There is probably someone on your marketing team who is always going on about their great idea for your landing pages. They make comments like “a picture of a dog would be better" or "a blue button would work better than that red one”. These are the sorts of comments that have formed part of campaign planning discussions in direct marketing for years. But when it comes to online marketing, there has never been a better time to prove that you know better!
Using any standard adserving tool it is possible to track activity from media click through to confirmation page. In which case, you can see the effect on results when you set the landing page live that has the “blue button” and compare it to the version with the “red button” (once a robust testing period has passed). You could even run the tests in parallel, by directing a share of your online activity at each page, and you’ll very quickly be in a position to judge the best converting page. What a great way to show the know-it-all that they were wrong!
Before you go ahead with this sort of bet (or test as we should call it) you must make sure you follow a few simple rules:
Keep a control test running. This is a standard direct marketing principal that is extremely valid for website optimisation. Ideally you should keep some of your online activity pointing to the original version of the landing page until you are confident in your results. If you don’t do this, there is always the possibility that the uplift in conversion would have happened anyway, due to other variable outside of your control.
Don’t make too many changes. If you are testing the effect of amending the button colour from blue to red, don’t change all the other elements on the page as well! Try various combinations, and use a multivariate testing tool if you get to an advanced stage, but keep each test as simple as you can.
Make sure you keep the test active for long enough. There is no set minimum time frame or volume of traffic for a test and it will depend on the size of your online marketing activity. However, as a marketing decision maker you should be well placed to judge this. It is common sense that you’ll need more than a handful of visits to be able to run a fair test.
Only use optimisation tools when you're sure you need to. There are some fantastic A/B and multivariate testing tools available; Google’s Website Optimiser being the most commonly used now, largely because it is free. However, you can make some significant improvements in landing page conversion without these tools and it can be tempting to overcomplicate things when you have them.
These tips are just a starting point. If you want to be sure that you are making the right testing decisions and optimising your landing page effectively then you could always get in touch with the conversion experts here at Equi=Media.