The Information Age has given us a wealth of consumer data. Your job as a marketer is to find the nuggets in that data that represent profitable business. By following four simple rules of data mining, you can make your lists or internal database produce gold.
1. The closer the data is to the consumer, the more recent it is and the more that it is composed of known data elements, the better the quality of the “ore.”
First, you need to ensure that the ore, or in this case data, will yield some nuggets. When you are out shopping for data to enhance your customer database or for a prospect list, ensure that your list or data supplier addresses these key questions to your satisfaction:
* How was this data collected?
* How old is the data?
* Is the data updated regularly and consistently?
* Has any data about individual consumers been appended from other data files? If so, ensure that you understand the nature of all of these sources of data in the mix and how they affect the balance within the database — for now and for the future.
You should look for a data source that has been created using actual or known information about the consumer, usually transactional behaviors or self-reported information. People move and habits change, so the names on your list should be no more than two years old and ideally just a few months old. As mentioned earlier, if data has been appended to a file make sure that you know the origin of the appended information. If some of the data elements are not “known” but have been inferred, you need to understand which elements are inferred and the probability that they are correct. All the analysis in the world on data that is suspect or fundamentally flawed will not provide the desired results.
2. The more you know about particular consumers and the more relevant that data is to your purchase situation, the better you will be able to decide if they are appropriate targets for your product or service.
Assuming that you have the right ore, examine how much relevant information you have in the file about each consumer. Traditionally, marketers have used demographic data, such as household income and ZIP codes, to target potential customers. But just because certain people live in an affluent neighborhood, doesn't mean they are interested in buying a swimming pool or central air conditioning. On the other hand, if you know that a consumer has or intends to buy a home computer in the next six months and you are in the computer accessories business, you've got a prospect. Your ore should have the most relevant information available about a household to allow you to select the consumers most likely to be your customers.
3. One of the best predictors of future behavior is past behavior; make sure your prospects are inclined to respond to direct-to-consumer offers.
Having an understanding of who your current best customers are and what characteristics they share will help determine what behaviors to look for in prospecting.
Making a list work for you requires understanding who is most likely to buy your product and finding those consumers in your ore. Identifying that a consumer is “in the market” is only part of the equation. You also need to know whether consumers will respond to direct-to-consumer approaches. Consumers who have made purchases by mail or over the Internet are most likely to do so again. Even a consumer who is interested in your product may not be inclined to purchase it directly from you.
4. Data mining is a team activity; make sure your team understands your business and is working with you to produce real results, not just selling you its products.
For most of us, our college statistics courses were the ones we endured because they were compulsory. That is why understanding the inferences or conclusions that can be drawn from data is something best left to professionals. But that doesn't mean that these people should be relegated to the back room to churn out numbers.
Think of the statistician as the refinery engineer whose job it is to get the most gold out of the ore. As a business manager, you need the engineer to understand your needs, to work with you in developing the performance expectations from your data. If the profile of the consumer and the goals are well understood and articulated to the statistician, you can get better results and save money by reaching only the right consumers with the right offer.
Statisticians use a variety of tools and techniques to predict the best prospects in a database. Their training and experience help them to test hypotheses to ensure that the prospects you approach represent your best chances of success. Your involvement with the statisticians in the early stages helps them determine the best process for refining your ore.
You need to have confidence in your mining team and its ability to deliver results. Make sure your team members can explain what they've done to find your prospects and why the prospects they put forward should be profitable for you. Have your team explain the refined data in terms of its similarities and differences to your current consumers or the characteristics that you feel make the best prospects. Can they explain the refined data in terms of the success criteria you've set? If you feel that you are being sold an off-the-shelf solution or that the team doesn't fully understand your business issues, move on. This is a partnership, and it should feel like one.
Turning information into consumer intelligence and business-building results takes teamwork and discipline. Your list needs to connect you with your best target consumers and meet your campaign goals.
These four rules will help you to ensure your data mining activity gives your business the best chance to accomplish profitable growth.
