
How behavioural design creates the best user experiences
Retailers we have a problem.
iPerception’s 2016 report shows that less than 25% of people who intended to make a purchase ended up actually buying something. That’s a huge leak in the buying funnel.
Why did 75% of people leave?
Using iPerception’s report from 2009 the top #1 and #3 reasons are usability-related:
Lesson: Fix your site’s usability issues.
But the problem is even more severe when you realise that this statistic talks about those who stayed on the website.
The second biggest leak is people who didn’t even stay on the website. The average bounce rate for e-commerce is reported to be somewhere between 35 and 57%.
According to a study by Chao Liu and colleagues from Microsoft Research, there is a 10-20 second window where a user decides whether a website is worth giving a chance or not.
What happens during those 20 seconds that makes people leave?
Research led by B.J. Fogg shows that many factors affect people’s trust towards a website, but the most prominent issue is design.
Lesson: Make your website visually appealing.
But here’s something interesting.
It appears that a website’s aesthetics and usability aren’t independent.
Research shows that users find visually attractive websites easy to use, even when actual ease to use is poor. In one usability study, on average, users failed to complete more than half of the required tasks successfully. When they worked with visually appealing websites they still rated their user experience highly.
So, maybe you could kill 2 birds with one stone: by improving aesthetics you might not only convince more visitors to stay, but also increase the chance that they will put up with your usability issues, ultimately buying.
This principle is known as the aesthetic-usability effect.
In the context of e-commerce, the aesthetic-usability effect is when users perceive more aesthetically pleasing websites to be easier to use than less aesthetically pleasing websites.
This means that better aesthetics can result in:
(With that said, there are studies that show that severe usability problems will force users to judge usability independently of aesthetics. Regardless of how great your visuals are, if a user can’t find a product, she won’t buy.)
We saw the aesthetic-usability effect work not only in academic labs, but in real-life business, too. For example, one experiment we ran was solely focused on improving the design of a check-out form. This test resulted in a +3.6% uplift. Additional revenue from that uplift is estimated at €450,000+ per year.
Some of the most important areas to apply that thinking are:
A landing page is the first page that users see when they come to your website. It’s crucial because this is where users form a positive first impression of your website. How well those pages look thus determines whether users will stay or leave, and what’s their level of tolerance with your usability issues.
Each website’s situation is different, but in retail most often visitors enter through your homepage, your category pages or your product pages.
As an example, let’s look at a traffic report for PC World, a British chain of computer superstores.
Based on the limited data that Semrush can provide me with, I can see that their homepage gets the most traffic. Equally, their category pages are right at the top:
Checkout Flow is the stage where people start to open their wallets. Not everyone does though with 69% of users abandoning the process. At least 2 of the top 10 reasons why people abandon could be mitigated by the aesthetic-usability effect:
As we already know, making this stage visually appealing can enhance users’ perception of the funnel being easy to complete and drive them to make a purchase.
A website needs to be attractive, but how do we make it happen? After all, it’s subjective. You might like your website, but do your users think the same way?
There are 2 approaches to answering those questions:
I’m going to show you how you can apply both approaches, looking at examples of PC World’s homepage (customer research) and NET-A-PORTER’s mobile checkout flow (heuristic analysis).
To understand if users perceive PC World’s homepage as visually attractive, we ran a small-scale survey, asking visitors to rate different aspects of that page.
For research we used VisAWI, an 18-question survey that was specifically designed for evaluating website aesthetics. The results of the survey showed that overall the website is visually pleasing. It scored 4.47 points out of 7. The benchmark for e-commerce is 4.05. This shows that PC World’s homepage is better than average:
In particular, users highly rated PC World’s layout as being “easy-to-grasp” and “well-structured”, indicating that it met their expectations as an ecommerce store. Overall, the majority agreed that the site is designed with care and looks professional.
But findings also highlight a number of areas that could be improved.
Some of the reasons why users might think so are:
Lack of whitespace
Text-heavy sections
The “reasons to shop” section is very crowded which adds to the denseness of the page. This is further complicated by how text-heavy this section is, which could leave users perceiving it as overwhelming:
This might be because PC World’s color harmony is disturbed. According to color harmony theory, bright blue and red colors that they use for their promo campaign wouldn’t necessarily match their main brand color, purple.
The theory of color harmony tells us that PC World would be better off following Hulk’s example…
In other words, using green as a complementary color to purple (potentially alongside a more saturated red):
There are 5 types of color harmonies, and depending on the one you choose, you’ll find other matches. Here’s another one suggested by Paletton.com, a tool for creating complementary color combinations:
To me personally (feel free to disagree), it seems that there is their branded color scheme. When seen in isolation its colors match well:
There is the promo-campaign color scheme. Again, when seen in isolation its colors match well:
It’s combining the two together that creates a challenge, and might have resulted in some people feeling that colors look “botched”.
When users say that the design of the site lacks a concept this might mean that a page doesn’t form a complete and pleasing whole. Gestalt theory of psychology suggests that we look at an object as a whole first and zoom in on its individual parts next.
When we have difficulty making sense of objects, we tend to perceive them as less beautiful.
As we’ve already observed with the colour scheme, the page doesn’t form a unified whole as two separate colour scheme do not match each other.
Similarly, the product catalog is chaotically organised. This makes it more difficult for people to group the objects together:
How about grouping those objects by a unified theme? For example, one row of laptops, one row of game-related products, one row of home office equipment. This would’ve made it easier for users to make sense of the page.
Finally, the page doesn’t seem to have a unified purpose. Yes, the main theme is the promo-campaign, but what action should users take next?
With so many competing calls-to-actions this isn’t clear:
Every page should have a dominant call-to-action (CTA). (As you should know from our article on the paradox of choice when there is no dominant option users are likely to get overwhelmed and quit).
Their “summer mega deals” banner takes the most space. So, I assume this was meant to be their dominant CTA. The problem is it looks like an image, not a clickable element which is why people might have been confused as to what they should do next on the page.
If this is the dominant purpose of this page, PCWorld would have been better off designing their “Shop Now” CTA not as text, but as a button.
Look at what Maplin is doing:
While technically, this banner is one single image, the call-to-action was designed to look like a button, so it’s clear to people that they can click it.
If PCWorld did the same and made that button as prominent as the rest of the banner, it would be easier for users to identify the main CTA.
To show you an example of how heuristic analysis can be applied to improve aesthetics, we’re going to look at NET-A-PORTER, a luxury fashion store.
Our designer, Josh Lenz, quickly identified a number of problems:
So, the checkout design looks not only hard to complete, but might also diminish users’ trust in the brand. To reverse that experience he made a number of changes:
All Steps
Step 1
Step 2
Step 3
My hypothesis is that since the page now looks modern, more clear and matches NET-A-PORTER’s brand, more users will perceive it as easier to complete. Not only will more users continue to complete their purchase as the page looks more clear, but users are also less likely to be scared by an experience that substantially deviates from the brand.
The next step would be for NET-A-PORTER to test this design change against the original, and measure if it actually impacts their bottom-line.