How to Measure Advertisement ?

"Half the money I spend on advertising is wasted: the trouble is, I don't know which half".
a statement by John Wannamaker.


That statement is still true today for the most marketers, unfortunately.
Fortunately, it doesn't have to be true.


Thanks to gigantic leaps in data generation, collection, and processing, and to improvements in mathematical and research techniques, it is now possible to use research to get an accurate picture of most stages of the marketing process, including contributions to the bottom line.
Thanks to recent developments in massive database crunching, it is also possible to make behavioral projections with greater accuracy.
The capacity to measure the effect of advertising variables on sales has actually been available since the invention of the split-run copy test in direct mail. An ad in a newspaper or in a magazine is printed in two or more slightly different versions, each coded with a different return address (e.g. Box 40, Box 41, Box42) or a different phone number. Doing this at a sufficient scale (reaching a certain number of people) cancels out the effect of other variables, and you know which version of the ad pulled the most results, which variable was most effective. With new mathematical models, it's possible to test recipes (multiple variables) at the same time.
Amazingly, the vast majority of business do very little rigorous market research, They make decisions about price, location, product design, packaging, service brand strategy, message strategy, and advertising based on  history or gut feel, not any objectively measured reality. they could significantly reduce the risk of their marketing investment by testing the waters before they take a plunge. After all, you can't manage what you cant measure.


Be the smart marketer who uses research to stay in touch with reality.