In a previous blog post, I introduced you to modeling. To refresh your memory, modeling is a process where you send the names of your existing customers to a huge database company like Epsilon, Experian, or Equifax. They match your names to the same names in their database, and then tell you everything they know about your people, including demographics, interests and hobbies, behavior, etc.
You can use that information to buy mailing lists of people with similar characteristics, and design email and direct mail campaigns and websites that will appeal to your target audience.
Today, I want to share some fascinating results I’ve seen, and more details on how you can use modeling for your next campaign.
A Modeling Study With Surprising Results
My first marketing job was working in the marketing department of a major seller of financial home study courses. We were always looking for ways to expand our mailing lists, but we wanted to do it intelligently by targeting our best prospects.
We decided to work with Equifax to build a model of our existing direct mail buyers. I sent Equifax 25,000 names of buyers who had purchased from us within the last 12 months. Equifax matched our 25,000 names with the names in their database and found 7,500 matches.
Equifax pulled out all the additional information they had on file for these overlapping names, revealing many important facts about our customers. We learned the average age and income of our buyers and how many of them liked to read books, ride bikes, hike, play basketball, and who owned a home, vacation home, boat, RV, etc.
Many characteristics about the course members didn’t surprise me. For example, it didn’t surprise me that many of them had investing experience. And it was no surprise that many of them had purchased self-improvement products.
But a few really interesting characteristics were also uncovered that I never would have guessed. One of them was that a large group of our customers had an interest in photography and developing their own photos. This was before anyone could easily print out photos on a home printer. In those days you needed your own dark room.
This made some sense to me since this would fit the category of “do-it-yourself ” … the very hands-on, do-it-yourself approach to making money we taught. Still, I never would have made the connection to photography buffs on my own.
I took this information to my list broker. Together, we identified a number of lists of prospects who had purchased photo developing supplies.
And what were the results? The results were good. Not great or outstanding, but good. We got enough of a response to make it worthwhile to mail the list a number of times.
The model we built with Equifax helped us discover more about our customers, and in turn helped us expand our mailing universe in a new area that we never would have thought of expanding into before.
Using Customer Modeling to Refine Your Marketing Efforts
If you have 20,000 people in your database, many of them may have some characteristics in common, but they are likely not a homogeneous group. Your buyers can be segmented into specific categories. And your modeling process can help you segment your list for more targeted campaigns to different groups of buyers.
For example, you might want to segment your buyers by age. Let’s say that modeling reveals the following breakdown of ages in your existing customers:
20-30 years of age 10%
30-40 years of age 25%
40-50 years of age 30%
50-60 years of age 25%
60 and older 10%
This tells you that 55% of your buyers are between 40 and 60 years of age. But a very respectable 35% of your buyers are under 40. These are two large groups of buyers that probably won’t respond the same way to the same message. It might be worth it to you to buy lists in both age ranges, and prepare two different types of promotional materials to each group. You would have to test to see what works best.
But let’s say your breakdown looks more like this:
20-30 years of age 5%
30-40 years of age 5%
40-50 years of age 20%
50-60 years of age 40%
60 and older 30%
For this breakdown of buyers, you know that targeting people under 40 may not be worth the effort unless you have a highly targeted piece explaining why young people should be interested in the product.
You might experiment with sending that piece to people between 40 and 50, which would bring in another 20% of your buyers. But your main focus should be on buyers over 50, and your sales piece should reflect that.
Other ways you might segment your buyers might be by income level, educational background, other interests . . . really, there’s no end to what you can do to identify your groups and then refine your message to match your prospects. Modeling will help you do that to the best effect.
Using Modeling to Make the Most of Carrier Routes
For some businesses, the critical factor for customer loyalty is location. If you have a restaurant or a dry cleaning business, your best prospects are the people who live within, say, a 30-mile radius of your store.
One of the best ways to reach your prospects is to send a sales piece to everyone who lives in a certain area. The Post Office designates all the names in a specific area as a Carrier Route. These are all the people that a mail carrier will deliver mail to.
But a 30-mile radius of your store is a pretty big area and may cover hundreds of Carrier Routes. You want to identify the ones that hold the best prospects. And as we know, that means identifying the specific Carrier Routes where your existing customers live.
In this case, your modeling procedure will involve mapping. Your data company will take the addresses of your customers, map them on a grid of the area, and identify which Carrier Routes have the greatest concentration of your customers.
Don’t be surprised if you find that these routes are located near streets that come directly to your store, making it easy for people to reach you. Modeling enables you to take advantage of these natural pockets of good prospects.
There’s a LOT of information that a data company can provide you about your customers… And what you find out may surprise you! But overall, all the that data leads to one thing… If you can get an accurate model of your existing customers, they will lead you to many others just like them!