|
THERE IS NO SUCH THING as an average customer. Every customer is different from every other, at least in some important way. |
For example, a company's average customer may earn $50K/year -- but there may easily be a group of customers who earn more than $90K/year and another group of customers who earn less than $25K/year. These two groups are quite different from each other -- and they are both probably different from those customers who actually have a $50K average income.
At the same time, all of your buyers have some things in common. For one thing, they have all spent money with your organization. They are certainly more alike than different in other important ways, too. Gaining a clear understanding of both the similarities and the differences among your buyers is essential to making better sales and marketing decisions.
WHAT IS PERFORMANCE ROFILING?
Profiling is the first step in gaining new insights about your customers. It is a series of simple data-driven ways to describe -- or "profile" -- your customers. Profiling does not suggest anything about any cause/effect relationships. Neither does it predict any future behaviors. It simply provides you with a clearer picture of what your customers "look like" in terms of one or two dimensions (variables) vs. one or two others. Usually, the best way to discover new insights is to prepare a series of simple graphs or charts (histograms) like the one below.
Performance Profiling starts with simple graphs like this one that use customer and sales data that you already have in your marketing database. For example, fresh insights often come from displaying counts of your buyer segments as compared to sales to those buyer segments. Then, compare counts and sales by segment to:
- Where are they located (state or sales region)?
- How long they have been buying from you?
- How much they have bought from you?
- How often they have bought from you?
- How many different kinds of things they have bought from you?
WHAT IS DEMOGRAPHIC PROFILING?
Often, dditional demographic information about your customers can be obtained from a "list compiler." Consumer-oriented list compilers collect, compare and consolidate demograhic data from census tract data, credit reports, public records of many types (including mortgage information) and so forth. Such data include age, size of household, presence of children of various ages, number and types of cars, trucks, motorcycles, several kinds of wealth indicators, ownership of residence, religious affiliation and sometimes hobby and reading interests.
List compilers can also often provide certain kinds of information about companies, too. These data are assembled from public documents, industry reports and voluntary disclosures. Such information usually includes annual revenues, company size, corporate ownership, SIC group and so forth.
Combining such demographic or firmagraphic data with your own company's data is called "overlaying" and can often reveal quite a lot about your customers. However, overlaying data requires that your customer files be clean and accurate (See Data Hygiene - Database Bullets to Dodge). Even then, only a portion of your customer list is likely to match up to the data available from the list compiler. And, these kinds of overlay data are also relatively expensive. Nevertheless, they be quite valuable in helping to understand your customers. Here is an example of what one company found when they compared the ages of their customers to some national averages:
HOW DO YOU USE PROFILING?
It is best to display only two or three kinds of data at a time because the graphs can quickly become confusing. It usually helps to print out each chart or graph on a single sheet of paper. Print it in color if you can. As you look at each one, jot down any questions, conclusions or ideas that the graph suggests. For example, look for "sanity". Does the chart confirm or challenge your assumptions or the "conventional wisdom" about your customers or patrons? Are there any spikes or dips that you cannot easily explain? Are there any categories that have something that surprises you? Is there any category or group that appears to be missing? Your answers may lead you to new questions that provide real insights and discoveries about your customers.
WHAT'S THE BOTTOM LINE?
Profiling is easy to do and is always the first place to start with any data-driven marketing effort. It helps spot data problems that must be fixed before they derail more complex kinds of analyses. Profiling often uncovers a lot about your customers that had been masked by thinking in terms of an "average customer".
However, this simple technique is limited to looking at customers in just a few ways (dimensions) at a time. And, it's rarely possible to predict which cross-tabulations or profiles will be useful and which will not. So, this technique can take some time -- and it reveals only the most obvious relationships. Still, it's easy to do and is especially good for spotting "low-hanging fruit" marketing and sales insights that you can use quickly and easily to boost sales and profits.
Questions? Just ask! [Back to How-to Resources] [Top] |