Customer reviews are powerful when it comes to increasing trust. But how much can trust impact your bottom line? If you are selling goods online, setting up a split test is a good way to find out. Here’s how you get started with split testing using customer reviews.
Split tests – or A/B tests, as they are sometimes referred to – is a great way to test and optimise key parts of your website or online shop. The principle is pretty simple: You have an idea that a specific change will improve your site, so you set up a split test that simultaneously shows multiple versions of your site. When enough visitors have visited your site, the stats will tell whether you were right or wrong.
You can split test pretty much anything:
And many other variables that you think can be improved. This introduction has more inspiration on common things to test for. But it doesn’t mention a split test using a variation where customer reviews are shown for maximum trust.
A lot of companies are using the Trustbox to display customer reviews directly on their website. By split testing, some of them have found out that overall basket size increases or conversion rate improves as much as up to 40%. Please have a look at our case studies for more details.
First task is to be clear on what you want to test. The best way to do so is by looking at your website as if it were a laboratory – and you are a scientist. You describe a single statement (the scientist will call this a “hypothesis”) based on your expectations. It can be a sentence that starts with “if” like:
If I place a Trustbox in the sidebar in the check-out phase, I expect the percentage of people completing the sale will increase.
It’s not about being right or wrong when doing split tests. Rather, it’s about learning from others’ best practices and measuring the results of the changes you make to your website.
There are a lot of split test tools available on the market. If you are looking for a basic solution to get started, Google Analytics Content Experiments (formerly known as Google Website Optimizer) or Visual Website Optimizer could be a good choice. Both tools are easy to use and guide you through how to setup your first split test without needing to involve web developers or an IT department.
When testing a single hypothesis at the time, the above mentioned tools will guide you through the process. But you will also get estimates on how many visitors you need in order to get results you can rely on. Statistical significance, in mathematical terms. Depending on how many visitors you have, it can take anywhere between a few hours to several weeks before you can be sure about the results, so from here, all you can do is wait and see how much increased trust impacts key metrics like your conversion rate.
Using your customer reviews in a split test is fairly simple. However, we have come across some examples where the version without the Trustbox performed better. When we helped the company investigating why, we came across one or more of these pitfalls that you want to avoid:
If you are in doubt about what works the best for your online shop when it comes to using customer reviews, additional split tests can give you the facts.
We’re always interested in hearing from companies who have tried this out. Please share your split test experience and let us know if results have improved as expected.