A good landing page is hard to come by. Despite the vast quantities of data generated by landing pages – and their very public analyses – few designers can nail a good landing page design straight away. For example, did you know you’re statistically more likely to see a conversion if you have three entry fields instead of one? In this article, we’ll explore some good landing page optimisation practices, and some services you can use to help tweak that conversion rate upwards.
- 1. Read the stats, but take them with a pinch of salt
Landing page optimisation enthusiasts and marketing gurus are full of all sorts of important-sounding facts. Often bulked up by numbers (why write ‘seven times’ when ‘7X’ grabs the attention so much better?), these statistical ‘analyses’ can, from time-to-time, reveal interesting user behaviours. However, quite a lot of their conclusions are unwarranted or unqualified – don’t forget, it’s a marketer’s job to get a user reading their copy and converting to their service through the desired call to action, not to produce insightful and thorough user psychology research.
Large-scale studies such as those available for browsing on academic publication or research group websites are generally your best bet for sourcing trustworthy, applicable material. Sources that fail to present their statistics in anything other than a barebones fashion may be easily digestible, but their content is mostly hot air; these sources provoke questions like the following: what was the sample size? Can we have examples of the pages? How did you control for confounding variables?
- 2. A/B test like mad
A/B testing has been much simplified of late, thanks to the introduction of dedicated A/B testing services such as Optimizely: there’s no excuse to avoid A/B, perhaps even A/B/C, testing of page designs. It’s not just for completed projects, either: Optimizely (for example) makes use of an in-browser page editor that allows you to resize and reposition elements live on the page. Then, you can instantly start receiving data from your visitors (including conversion rates and hotspot analysis). A/B testing should form part of any web design or development process, especially if you’re targeting mobile devices. The kind of design that’s suited to your customers may well be the kind of design that’s suited to their device – and, depending on your target demographic, that device will vary from person to person.
You can also configure A/B testing services like Optimizely to serve different pages to users depending on their device type (an extension of CSS media queries, with analytics tagged on). Next time you’re designing for a page that is likely to be read by people using the latest tablets, try switching out your ‘Click here’ call to action text with ‘Press here’ instead, and start gathering feedback on the change straight away.
- 3. Get to understand how human vision works
Visual neuroscience constitutes one of the largest research bodies in the world right now. They frequently run public outreach programs, and publish jargon-free versions of their research findings. Even Wikipedia can confer a pretty good insight into theories of vision. Learn them, think about them, and use them in your next project.
If you’re more visually-oriented, iTunes U publishes hundreds of lectures on the cognitive science of visual perception, all of which are available to download free of charge.