It’s often said that the best design is invisible. After all, a user trying to accomplish a task on a website or app is likely thinking about their goal and not the site (in the same way that a person cooking dinner is thinking about the food and not the spatula). More often than not, the goal of the designer is simply to get out of the user’s way.

To the end-user, it isn’t always apparent how much work has gone into making something intuitive and elegant. In fact, they shouldn’t be thinking about the designer (or the design) at all.

The goal is really to create something that ‘just works’ in such a way that it doesn’t draw undue attention to itself – and it’s been said that some of the most elegant pieces of design in human history are so good and ubiquitous that we almost universally forget that somebody had to create them (the concept of the T-shirt, for example).

This ideal should be the goal of designers across many fields, and website design and user experience (UX) are no different. To that end, let’s look at some ideas and examples of ‘invisible’ UX optimizations that users will never think about…

The labour illusion

A team of scientists at Harvard Business School once conducted a study into the psychology of being made to wait for loading bars on websites. Our intuition might lead us to believe that being made to wait is invariably bad and that users will always want everything as fast as is physically possible, but the researchers found that this wasn’t the whole story.

Their study consisted of recreating and rebranding a popular travel booking website and presenting modified versions of it to different groups of participants – each person was instructed to try to search for and book the same holiday, and was later surveyed for their opinion of the website after completing the transaction.

50% of users saw only a loading bar, the other 50% an additional explanation of what the site was working on

Some users were shown their search results instantaneously; others were randomly made to experience a waiting period of anything from 10 seconds up to a minute. All of the users who had to wait were shown a progress bar, and some were also shown a visual display of the various processes undertaken by the service to populate the results page as an explanation for the delay.

The results, surprisingly, showed that the site was consistently reviewed the most favorably by users who had experienced a wait of up to 50 seconds but had been shown an explanation throughout of what the site was working on – even more so than the users who had received their search results immediately.

The researchers conducted other experiments, such as a simulated dating site that in some cases would deliberate for a time over the production of a ‘perfect match’ – in each case, users consistently responded more approvingly to the site that had taken a little longer but had provided some transparency about what it was doing.

They ultimately coined the term ‘labour illusion’, a concept with a subtle but important difference from operational transparency (an idea with which it sometimes overlaps). In other words, even if a task doesn’t actually take that long for the website to perform, there may actually be value in providing the impression of a system that is working hard to fulfill the user’s request.

Especially in the case of travel bookings, romantic matchmaking, and price comparison websites, a user might be slightly distrustful of a site that returns the answers too quickly. How thorough was it, we might wonder, in finding the very best deal for us? In a strange way, we are perhaps reassured that our query is being taken seriously by the system and chewed over with the appropriate meticulousness.

Airbnb’s machine learning secret

Another example of behind-the-curtain magic making a big difference to the user experience of search results has been employed by accommodation service Airbnb. It’s something a user would scarcely think about – you just type in your search for where you’d like to stay and then potential hosts come up. Simple, right?

In reality, there is an enormous amount of technology working behind the scenes to deliver the goods. After all, Airbnb offers a two-sided marketplace that wants to provide value for both guests and hosts – and there are many reasons why a host might not accept a request from a user looking for a place to stay (timing conflicts, amount of notice given, and so on).

To keep both sides happy, Airbnb implemented machine learning in 2015 as a means to make predictions about the preferences of individual hosts and the types of requests they would be most likely to accept. They recognized that for guests, sending a lot of requests and getting repeatedly turned down would be a bad user experience – and hosts wouldn’t want to be constantly receiving requests from guests they couldn’t accommodate.

Accordingly, Airbnb’s machine learning model pays attention to which enquiries each host does and does not accept and uses that data over time to predict the types of requests that are likely to be preferred. Thus, a potential guest searching the service for a place to stay is shown only those hosts who are most likely to respond favorably to the user’s request.

It is an elaborate system of data collection, mathematical weighting and complicated algorithms that matches Airbnb’s users to the perfect hosts – but none of it is obvious to said users, who rarely need to think about the how and the why of things that ‘just work’.

The dark side of UX

Of course, techniques that affect user experiences in ways that users don’t notice could be used for good or evil. Sometimes, unscrupulous UX designers can use hidden methods to manipulate or trick users into acting against their own best interests – signing up for services they didn’t actually want or clicking advertisements disguised as part of the page.

These so-called ‘dark UX’ techniques can take many forms, and at their most insidious can use the visitor’s own psychology against them – whether by exploiting the tendency of many people to skim-read content and hit the ‘continue’ button or by reeling users into a free trial that quickly becomes a paid subscription they can’t cancel.

Of course, if users do work out that that they’re being exploited it can spell trouble for the company caught employing these techniques. LinkedIn, for example, was found guilty of essentially tricking its users into email spamming all of their contacts in 2015 and ordered to pay a $13 million penalty.

As any Spider-Man fan will tell you, great power must come coupled with great responsibility. Much of what UX designers do is often visible to the end user in the form of carefully coloured call-to-action buttons, the arrangement of navigation elements and so on – but the world of ‘invisible’ user experience practices is an unintuitive secret to most.

Where ‘dark practices’ are concerned, at least, users generally work out what’s happened after a fashion – but when UX is done properly and ethically they should never know how it works or even think about it, and therein lies the beauty of the craft.

The practical details of Airbnb’s algorithm or a travel agent’s loading icon don’t have to be apparent for a user to appreciate their effectiveness – in much the same way that diners in an exquisite Michelin star restaurant don’t need to know the specifics of what happens in the kitchen to appreciate the craftsmanship inherent in a wonderfully prepared meal.

The reality is that apparent simplicity is sometimes the product of a great deal of behind-the-scenes complexity, and in an ideal world the user should never be encumbered with that knowledge. Much like the inner workings of a magic trick, the ‘how’ of a great piece of design is often a secret for only its creator – and perhaps a few fellow professionals – to know and appreciate.

—–

This post was contributed by Angle Studios – expert Kent web designers and UX specialists with over 15 years of experience delivering high-quality website and branding services for businesses across Kent and London, UK.