3 Ways to Evaluate Your Design Using Data
Every designer faces hard to make design decisions throughout a project. Making the right call in these circumstances is what separates a great designer from a good one.
Yet, design decisions shouldn’t be always coming from the gut feeling of the designer. As gut feeling can result in unexpectedly good results from time to time, there is always another option if you are looking for a more stable option; using data.
Being able to present a solid reasoning of why one design is better than another would both help you sleep better at nights, and also make your team/manager/client to have better trust in your designs.
So here are 3 methods to get you started:
Conduct A/B tests
A/B tests are perfect for you if you are working on a project that already has active users which you can use as guinea pigs.
The key to a successful A/B test is to keep the scope small and to set a measurable goal before you even start running the test. First define what is a successful design, are you trying to make more people sign up for your product, then your measurable goal would be something like number of sign ups per new user. Or do you aim to keep your users longer in your website, set your goal as the average session time. You get the point.
When it comes to running your tests, make sure you do not have so many variables in place. You want to make sure that after running the test for a while you will know which of the design options you want to take. If you keep adding more tests into the mix, it might be difficult to actually figure out which of the tests is making your numbers go up.
So all this is assuming you are using some kind of analytics software to actually gather up this data from your users. If you don’t know where to begin with analytics, check out the free online courses by Google for its analytics tool.
Utilize heatmap software
I can already hear you, what if I don’t have active users yet? No worries, we’ve got you covered as well.
This one is I believe a more uncommon method, but when I saw it for the first time, it basically blew my mind. The idea is that we will evaluate two different designs with the help of a software. It is powered by machine learning and has been fed with thousands of eye tracking study data to simulate user behaviour as accurate as possible.
I was introduced to this technology by a Berlin startup called EyeQuant during a design meetup they were hosting. EyeQuant takes your design as an image file and then presents you with a heatmap and several other measures regarding how exciting or clean your design is. The results are pretty accurate and the heatmaps are a really good way to see if the correct things stand out in a page.
If you are into heatmaps, you can also try out some heatmap analytics software (e.g. Hotjar) to see heatmaps that show your actual users’ behaviour.
Compare and analyse with competitors
Let’s be honest to each other 99% of the time, your product won’t be the first of its kind. That’s why you need to admit that you need to check your competitors before going too far in your design process.
You don’t have to discover America again. There is a bunch of proved and accepted design decisions, which your competitors had already went through, tested with real users for you and succeeded. So good news, there is reliable data you can use when making your design decisions. You can investigate their solutions to similar problems and also can make use of the public feedback your competitors receive to understand how the users react to these design decisions.
Looking at your competitors would also give you a clear idea of your strengths and weaknesses. So you can easily figure out which aspects of the design you need to improve to make the users choose your product among others.
And lastly, none of your competitors have “the perfect solution” for every design problem you are facing. You need to be aware of the good and bad parts of each design and take the good parts as much as you can and harmonize it with your design and serve to your users.
There is lots of other methods to reach better design solutions which you can benefit from. Leave a comment below which other methods you use at work to improve your design decisions and to convince your colleagues.