Split Testing and What to Split Test for Easy and Highest Impact
You have undoubtedly come across an article or post that talks about the advantages of split testing for web site conversions and success. I think every marketing or e-commerce centric blog has a post about split-testing. It’s a fantastic technique that can produce measurable results. But how many of us actually do it? I’m betting the percentage would be on the low side.. Yeah sure, we’ll make changes to our pages, especially for a page that isn’t performing up to expectations. But pro-active split testing is a whole different ball game (even perhaps on a page that IS performing well).
Consider your 5 best converting pages or products, and imagine being able to increase their sales and performance by 5–10% with just a few tweaks. And these wouldn’t be guesses, these would be measurable quantifiable effects from tweaks. If you could increase your income by 5–10% (perhaps more) a month, isn’t that worth a little bit of elbow grease to do some split-testing?
It doesn’t have to be difficult either. Google offers a free tool, Google’s Content Experiments in Google Analytics allows you to conduct split testing easily. It enables you to:
- Compare how different web pages perform using a random sample of your visitors
- Define what percentage of your visitors are included in the experiment
- Choose which objective you’d like to test
- Get updates by email about how your experiment is doing
Now, do you have to make 2 completely different pages to test? No, not at all. I’ll give you a couple tips on a few of the easiest things to tweak that will give you the most impact:
- Page headlines (which headlines attract the visitor to stay on the page).
- Product images and copy on the page (which will keep the visitor on the page).
- Button text (which text gets more people to click and buy or sign-up for a mailing list).
Thing about the series of actions you want your site visitor and potential customer to take. First thing that happens is that they land on the page and take in the headline. Are they sticking around or bailing out immediately? If you’re having an issue with people bouncing immediately, change around the headline a bit and split test to see the results.
Once they’ve taken in the headline, and they’re deciding to stick around, now they delve into the content and pictures on your page. Now, are they staying long enough? Are they taking in your content and your value proposition? Perhaps your content is too long. Maybe the bullets aren’t snappy enough. Perhaps it’ll just take a few different product pictures. Is your affiliate video helping enhance their post-lunch food coma? :) Make a change, split test, and see for yourself if the change encourages people to stay longer and get through to your value proposition.
Ok, so now you have your audience sticking around and taking in your whole page. Are they clicking where you want them to click? If not, then it’s time to tweak and test your offer and/or your call to action. Don’t be afraid to experiment either. Every situation and every page is different, and sometimes what the marketing guru’s prescribe isn’t always the right answer. I’ve seen situations where toning down the call to action button lead to more clicks and conversions. I’ve seen situations where lowering a banner or ad on a page led to more click throughs.
But, in any situation, you’ll never really know unless you try it and you test. Test, test, test, and see the results for yourself.