Google Optimize is an A/B testing platform offered by Google that allows you to create, run, and analyze A/B tests on your website without having to write code or add any additional components. The best part? It’s totally free and it provides you with actionable results in real time. Whether you’re new to A/B testing or you’ve been doing it for years, the following post will help you get started with Google Optimize and give you tips on how to get the most out of your tests.
Why Google Optimize
There are dozens of other web testing platforms out there, but not all of them offer mobile support and that’s where Google Optimize shines. It’s an easy-to-use platform for creating multivariate tests on your website or mobile app. It also offers multivariate testing and single-page app testing capabilities, as well as split URL and redirect detection—you can do it all from one place! More important: it’s free for small businesses (with a cap of 30,000 pageviews per month). And for larger companies? Well, let’s just say you won’t get charged a ridiculous monthly fee that makes it impossible to experiment.
The basics of A/B testing
There are numerous factors that can make or break your website, but one of them is testing. When you’re making significant changes to your website, it’s important to be confident in what you think will help it convert better. And if you don’t know what type of layout or message will help bring in more conversions, test away! One way to accomplish that is through an A/B test. This type of experiment allows you to see which version of something performs better and get actionable insights into how you can improve your site moving forward. Before diving into examples or any specifics about how web testing works, let’s talk about multivariate testing. What’s It?
How many visitors do you need for A/B testing?
Typically, you need at least 500 visitors per day for Google Optimize, and 10,000 total visitors over a 14-day period. If you don’t have that much traffic, consider a multivariate test instead. Multivariate testing uses one experiment to track many page variations at once. Using it effectively requires more design and analysis time on your part, but its scalability makes it worth investigating if your site has at least 20k monthly visitors. Google Optimize or Google Analytics?: While Google Analytics gives you insights into how people are using your website right now, Optimize lets you try out new ideas without losing any data. That said, while Analytics is easy to use (and free!), some of its functionality may be limited in scope compared to Optimize—so it’s best used for continuous A/B testing of specific pages or experiments with a very limited scope.
What is multivariate testing?
Multivariate testing is a type of website testing that allows you to simultaneously test changes on multiple elements on your page. Multivariate tests are sometimes called A/B tests or split tests, because you’re dividing up traffic for your webpage into two groups, which receive different pages. One group might receive one version of your homepage, while another gets a slightly modified version. With multivariate testing, you can test changes on individual elements—images and text being popular choices—but also changes across whole pages at once (e.g., different headlines, text sizes, designs). It’s no exaggeration to say that multivariate testing can be worth its weight in gold—if done right!
Real Life Example
One of these examples is Door changes which gave them a pretty big increase of 2.4% in new signups. By simply changing how many visitors saw text vs icon for signing up, they were able to significantly increase their conversions.
Can I use Google Optimize for free?
Yes. Google Optimize is free and it’s super easy to get started. As long as you have a Google account, you can use it right away. If you don’t have one, no worries – just set up an account and follow along! The first thing you need to do is configure your experiment. At that point, you simply copy and paste a piece of code into your website’s source code, add some simple configuration parameters like the variations and visitor limit (if any), then boom: You can begin testing different variations of content in real time. See? Easy.