How do you measure the real uplift your advertising creates?

We all know that if we did nothing with advertising: no TV, radio, direct mail, or e-mail, there would still be a percentage of people that buy from us. It may be a small percentage, but there are some that would still buy because they know us from our location, from their friends, or for some other reason. Of course, it is a little more complicated than that because almost everyone does some kind of advertising. In addition, there are times we experiment or try new approaches.

So the real question becomes: When you try these different approaches, what is the real uplift it creates?

What is the actual increase in sales or profit generated as a result of your advertisement? That is the answer we really need to know.

There is a simple way to measure your success with e-mail and direct mail. Regardless of whether or not you use mass mailings, segmented lists, or decide to use data analytics to nail down your target list, you can do this simple test.

Take your e-mail or mailing list and pull out a minimum of 10% and use it as a control. That 10% should not receive the mail or e-mail advertisement you plan to send to the other 90%. The goal is to discover the exact ROI of the 90% you advertise this specific ad to and compare that number to the 10% control list that receives your normal communication message. For sales, measure the ROI over a 30-day period. For service, you will want to measure for 45-60 days depending on your offer.

“We would have sold or serviced these people anyways!” How many times have you made that comment?

Hundreds of factors can be measured in direct mail and e-mail campaigns. The list, headlines, offers, and creative can all be measured and need to be isolated to know the true impact each factor has. But, to analyze from an ROI standpoint whether or not the advertisement provided more sales and profit over and above what you normally would receive, start measuring against a control list.

Below is a study we did using this control method.