What’s wrong with users?
Anyone who has experienced a user testing session on one of their websites will be familiar with the frustration of users missing the blindingly obvious. It’s a good job most testing sessions involve 2 rooms, as the design team’s language is usually fairly blue!
Don’t forget that the act of testing itself influences user behaviour to a greater or lesser extent depending on the skill of the tester and the susceptibility of the subject. A/B and multivariate testing help to eliminate some of these problems, but the results only show you ‘what’ works not ‘why’.
The influence of size and colour (http://www.getelastic.com/cart-button-size/) on the click rates of action buttons is interesting, but the context of the buttons is as important as the button itself. A single big orange button with an arrow, on and page with a lot of white space, will achieve a high click rate. Add another 5 similar buttons, and a lot of graphical full colour images, and the rate will drop off.
Just a note of caution though. If you A/B test each element on a web page to find its optimum setting, the page with all elements optimised will not necessarily outperform other combinations. Think of all your favourite flavours – chocolate, strawberry, cinnamon etc, and then think of all of these combined into one cake – not quite Jamie Oliver is it? Stats and user testing provide great information, but you can’t beat expert interpretation to add real value and show you what to test in the first place.
Breaking site content down so that pages have only one or two main calls to action really helps. However, faced with the need to design a home page and satisfy multiple competing stakeholders, this can become impossible. So you need to allow your users to browse easily to identify their sales channel before pushing them to buy. Once into the sales funnel, you need to direct the inexorably towards the sale whilst re-assuring them at all stages.
Those of us men who are allowed to do the shopping, - and God help us if we come back with the wrong brand of washing machine tablets, - will be familiar with the ‘last packet on the shelf’ syndrome. In this situation, the item on the shelf matches what we want, but because there is no alternative we don’t want to commit ourselves to buying it. This perverse human psychological trait can also be seen at work on website user journeys.
Web developers know that the correct name for the button at the bottom of a form (and default browser setting) is ‘submit’. However, this word is psychologically final, - we only submit something when we’re completely happy with it don’t we? Once submitted there is no going back is there?
So what happens if we change the word ‘submit’ to ‘continue’? Well, ‘continue’ implies two things: we are not at the final stage, and we can go back if we want. And guess what? The number of clicks on the button goes up substantially. So more carrot less stick may be the answer to increasing conversions. This means thinking about button text and sales funnel steps making actions sound less final and providing users with escape routes. The best sites allow you to step back and forwards within a process without having to re-enter data, - there is still the perception among some older internet users that clicking on a ‘Buy Now’ button will somehow automatically debit their bank account - including adding service charges and carriage - with no further steps.
On the other hand, if your site is for kids, maybe your action buttons should contain the text: ‘whatever you do don’t click this’!
Tags: internet users, useability, Web developers, websites
Leave a Reply