Love your forms: Does size matter?
As part of the experimental approach, we run a lot of experiments. And many of these experiments are about forms. One piece of advice I can give all you form-users now: You have to love your forms for them to perform.
Background & Hypothesis
Trustpilot is an open review website where everybody can write reviews about their shopping experiences with online businesses. As with many other websites who use forms, many users start writing in our form but do not complete it. As more than 2,5 million users see this form every month, getting just a small performance improvement (more people completing the form) is important for us.
Lower the field height
From a single user interview, we learned that the user found it more compelling to complete the form if the task looked easy. We assumed her reason for saying this was that we (by mistake) had made the review text field smaller — it was lower in height. We could have spent countless interviews and internal discussions figuring out if this was true, but instead, we followed the experimental approach and ran a kill experiment.
The experiment
The experiment was quite simple: Change the height of the central text field to half the size/height.
As we have a lot of traffic on this form, we ran the test for less than a day and had around 90,000 users seeing one of the three variations. So did more people complete reviews when given the smaller text field?
Nope. Not at all. If you want to see what a disappointing result looks like, take a look at the graph above! Not statistically significant, and really just not significant at all — even though we ran 30,000 users through each of them. The different was less than 0,1% in completion between the variations. So we avoided long discussions, saved ourselves time, and killed the idea in less than 24 hours.
Read more about experiments in our Experiments publication.
