Most of you are familiar with the six sigma programs that have raised quality standards for products and processes around the world. Six sigma programs are important and they work—when used under proper circumstances. Unfortunately, when devotion to quality is taken to extreme levels under the wrong circumstances, it leads to undue caution and lack of experimentation. So, should your Internet marketing program be six sigma? I’m here to argue it should be maybe one and a half sigma.
OK, I thought one and half sigma was kind of a catchy title—I don’t even understand the difference between six sigma and “one and a half” sigma mathematically. But I am throwing around “one and a half sigma” to make a point.
The point is that Internet marketing is about experimentation, not mindless devotion to quality. Making no mistakes in your marketing program is the wrong goal. Six sigma is an excellent way to deliver high quality with repeatable processes, such as manufacturing your product, but it is exactly the wrong goal in Internet marketing.
Here’s why. You know exactly what your products need to look like to be perfect replicas of the design. You can test them and make sure they are uniform, consistent, reliable—all those great things that products should be. But those are not the goals of Internet marketing.
Your goal in marketing is to be fresh, new, impactful, and most of all, effective. And the only way to test effectiveness is to see how many people actually buy from you. Whether you sell online or offline, you can count how many visitors to your Web site perform conversions. And you can see whether that rate is improving with each Internet marketing campaign.
And to continuously improve, you must experiment. You must make mistakes. You must set out knowing that you don’t know what a perfect campaign is and knowing that you’ll have to keep changing what you are doing to optimize your results. In short, you must do it wrong quickly, and then fix it.
The worst thing you can do is to try to avoid making mistakes. You’ll do something safe (which is also boring and ineffective). You’ll reach group consensus and you’ll test it and you’ll prove that it isn’t an enormously embarrassing error. You’ll prove it is acceptable. But, in truth, it will be wrong, because no one ever gets these campaigns right the first time out. They optimize their results only through ongoing trial and error. You shouldn’t be settling for acceptable.
So, instead, you need to experiment—to do it wrong quickly. You don’t set out to do it wrong, but you need to accept the fact that your first shot will be wrong and you need feedback to gradually home in on the target. So I am using the cheeky name of “one and a half sigma” to get that across. Just try something. And make sure you have the measurements in place to see whether it worked. Then change it and try again. You may not win any quality awards, but you might start selling something.
Improving your results based on the feedback of measurements is what six sigma is all about anyway. Just make sure you do it the right way so that you encourage experimentation rather than playing it safe.
