Direct Mail Tip – A/B Testing

Post image for Direct Mail Tip – A/B Testing

by Tom Watson on June 11, 2010

Many small business owners use direct mail as a component of their overall customer recruitment strategy.

This is because direct mail can be focused like a laser on potential clients with a message crafted just for them. It is very effective to say the least.

But, it can be difficult to determine how effective the direct mail piece is without some way to measure it. What if the sales letter was worded a little differently? Would it create a better response (would you receive more calls)?

This is really worth finding out because if you could increase your response rate (how many responses you received divided by how many letters were mailed) you would increase your revenue at some point.

A small change in response rate creates a big difference!

If you could increase your response rate from just 1% to 1.5% that could be tremendous. Why? Lets create an example.

Lets assume your average customer invoice is $600.00 per month (excluding sales taxes). Lets further assume your profiting about $150.00 per month when all is said and done. This means that in this hypothetical example you make $1,800 per year off each customer ($150.00 X 12 months). So now we know what each new “average” customer is worth.

Now lets further assume that for each 200 sales letters you mail out you get two calls (1% response rate). For some sense of accuracy I will assume that not all new customer calls (that came from the sales letter) will result in a sale. You will simply not close the deal 100% of the time. So lets say you get 50% of the customer inquires to sign up for service (at the average of $600.00).

So this means that for each 200 letters mailed, two people will call you and you will get one to sign up for service (at a profit of $150.00 per month). So far so good! Now we know that you will earn $1,800 per year from each two hundred letters mailed on average.

Now this does assume you are using a good sales letter and you are following through with all the important stuff (ie: What Your Customers Are Really Buying).

Lets set this example in motion and expand it a little bit. If you make $1,800 for each 200 letters mailed, what happens when you mail 1,000? Well…being 5 X the amount you would expect 5 X the profit. This would be $9,000 for a 1% response rate.

But what about increasing the response rate that 1/2 point we mentioned earlier?

That 1/2 point would increase your profit by 50%. So instead of a $9,000 yearly profit you would have a $13,500 profit. Big difference huh? Sounds like it’s worth determining if you can squeeze that extra 1/2 point out of your direct mail piece.

A/B Testing is a tool to help you do just that.

This method is very simple and quite useful in determining whether or not “version A” works better than “version B”. What you do is create two versions of the same direct mail piece. Version A would have one special offer (something like “1 week free cleaning for all new accounts”). Version B special offer would be different (maybe “2 free weeks of cleaning”).

Keep in mind that this does not have to be such a minor tweak such as just changing the special offer. It could be two entirely different letters. The difference could be wording or maybe color (colored paper vs white paper). If you can think of it you can test it.

The next step is mailing letter A to one group and letter B to another. Now all you have to do is see which group responds better. This will require you to keep track of who received what letter. Simply put the names of each list into a database (or notebook) and cross reference the calls as they come in to see which offer they were sent.

To get the best results you will want to have your sample size be as large as possible. Mailing ten letters to group one and ten to group two isn’t going to provide much useful information. For a bare bones budget test I would start making my sample sizes at around 250 for each group. As you get some experience doubling (or tripling) that number would provide better data.

The testing of your ads should be a continual process.

At the end of the day, if your going to go through the trouble of sending out direct mail letters, sending the one that will make you the most money should be your priority.

{ 0 comments… add one now }

Leave a Comment

Previous post:

Next post: