Marketing isn’t an exact science. In a scientific experiment you can repeat it again and again and consistently get the same results. People, fortunately, aren’t that predictable and unchanging.
Fashions, technology, society and trends are constantly changing and it’s impossible to hit the nail on the head every time with marketing. However, there are some general rules which you should use to guide your marketing (many of which I’ve covered in previous posts and my overview of SMART marketing) and to perfect your marketing you need to do “split testing” or “A-B testing” as it sometimes called.
The approach is simple:
- You develop two pieces of promotion in the same media
- You send it to two different samples of your target market.
- You measure the return
- You send the most successful piece to the remainder of your target market
Let’s have a look at this in practice. Bob runs a training business and has the addresses if 1,000 local HR managers. He wants to do a direct mail campaign so using best practice and guidelines (such as using a headline, offer, deadline, a P.S. etc.) he writes two different letters (A and B).
He then takes a sample of 100 people from his database and sends the first 50 letter “A” and the second 50 letter “B”. He then measures the response rate he gets, say letter A gets 4 replies and B gets 6 replies. So he then sends the remaining 900 people in his database letter “B”. This is simple but as you can see if he had just written and sent out his first letter, A, he would have lost 1/3rd of his possible responses. In this example, he would have got 120 responses from letter B and only 80 from letter A.
It is worth noting here that as we have covered in other articles, you must have a way of measuring the response you get. If you don’t, you don’t know what is working and what’s not and you can’t do this type of split testing. Your promotion must ask people to respond in a specific way whether it is through visiting a web site page, contacting someone, ordering a free e-book etc.
The whole split testing process might seem like a lot if effort but it is surprising what difference a few changes (or even one word) can make.
I remember one Adwords campaign where two adverts were run for the same service to see which one would perform the best. The ads were identical in every way with the only difference being one used the word “fast” and the other “easy”. The ad containing “easy” got 5 times more response than the one using “fast”. No one could have predicted this massive difference and no one can write the optimum piece of promotion every time.
The only way to get get it right and get the highest response is to “split test” your main campaigns and promotions.
