Track. Test. Implement. You Are A Scientist.
I’ve stated in the past that Internet marketing is a lot like being a scientist. This is because you try all sorts of stuff for people to easily purchase from you. While I know that a relationship founded on trust, cred and authority absolutely makes selling easier, still, finding new and better ways is a way of honing the Internet marketing craft.
Introducting split and multivariate testing – the two types of tests that help you catch attention, get more clicks and make conversions easier. In this post, I’ll describe how to implement these two tests. Check it out:
Split Testing gets you to test with two variants – variants A and B – and is arguably the easiest to track. Let’s use Facebook ads as an example.
You have FB Ad A and FB Ad B – both ads have the 4 variables, namely: Headline, Copy, Image and Bid. With Split testing, you get to test with one variable only. You can have the same Headline, Copy and Bid for one ad and test two different types of images.
From there, you can track which one works better for your ads. The advantage, as mentioned earlier, is that the it’s easier to track. In the case of Facebook ads, I reckon it’s much cheaper too.
Multivariate testing gets you to test with 3 or more variants. While it’s “harder” to track (by that, I mean, if you’re dealing with 26 or more ads, it’s going to take more than an hour to figure out what worked and what didn’t), it’s waaay more comprehensive AND gets you to find out what works in one fell swoop of a campaign.
One recommendation I have with multivariate testing is to group them into two – draw out which ad works best for you according to maybe clicks and actions and have two ads compete in one final showdown to determine which works the best.
In any case, whether you do a split test or multivariate test, make sure you set aside a reasonable amount of time for all variants. Time, is, after all, a very important factor in testing.
That’s about it for this testing, folks. Let me know what you think in the comments section below.