Why you should be data-driven instead of opinion-driven.
Everyone wants a website with high conversion rates.
Everyone wants a website that is easy to navigate.
Everyone wants a website that looks like the 8th world wonder.
But Rome was not built in one day.
Slow but steadily, businesses are coming to understand that you do not build a complete product and then let users test it. Instead, you start small and test as soon as you can with the end user.
In this particular case, we are helping a client optimize different parts of the funnel. Focusing on one part of the funnel at a time. Here we focused on the conversion part. If you would like to read six tips to increase your website conversions click here.
Status quo: a conversion rate similar to the industry average. But lots of ideas for improvement of the website’s user experience and user interface.
We started with one product page. Redesigned it based on less is more. A clear CTA, well thought out contrast between background and text plus optimizing the user-friendliness.
The goal was to increase the conversion rate of the landing page by getting rid of unnecessary distractions and building an intuitively easy interface.
Then we used A/B testing to compare the old landing with the new landing. Via Facebook Advertising, we could exactly measure how well each landing page performed coming from that same advertisement. This works by building a conversion campaign with two advertisements. One advertisement linked to the old landing page and one advertisement that linked to the new landing page. Facebook then automatically divided the target audience into two groups that are shown one of the two advertisements.
Result: the old version outperformed the new version by a mile. Totally against our expectations.
Should you consider this a failure?
I think not.
Too often businesses make opinion-driven decisions. Instead, let the data speak. Rather than redoing the whole website, we tested one page first. It would have been a complete waste of our time and the client’s money.
Of course, the story does not end here. We now know what is happening. Time to find out why it is happening. This is the difference between hard and soft data. Basically, hard data is the numbers, the percentages and the ratios that give you an overview of what is happening on your website. Soft data is the questionnaires, the polls, the interviews and user tests that give you valuable insights about why users behave like that.
For example, we used usabilityhub.com to perform a ‘Preference test’. The testers are shown both versions of the landing page and then have to choose the one they intuitively like better.
From there the test proceeds and the participants are asked in-depth questions about the landing page of their choice.
Both hard and soft data are invaluable to create an amazing user experience.
It’s easy to generate ideas. Ideas are cheap. Execution is what counts. Execute, learn from it and improve with the next experiment.
“Continuous improvement is better than delayed perfection.” — Mark Twain
The lesson is to start small. Test new ideas first before fully implementing them with the possibility to waste resources. Follow the lean method. Build measure learn.
