Back to Listing

Marketing Mistake #3 | Not Testing Your Marketing Strategies.

05 May 2015

People can spend so much time developing the “ideal” marketing program or strategy, that their objectivity goes right out the window.

It’s important not to get blinded by how well you think your marketing efforts will pay off. Even with all the necessary research in place, nothing is certain. The real key is how the market responds to your efforts and offers.

Before you invest any significant amounts of time, money or effort in any marketing or promotional effort, it would be wise to test the effectiveness of your campaign. That way, if it doesn’t work the way you thought it would or should, you haven’t wasted unnecessary resources.

Here are a couple of examples to illustrate this point:

Let’s say that you want to do a direct mail campaign to 20,000 homes in a certain radius around your business. So, you rent the list of homeowners, print up the pieces, have someone stuff the envelopes, affix the postage and haul them down to the post office. Then you sit back and wait for the calls or orders to come in.

But after a couple of weeks, you look at your results and find that you have only received 19 responses. You quickly take out your calculator and figure up that you just pulled a whopping .00095 percent!

Doesn’t sound like a very good return on your investment, does it? Well, it’s not. But, sadly, that’s the way most businesses operate.

But now, let’s suppose that instead of all 20,000 pieces being identical, and mailed at the same time, you decided to only mail to 5,000 the first time. And, let’s say you came up with five different approaches or ideas.

Now you go to the printer, have 1,000 of each idea printed, and get them in the mail. After a couple of weeks, by monitoring the results, you notice that idea number three had the best response... 12 orders, or, .012 percent.

Not a great response, but far better than all the other letters combined. And, you’ve not only not spent a ton of money on your marketing campaign, but you haven’t tipped your hand to all your prospects with an ad that doesn’t work.

Now, let’s go back to the drawing board and fine-tune the letter that pulled the best response. Make the headline more attention-getting, the benefits the reader will gain more appealing, and the offer more attractive. Then, send it out to another 1,000 names.

This time, your letter pulls 47 responses. That’s a 4.7 percent response! Not bad in anyone’s book. And, you’ve only used 6,000 of your original 20,000 names. You still have 14,000 left.

You now have a “control” letter. By tweaking the letter again, you may find that it pulls even better. If it does, then this letter becomes the “control.”

Just because you rented 20,000 names, doesn’t mean you have to mail the same piece to all of them at the same time. That’s what most businesses do, and it’s why so many of them have given up on direct mail.

Once you have a letter that is pulling well, keep mailing in short runs, always trying to beat your control letter.

Do you see what has just happened? By monitoring the results and by testing your mailing you were able to turn a complete flop of a promotion into a huge success.

Your testing is never complete. Just because you’ve run a couple of different tests, doesn’t mean you stop testing. You should always be testing something different to see if you can beat your control.

For instance, suppose you’re doing a direct mailing to residences and want to know whether you should use an envelope with teaser copy, or a plain white envelope. The answer can be found by testing. Here’s one way you might do it:

Split your addresses into two groups; Group A and Group B. These can be arranged by post codes.

Group A will consist of addresses in certain codes and Group B will have addresses in different postcodes.

Group A will have teaser copy on the envelope, and Group B will be in plain white envelopes.

It’s important that the contents of all envelopes be exactly the same. If the test is going to be accurately measured, you can only have one variable, and that variable in this case, is the different envelopes.

Now, when your responses or orders come in, simply keep track of the post codes they came from and you’ll know very quickly which envelope pulls the best. (This is assuming, of course, that the various post codes consist of similar demographics.)

Once you’ve determined which delivery system is best, you can go on and test other components of your mailer.

Some things that are worthy of testing are…

•Super-heads (or pre-heads)
•Body copy
•Bullet points or statements
•The offer
•Return policy
•Whether or not to use graphics
•Which graphics pull best
•Pictures of products
•Pictures of people
•Charts and tables, etc.

If your marketing can’t be segregated by post codes, you can use specially coded coupons or response cards. Use different codes for different variables that you’re testing.

Or, you might use different telephone numbers. Some businesses have the respondent ask for a specific person or department.

For instance, if they ask for “Susan,” you know they’re responding to one particular offer, and if they ask for “Heather,” they’re inquiring about the same offer but with a different variable being tested.

Such persons don’t really have to exist; you just use the names as codes to measure the responses.

When the phone is answered and the caller asks for “Susan,” you simply say, “I’m sorry, Susan is not available right now. May I help you?”

You told the truth. Susan isn’t available. And, you’re able to help them. What’s more, you’ve identified which offer the caller is responding to.

Testing is such an important part of any successful marketing effort that nothing should be rolled out in large amounts without first testing.

Just using this simple strategy can help you maximise your returns, minimise your costs in time, money and effort, and add significantly to your bottom line.

Whatever you do, don’t fall in love with your project so much that you can’t be objective. And don’t spend too much time trying to analyse and figure out why a particular thing works and why another doesn’t.

It really doesn’t matter why. What matters is what. Put small test runs out into the marketplace and let your customers tell you what works and what doesn’t.

Thanks for reading, and if you want a chart that can help you keep track of the variables you’re testing email


Lisa Ormenyessy is a seasoned business coach who helps small business owners to gain leads easily without breaking the bank.