A/B Testing On A Budget: 4 Free Tools You Can Use
Something on your website is performing poorly. I can almost guarantee it. Unfortunately, I have no idea what it is. And unless you’re constantly doing A/B testing, neither do you. So let’s fix that.
A/B testing is the best way to learn what is working and what isn’t on your landing pages. That’s because it eliminates the hunches (“that logo looks ugly”), the old wives tales (“green means ‘go’ for submission buttons”), and the tradition based reasoning (“always put your sidebar on the right”), and replaces them with cold hard numbers.
Unfortunately, A/B testing has traditionally been very expensive to implement. But, if you’re the DIY type, you can conduct A/B tests for free using some of the four tools outlined in this article:
5 Second Test
The easiest free option is Five Second Test. It allows you to optimize buttons, calls to action, and landing page elements by setting up a quick five-second test for random strangers to take. You can present to testers things like; two alternative submission buttons, or two landing page images, and ask which users prefer.
There is a paid plan, however, you can receive tests entirely for free simply by taking the tests of other designers. For each test that you complete, another Webmaster will take your test.
This isn’t an effective way to gain thousands of data points, but if you’re trying to quickly answer the question of which photo is more appealing on your landing page, it’s a great free tool that requires little setup time and no programming ability.
Google Analytics Experiments
If you’re currently using Google Analytics successfully, you already have a robust A/B testing tool at your fingertips, Google’s Analytics Experiments.
Analytics Experiments differs from traditional multivariate testing programs in that it doesn’t allow you to change just one element of the page, but instead allows you to test up to 10 full versions of a single page by creating alternative URLs that rotate through. This feature makes Analytics Experiments ideal for the early stages of a website designing process when you have disparate ideas about how your landing page should look. Long term, however, there’s a danger of misattributing what is driving the increased conversions, so after picking a winning initial layout, you should test only one element at a time.
The biggest benefits to Google Analytics Experiments are that it’s free and relatively easy to setup with only a basic understanding of HTML.
The leading optimization software, Optimizely offers a full suite of A/B and multivariate testing options that extend beyond a single Webmaster’s needs. Despite it’s extensive capabilities, however, the free starter package, available to websites with under 50,000 visitors per month, is easy to use. It requires only a single line of code and required no technical expertise.
Additionally, it offers some features that Google Analytics Experiments does not, like allowing you to segment users by source, browser, and mobile v. desktop so that you can see how different segments of your traffic respond to your design changes.
If you’ve got decent developer chops and want something more robust and customizable, then A/Bingo is a good free open source option. Deployed as a Ruby on Rails plugin, it can be implemented via just a single line of code, it has minimal impact on your server load, and it can be modified to suit your particular needs.
Because it’s not packaged quite as cleanly as Optimizely or Analytics Experiments, it’s not ideal for someone with very little programming experience. For those that do, however, A/Bingo provides additional flexibility in defining conversions, meaning it can track things as unique as whether a player reaches level 5 in your online game, or how many points a user accumulates. Best of all, A/Bingo is free.
Continually optimizing your website and landing page for conversions via A/B testing is one of the easiest ways to improve your website’s revenue. And by using the free tools described in this article, it can also be one of the most profitable things you can do, because the costs are simply your setup time.