Its purpose is to get ever closer to the truth.

While certain people*** with an axe to grind may well criticise the scientific method, nobody with working grey matter denies that everything useful we know about the world (including the device you’re using to read this blog) is a gift of science.

Marketing is not a Science. Marketing isn’t concerned with understanding the truth about something, and has other objectives. However, there is a branch of marketing that is concerned with measurement, repeatability and falsification.

Direct marketing attempts to borrow those aspects of Science that help it get closer to the truth about certain things. In particular…

• Who is more likely to buy a particular product?
• Which ad headline delivers the most visitors to a website?
• Which offer produces the most sales?
• Which price produces the most sales?

And any other question a business owner might ask. Such testing is sometimes called split testing, or A/B testing. It works like this…

• Create 2 versions of something and put them in front of a statistically significant number of people (usually at least 500 unique individuals)
• Make sure you know specifically what you’re testing. For example, if you’re testing one headline against another, everything else must be identical
• It is possible to test more than one thing, but you must be able to show a statistically significant sample for each thing. For example, if you’re testing one headline against another, and one free gift against another in the same sales copy, you’re actually running 4 tests…

1. Headline A – Offer 1
2. Headline A – Offer 2
3. Headline B – Offer 1
4. Headline B – Offer 2

• Measurement is the key to finding out what works. In the above example, you’d need to know the number of sales for each of the 4 categories. You can then report on which of the 4 combinations delivered the most sales.

This approach to marketing gets close to Science. In theory, anybody could rerun the above test and will generate the same result as you. In practise, that probably won’t happen because there is always a random variable beyond our control – the audience. This is especially so for web marketing.

I might test 2 headlines and 2 offers on one of my websites. You might do the same on yours. As your site gets different visitors to my site, there’s no guarantee our results will match. And in fact, if I run the same headline/offer combination several times throughout the year, I’ll be reaching different people each time.

We can never really know whether one headline is always more powerful than another. What we can do, is get closer and closer to a powerful ad. Or a wildly successful offer. That’s still more reliable than somebody’s intuition, or their experience (i.e. their preconceptions).

* If my simplified definition bothers you, please don’t take the time to let me know.

** Yes, and other things too. But this article has a point to make about marketing.

*** People who fall into 2 camps: Those with a financial incentive (e.g. floggers of ‘alternative’ medicine), and those with a tenuous grip on reality.