Insight | UX

Dark patterns

Dark Patterns; they’re evil, nasty, horrible things and it’s a fair bet you’ve come across them before. But why do these UX behaviours grind my gears so much (even more than the phrase “grind my gears”)? And why should you be on the look out for dark patterns creeping into your web project? Let's take a look at some examples and explore how they make their way into our everyday lives.

What is a 'dark pattern'?

In the early/mid 90s, Jakob Nielsen (the Mr Miyagi of web usability) published a list of 10 heuristics for User Interface design. This list became the bedrock of what would be considered ‘best practice’ in terms of creating a positive experience for a user.

Obviously, not all systems adhere to these principles and on occasion mistakes will be made; sometimes the implementation will be rushed, sometimes the design will be flawed or the initial idea poorly thought out. Sometimes though, bad design happens on purpose and elements of the user interface are crafted with the aim of deceiving or misleading users, essentially reversing the best practices that Nielsen outlined. These are Dark Patterns, and they make you feel like this:

Vader realised he’d forgotten to cancel his free trial.

Types of dark pattern

There are a number of types of Dark Pattern, and I won’t bore you by going through them all, but here are a few of the more common patterns to give you a flavour of the kind of stuff we’re talking about here. If you haven’t come across one of these before, then I can only assume you’ve never used the Internet.

Bait and switch

The oldest trick in the book. The user sets out to do something, but during the process something else happens as well that they were not aware of and were not expecting would happen. To give you an idea of how prevalent this approach is, Microsoft made the news last week after changing the behaviour of their 'Please upgrade to Windows 10' popup so that clicking the close button activates the upgrade instead of just closing the popup.

Forced continuity

This is where the user signs up for a free trial but has to enter their credit card details. Then at the end of the trial they automatically get put on a paid membership service and billed accordingly. Sound familiar? You can find this on a large number of subscription-based websites that give out free trials. This pattern is often used in conjunction with…

Roach motel

Where the user finds it very easy to get into a certain situation, but getting out is another matter! You know the kind of thing, you sign up online and it takes about 10 seconds, but when you come to cancel you find yourself trawling through options menus and help pages only to stumble upon a hard to see link that tells you that you can only cancel over the phone. Grrrr.

Trick question

Where the user is prompted to answer a question that can easily be read to mean the opposite. I.e. “Tick this box to not sign up to our mailing list”. Putting a check in a checkbox is a positive action, so assigning a negative action to this is deliberately trying to dupe the user.

There are many more types of Dark Pattern. If you want to read more about them then check out the Dark Pattern Library for descriptions and examples of each one.

When a measure becomes a target, it ceases to be a good measure.

So, let’s look at a fictional example. Fudgetech Ltd has a KPI that measures the percentage of new customers that have checked the box to sign up to the monthly mailing list during registration. At a review meeting, the guy in charge of website updates (let’s call him Edgar – I’ve been re-watching 24 recently, and Chloe’s probably above general support tasks) is told that this percentage is far too low and needs to be increased. I say told, I mean shouted at, and Edgar knows this means it needs sorting ASAP.

Edgar scurries back to his desk and thinks about different ways to get more users to sign-up to the mailing list. With the rollicking fresh in his mind though he starts leaning towards the quickest way to get users to sign-up to the mailing list, because increasing the result on the KPI is the best way to prevent another verbal barrage. Edgar has a light bulb moment; why not switch the ‘opt-in to the mailing list’ option on the registration form to opt-out instead. Better still; why not make the font smaller so users are less likely to notice. “If they don’t actually want it, they can unsubscribe later” reasons Edgar, because all he’s been asked to do is increase sign-ups of new users.

Eventually the changes are launched and sure enough, the majority of users registering fail to notice the checkbox with the tiny “click here to not sign-up to our mailing list” text. At the next review meeting, Edgar displays a shiny graph showing a sharp rise in the percentage of new customers signing up to the mailing list. It was considered a job well done, and there was much rejoicing.

It was considered a job well done, and there was much rejoicing.

What we see here is a classic example of a ‘bait and switch’ dark pattern, with a dash of ‘trick question’ added in for measure. In this case the root cause was a subtle, but fundamental shift in the objective that put the focus of the end-goal on metrics, rather than the user experience. What the KPI doesn’t show is the dissatisfaction that this negative experience causes, and how this affects the user’s perception of the company. In future, there may be a drop-off in completed registrations or there may be a subsequent increase in people manually unsubscribing from the mailing list, but for some users the damage will already have been done. It may only seem like a minor irritation but put yourself in the user’s shoes and ask, ‘what would I think of a company that did this to me?’

Understanding 'bad' design

Just as it's important to understand what constitutes 'good' design, it's equally important for everyone involved in a project to understand what makes a design 'bad' so that we can guard against these things creeping in, especially when projects hit crunch-time. We all know from our own experiences that transparency and honesty improve the customer’s relationship with ‘the brand’ and at the end of the day, we need our users to have trust and belief that we are looking after them. The use, intentional or otherwise, of Dark Patterns may help make stats look good, but you take a big hit on customer satisfaction and trust, and that’s not a trade worth making. So whatever you do, keep your focus a user-centric one and don’t be tempted to join the dark side; they’re all miserable sods and no one likes them.