Does It Really Cost Too Much Or Are We Leaving Money On The Table?

Download to pdf

A number of years ago, I was at the county fair with a friend and, as we walked down the fairway, a “barker” came out and offered us one dollar for 50 cents. Immediately I gave him two quarters and accepted my shinny silver dollar but my friend refused. Why? She said, “I don’t want to spend the money.”

Unfortunately, that’s how too many people run their direct mail campaigns. And at some point in our careers, we’ll introduce a test idea that’s refused because it’s thought to cost too much. But does it?

For example, if a test package costs 36 percent more than the control, do we need to increase response by more than 36 percent to make it work?

Let’s do the math.

First, let’s make sure we have rollout cost.

A test mailing of 50,000 pieces is going to cost more than the same exact package when you rollout in larger volume. Get the rollout cost before you mail and use this number when evaluating results.

Now, let’s see if, as in our example, our test mailing costing 36 percent more than the control must increase results by more than the same amount to beat the control. Here’s a simple example of what I mean.

The control costs $279/M and is pulling a 1.5% response (15 responses per thousand pieces mailed) with a $25 average order.

The test package has a rollout cost of $379/M. ($100 or nearly 36% more than the control.)

Yet with a $25 average order, the test needs only four additional orders per thousand pieces mailed to match the control. This 2.1% response rate (21 orders per thousand pieces mailed – or four more than the control) is a difference of about 24% not the 36% difference in cost.

Discovering a new package beats the control by 24 percent will have a tremendous impact on the client’s bottom line. Along with the immediate increase to revenues, the new package gives us a whole new universe of names to mail to. Lists that once didn’t work suddenly become profit makers.

Yet if we didn’t recognize the difference between the test package’s higher cost and the number of additional responses it needed to beat the control, we would have left a lot of money on the table.

We have many other things to consider when developing a testing strategy and we’ll talk about those later. But for now, get your rollout cost, do the math and see what new opportunities you can find.


Tags: ,

Leave a Reply