Hello everybody it’s Brian Lewis here again to talk to you about your digital marketing. You know if you want to really make sure that your website is the most effective marketing tool it can be – you’re probably going to want to do testing. In other words after your site is redesigned you’re not done. And when it comes to testing, the most traditional way that marketers do testing is what’s called A/B split testing, it’s one of the ways how to run successful digital marketing testing on your website. It is where you send a portion of the traffic to an existing page and then a portion of the traffic to a test page. Then sit back and wait until you’ve reached what’s called statistically significant results and then look at the data and choose which one won. Now what can happen if you have a site that doesn’t get a lot of traffic, doesn’t have a lot of conversions, you may have to wait a long time and in that sense time can actually be your enemy.
And there’s a few reasons why you don’t want to wait a long time when you’re testing. One is the seasonality and what happens if you’re running your tests. It takes a couple months and then you run into let’s say the holiday season or some other season that impacts your business. That’s really going to invalidate your test. Another would be maybe you’re changing your product or service during the time that this test is running. The test runs long enough perhaps you’re changing your marketing tactics or maybe there’s a competitive change. So what happens when we have to spread that timeout there’s a lot of exhaustion, these are factors that could play into the test results.
So you know and actually the situation is worse too if you have a small traffic site because really the best way to test is the test by segments. In other words you’re your mobile traffic probably gonna behave differently than your desktop traffic and probably want to do a separate tests for that. Your visitors may actually behave differently on your site based on what traffic source they’re coming from. Whether they’re coming from Google organic or paid search or email, and so you want to segment that even more. So what happens if you have a site that’s got small traffic, small conversions, then you segment it out – now you’re really talking about tiny traffic and tiny conversions.
So what happens is it’s going to take a long time for your test results to be meaningful. Some rough guide lines, you probably don’t want your test to run more than about six weeks. Anything beyond six weeks you may start running some of these problems. So what is kind of a good benchmark to decide if you have enough traffic and conversions to do A/B split testing? Well think about or check your analytics and see do you have about a thousand visitors unique visitors per week and are you getting roughly 50 conversions per week for each segment of the test. If you have that much then A/B split testing is probably going to be a good solution for you. If you don’t then stay tuned because the next segment I’m going to be talking about are some alternatives. If you would like more information on how to run successful digital marketing testing, please feel free to contact us by clicking HERE.
