We previously ran an article which showed how long sales pages work. Another previous case study talked about anti-A/B testing on how removing header from our blog increased signups by 60%. Now in this post, I will be talking about how removing content from a landing page and hence shortening it increased sales by 62%. So, what is the real truth? Does more content convert better or less content? The short and sweet answer to this question is that there is no eternal truth. You must A/B test your way to more conversions. Sometimes long copy works, sometimes short copy works.
A/B test case study
AssessmentDay is a leading supplier of practice aptitude tests to graduates and job seekers. They help candidates prepare for their assessment centres and psychometric tests by showing them what to expect. Based on recommendation by their affiliate partner WikiJob (who are also our customer; previously we had featured one case study from WikiJob on how they increased sales by 34%), AssessmentDay ran an A/B split test on their co-branded landing page. Traffic on the landing page was entirely referred traffic from an affiliate’s website. The affiliate’s traffic is all organic and they have banners and text links sending visitors to the landing page.
Cutting corners: variations designed for the landing page
Their original landing page had a screenshot section and a FAQ section. Here’s how it looked:
For variations, they tried out the concept of ‘less is more’. Sometimes people get bored by reading long sales pages and they just navigate away, so we stripped out various parts of the page altogether. Having said that, the original sales page didn’t exactly have lots of text in the first place. For relatively low-ticket price purchases we think people want to quickly see what it is they’re buying and then make a decision; they don’t want to read huge reams of a sales pitch.
A/B test results: 62% increase in sales
The variations we tried were simply removing each of these sections in turn and then removing both sections together, so there were three variations in total. In first variation, they removed the FAQs section. In second variation, they removed the screenshots section.
Split-testing is always surprising! That’s half the fun of it. The result was a 62% improvement by simply removing the FAQ section, or a 56% improvement by removing the screenshot section.
Drop in sales if you remove too much content
What happens if you remove too much content from the landing page? Out of curiosity AssessmentDay then went one step further and deleted both sections of the page but this proved too much and the conversion rate actually dropped by about 3%. Here’s how this variation looked like:
Their hypothesis is that people like to have enough information to help them see what the product is, but any more and they start to get confused or distracted.
Lessons from the A/B test
When we asked if there were any lessons that AssessmentDay learnt from this test, here’s what they said:
- Less is more, up to a point. It’s surprising how you can spend so much time deliberating over the content of a sales page when in fact that section of content actually puts people off and you’re better without it.
- Split testing wins out over gut feel almost every time.
They also had great things to say about the split testing tool they used (Visual Website Optimizer):
Visual Website Optimizer has been extremely valuable, not to mention fun to use. The service has paid for itself several times over.
This was an exciting test with unexpected results. A classic case of trying bold changes and finally getting a much deserved increase in conversions. If you have any comments or feedback on this case study, please let us know!