When it comes to websites, how they should work, what they should look like; pretty much everyone has an opinion.
“What colour should the buttons be?”
“Should the navigation menu be vertical or horizontal”
And so how do you decide what is best? Maybe the web developer will decide, maybe decisions are made by committee or simply made by the HiPPO (Highest Paid Person’s Opinion) or the loudest voice. But decisions based on opinion or even on what we think we know, are probably wrong.
An alternative is to study the evidence. There have been a number research studies to determine people’s favourite colours. Here are the results from a colour survey.
Obviously, colour is a is a personal choice but when 42% of the people asked say blue is their favourite, then at least you know it’s a safe bet. Do you think it’s a coincidence that the Facebook, Twitter and LinkedIn logos are all blue?
How visitors view your website can be seen by data gained by eye tracking research. If you could determine where on your webpage your visitors’ eyes will be attracted to, would you not design or redesign accordingly? By analysing a huge amount of eye tracking data, the clever folk at Eyequant predict where on your web page a user’s eyes will be drawn. And with just a few seconds to grab their attention, you do want to be sure that they see what you want them to see. This is especially true if you’re using paid for marketing to deliver visitors to your site.
Will you increase sales by changing your ‘buy now’ from green to red? Red will instill urgency in the user but on the other hand it also symbolizes ‘danger’ and ‘stop’. Again, you’re quite likely to receive plenty of opinion regarding what will work best.
The right option is the one that works and the only way to know that is to let the two compete. This approach is known at A/B testing which simply means that you test option A and option B under controlled conditions to determine which performs better. You will want to do this with enough visitors to make the results statistically meaningful, i.e. think 100s not 10s of tests. A simple, low tech option would be run the webpage with a red button for a week and then for another week with the green button. You’ll need an analytical tool such as Google Analytics in place to monitor the results. Remember, we’re not counting the number of visitors to the page, we’re counting the percentage of visitors that hit the green button compared the percentage hitting the red button. Although it must be said that running A/B tests simultaneously alongside each other is a much better test.
For a higher tech option to A/B testing and for a few dollars a month, you could do worse that use a tool like Optimizely. What I love about this service is that you can create alternative pages without needing technical expertise to keep changing your webpages. Perfect for the small business! It allows you to create multiple versions of a page and then effectively rotates them while keeping track of the results of each page.
And so it’s time to stop the opinionated debate in the office, let the actual results set you free!
If you use any other tools and techniques to discover what works best, please share in the comments.