In this multichannel age, direct mail is still the largest single power in direct marketing. The U.S. Postal Service handled 102.5 billion pieces of standard mail (direct mail advertising) in fiscal 2006, nearly half the year's total mail volume.
Direct mail is a great medium. Measurable, scalable, highly targetable, quickly adaptable to changing environments and opportunities, it's hard to beat for marketing success.
But that success hinges on how well we mailers practice our craft, especially the basics. Here are three rules of data handling that will improve your chances:
-
Use data hygiene and prep procedures to eliminate sheer waste.
-
Use house data to help segment customer files and focus marketing dollars on the best customers.
-
Couple house data with external data sources to profile and model, and drive prospect targeting.
MINIMIZING WASTE
List hygiene and solid data prep are the foundation of success. They aren't sexy, but get them wrong and you'll pay. For example, the latest U.S. census figures indicate there are more than 301 million Americans. A recent relocation study by Mayflower Transit projects that 43 million of us move every year. That's one in every seven people, or 14% of the population.
If you fail to update those addresses regularly, some 14% of your house file will become undeliverable every year. By the end of three years a file of 100,000 customers diminishes functionally to just shy of 64,000. If your mail package costs 46 cents to print, process and mail, your oversight made it $16,560 harder for the campaign to be effective.
Compound that with other issues: a small percentage of your customers die every year (usually between 1% and 3%); apartment numbers aren't correct; you send mail to people who don't want mail solicitations. These can add up to pack a negative wallop.
Here are some simple things you should be doing:
-
Standardize data and consolidate it in a master file. Keep this file as current as you can — that is, update records as you learn of changes.
-
Clean your list twice a year. Run it through the National Change of Address process; a deceased filter; a local address correction service that converts rural route numbers to street addresses; and Apartment Plus, which appends correct apartment numbers to accurate street addresses. There are other hygiene procedures available, but these are the essentials.
-
Bounce your file against in-house suppression requests and the Direct Marketing Association's suppression file. Respect your customers' wishes regarding mail.
-
CASS-certify your list (be sure to dedupe), then automate and presort it. You'll eliminate wasted mail and pay the lowest possible postage.
DATABASE SEGMENTATION
The next step? Don't mail every offer to every customer! Segment your database so you can focus on the best customers.
There are many ways to do this, but RFM (recency, frequency, monetary) analysis is one of the old workhorses and where you should begin if you haven't segmented much of your customer file.
Depending on how complex your data is and what your objectives are, here's a simple start: First, convert your information so that it can be arranged numerically. Transaction dates would range from the most recent to the oldest. The number of transactions can be totaled and listed from most to fewest. Transaction amounts could be added together and run from largest to smallest.
Array each of the three types of data and divide them into quintiles (blocks of 20%). Customers in the top 20% of spending would be the top quintile and each would be coded with a “1” for “monetary.” These are the people who've spent the most with you. If the top 20% of customers ranked by total spending is coded “1,” then the next 20% of customers (or the second quintile) would be given a “2” and so forth down to the bottom quintile (oldest transaction dates, lowest dollar totals, etc.), which would be assigned a “5.”
You'd do this three times: once for recency by listing dates of the latest transactions, once for frequency by noting the total number of purchases, and once for monetary by indicating customers' total spending.
Next, multiply the three codes in your data set to come up with an RFM index. Notice from the chart on this page how easy it is to quantify customers' relative value with a simple RFM index. The most valuable customers have the lowest score. The best possible score is a 1 (1 × 1 × 1), the worst 125 (5 × 5 × 5).
You can use the index in different ways. For example, to focus resources on your best customers, stop spending money on the bottom 20%. Electronics retailer Best Buy landed on the front page of The Wall Street Journal about a year ago when it made this audacious move. It cut all spending on that bottom tier because every mailing to it lost money.
This is only one way to use RFM, and RFM indexing is just one segmentation tool. The key is to start thinking about your house file as a collection of groups and not a single group.
BOB MASSIE is CEO of DM firm Marketing Informatics, Indianapolis.
| Customer | |||
|---|---|---|---|
| A | B | C | |
| Recency | 1 | 5 | 2 |
| Frequency | 2 | 4 | 4 |
| Monetary | 1 | 2 | 3 |
| RFM Index | 2 | 40 | 24 |
