Whether you’re a direct mail copywriter, an account executive, or a group manager, here are 5 things you’ll want to know when planning — or approving — a test mailing.
1. The mailing’s objective
For example, some mailers may be content with tweaking the control with hopes for a 10% – 15% lift in results. Others need breakthrough results.
When you mail millions of pieces per year, beating the control by a few percentage points — or even by half of 1% — can mean big money. But a smaller mailer may need breakthrough results to justify the test costs.
Before you spend a dime to produce a test mailing, everyone needs to agree on what results the mailing needs to achieve before it can be considered a success and how quickly you need to recover your test costs.
2. Does the proposed test have a chance of winning?
Rather than become stuck in a “make it cheaper” mode, test adding features and benefits to your package. More often than not, you’ll increase profits by adding features and benefits to the package rather than by taking them away.
But before you commit to any test mailing, get the roll-out cost of each package and calculate how much it must improve results to beat the control.
It doesn’t make sense to test a package that requires an unrealistic rate of response to beat the control — or one that can’t recover its test cost in a reasonable time. (See How to Use a Pre-Event Routine for More Profitable Direct Mail Testing.)
3. What makes a difference?
Before you can determine whether the test mailing has a chance of beating the control, you need to understand how particular elements of the mailing can affect results.
Your mailing has 5 major components that affect results. In order of their influence on results — from most to least influential — they are the list, offer, format, copy, and timing.
Consequently, when you need a major breakthrough, test the components that have the greatest influence on the mailing’s success — the lists and the offer.
When you can afford to test for incremental gains, testing a new envelope teaser, formats, copy leads, graphics, and package inserts can have a positive effect on results. But when you need a breakthrough, focus on your lists and your offer. Nothing will make a bigger difference in your mailing’s success.
4. When is the best time to test?
The best time to test is when you have a successful control. Waiting for the control to fatigue before you begin testing can destroy your cash flow. Test from strength. However successful your control is, and regardless of how long it’s been the control, it can be beaten.
5. How much to test?
As a rule of thumb, you’ll want to limit your total test volume to no more than 20% – 25% of the total mailing. This way, if the test fails, the returns from your control will help absorb the loss and preserve the operating budget.
But when deciding how many pieces to allocate for each test, keep in mind that the important thing isn’t how many packages you mail but how many responses you get back.
Your test volume should be no less than the number you need to produce statistically reliable results. Personally, I like to see 100 responses. But mathematicians say as few as 39 responses can be statistically valid with a 90% – 95% probability of accuracy.
You can use an online calculator to tell you how much to mail, but the concept is easy. Just take the control’s response rate and see how many pieces you need to mail in order to produce 100 responses. For example, with a 1% response rate, you would need to mail 10,000 pieces in order to produce 100 responses.
Testing is essential to profitable direct mail but test smart. Don’t take shortcuts. Do the math first. Test things that matter most — lists, offer, copy, and format. And don’t be afraid to test things that increase your package cost.
Test smart and have the discipline to act on the results.