperative that field representatives are included. Enlist champions and involve them in the upfront decision-making process.
It is also important that the information technology department be intimately involved from the beginning because these employees are probably already working on some parts of the solution set you seek to deploy. Almost every IT initiative in today's world of enterprise-class applications and customer relationship management has some elements that can contribute to the success of database marketing initiatives.
Flowchart your company's sales processes. Start by creating a detailed flowchart of the company's sales process, from identification of a prospect to closing the sale. It sounds simplistic, but sometimes a visual representation of all the interrelated work elements and information flows involved can be revealing.
Successful database marketing is designed to take advantage of the company's existing processes, the myriad ways that people communicate and get things done within the enterprise. Of course, if database marketing is new to the enterprise, the strategies and tactics employed almost surely will require some new processes. Just don't assume that because you have a better idea, people will change the way they work. Quite often, they won't. Instead, look for ways that existing processes can be used with minimum alteration when possible.
Define the objectives. What problems are you trying to solve? What opportunities do you want to take advantage of? Answering those questions as specifically as possible can define your objectives with clarity. Beginning with the end in mind will ensure your success in creating a database and correctly choosing which tools you will deploy to access it and use it.
Take inventory. No matter where you're starting, it's likely that you already have some experiences and existing tools that can contribute to launching your program. Some examples:
o Customer communications that are already working for you. Leverage the experience and make those communications a part of your initial deployment. This might include customer care calls or effective direct mail lead-generation mailings.
o Field sales and promotion techniques that can be replicated and extended. The idea is to find out what the best agents, branches, dealers or stores are doing, then use those strategies and tactics to raise every location to a minimum level of marketing effectiveness.
o Metrics that the company's managers are used to seeing. Improve those reports and build them into your program design.
Database design. More than any other part of the task, the data model - the data elements that will be made available in your database, data warehouse or data mart - will determine your success or failure. Some people advocate throwing everything possible into the database, but that approach has serious drawbacks.
Aside from the storage and processing implications, in my experience having hundreds or even thousands of data fields to choose from makes the analytical environment confusing and unwieldy. Rather than carrying massive amounts of high-granularity detail, the best place to start is to determine what data elements are really meaningful, then concentrate on ensuring that the information in those fields is as reliable and accurate as possible. Some companies talk about their multiterabyte data warehouses as though they were badges of honor, but what really matters is the usefulness of the data and how easily it can be accessed and used to drive business.
Initiate an extended dialogue with the keepers of the data in your organization to rationalize data inputs with output requirements. You'll need a complete understanding of incoming data file field definitions and integrity. Plan to add further information about your customers by appending data from external sources. Then develop a list of the data "voids" you discover - information you really need but can't get. Sometimes those information elements can be created by summarizing, consolidating or aggregating other data points that are available. For example, you might need to know how much a customer has spent over the lifetime of a relationship. That number might not be available but could be derived by totaling the customer's individual purchase or transaction records.
Define information needs. Never ask the stakeholders what kind of reports they want to see. Instead, get a list of the questions they want the answers to. Not only will this help you add specificity to the decision support and analytical requirements you'll need to satisfy, it also will help you define what data elements need to be included in your data model.
Understand technology's place. Don't assume that having the technology will make you a successful database marketer. The most you can hope to gain from your investment in technology is an increased ability to develop an understanding of customer behavior and the basic tools to implement communications intelligently. The part of the process that matters most - creating sales - requires a skill set that has little to do with database technology. Audit your organization's ability to deliver those skills as well.
Richard N. Tooker is senior vice president of database marketing at DMW Worldwide, Wayne, PA.