Summary: Dark patterns have always been regarded as bad design practices that ultimately hurt businesses and users more than they help them. But there isn’t much data to prove this. If we want to defeat dark patterns once and for all, we need to measure them in a quantitative format and thus demonstrate that in the long term, they will net a negative impact on the KPIs that they improve in the short term. Until we can do that, we should remain open-minded and unbiased in our approach to these issues.

It’s time for a reality check. Classically, the UX community has been strongly opposed to the usage of dark patterns in design. If you’re not familiar with what dark patterns are, they’re essentially user interfaces that have been carefully crafted to trick users into doing things, such as buying insurance with their purchase or signing up for recurring bills. At first glance, a dark pattern may simply appear as a mistake in a design, but it isn’t – it’s a very deliberate element that has been crafted with a solid understanding of human psychology. For this reason, dark patterns are commonly seen as unethical design practices that are detrimental to UX and erode a user’s trust in a product, website, or brand.

In one recent manifestation of this, Dan Schlosser published an excellent article detailing the dark patterns that LinkedIn has seeded into their sign up flow. In a nutshell, he was able to demonstrate how LinkedIn was using it’s sign up flow to trick users into surrendering their email contact list, under the guise that it would enhance their LinkedIn experience. What LinkedIn neglected to communicate was that they would be using the contact list to send numerous emails to all of the contacts that hadn’t yet signed up for LinkedIn, on behalf of the user and without their knowledge. But this wasn’t just a one-time occurrence. In fact, Dan was able to document 8 separate instances in the sign up and onboarding flows where LinkedIn attempted to trick him into syncing his contact list.

Dark Patterns in the signup flow

Dark Patterns in modals

Unsurprisingly, this sparked a massive uproar in the UX community, with designers shaming LinkedIn for their shady practices and claiming that it would undoubtedly lead to their downfall. Yet, in the midst of all this craziness, nobody was able to provide any conclusive data to support why dark patterns are actually bad for businesses. Nobody was able to demonstrate how these practices would indeed lead to the demise of LinkedIn. And perhaps more importantly, nobody was able to prove that LinkedIn was even making a bad decision in the first place.

We’re All Underinformed

Despite the fact that UX is a very data-driven profession, we spent days sharing our opinions on the matter and speaking in ominous tones about what we “know” will happen to LinkedIn. We did all of this in the full absence of any actual data from the company. Sorry guys, but until we somehow manage to obtain LinkedIn’s safely guarded qualitative and quantitative data, none of us really know what we’re talking about (myself included). In theory, this entire argument is pointless until we gain access to that data.

Discussion on the Medium article

Discussion on Reddit

But we know that LinkedIn probably won’t ever publicly comment on this matter, nor will they hand over access to their analytics accounts. So instead, I’m going to share an opinion from the other side of the argument. But I urge you to take everything that is said (from both sides) with a grain of salt. We shouldn’t be arguing based off of expert opinions or personal observations alone, but alas, here we are.

UX Doesn’t Just Serve Users

I want us to begin by checking our understanding for the purpose that the UX Designer is meant to serve. In recent years, there has been a massive proliferation of the idea that UX and business goals inherently sit in opposition of each other. In one corner, we have the altruistic UX Designer fighting for the good of the user, and in the other corner, we have the greedy company fighting for more revenue. This represents a fundamental misunderstanding of UX.

The UX Designer is responsible for serving the needs and interests of both the user and the business. Not one or the other. And the idea that these interests must be mutually exclusive (as in, you can either optimize for the user or the business, but not both) is, yet again, the mark of a fundamental mistunderstanding of the role that UX serves. The most ideal and optimal designs will take opposing pain points from both the user and the business, and devise an outcome that solves for both. UX Designers sit on the same team as their company, and as a result, they need to solve for the company’s interests as well.

But Dark Patterns Are Still Bad

Are they? How can we really know that? LinkedIn is a gigantic and highly advanced tech company, representing the third largest social network in the world. There are more than 350 million people using LinkedIn and the company has a goal to reach 3 billion users in the long-term future. In 2014, they did more than $2 Billion in revenue. They source, analyze, and maintain one of the largest and most valuable data collections known to man. Can we really say, conclusively, that all 7,600 of LinkedIn’s employees (not to mention their expansive number of advisors and shareholders) are completely ignorant to the effects of dark patterns? Are we, the outsiders, aware of something that they aren’t? Or would it be more reasonable to assume that LinkedIn has likely spent immense funds and man hours testing, refining, and verifying the very practices that we’re debating about right now?

This is an InMap. It represents one user’s relationships, usage patterns, etc. This is the data that LinkedIn previously let any user access about themselves. You can rest assured that LinkedIn’s internal data is significantly more robust than even this.

Now, you could argue that this all could merely be the result of conflicting corporate interests, selective data collection, and misinformation. If so, I would agree that you’re on to something there. Considering the aggressive nature of LinkedIn’s user acquisition goals, I would not be surprised to find out that they’re resorting to irresponsible methods to hit their numbers. But we simply can’t ignore the fact that we’re dealing with an issue that would likely have been very obvious to any notable company, much less a sophisticated tech giant like LinkedIn. For that reason, the approach that we take to this argument has to change.

How To Win The Fight Against Dark Patterns

First we need to accept that, by all accounts from LinkedIn’s perspective, dark patterns may in fact be working very well. Second, we need to understand that it is our job as UX professionals to serve both the user and the business. This means that if all data points to the dark patterns being good for the company and not bad for the user, then we need to respect that. We shouldn’t be making decisions based off of gut feelings. Third, if we still know that the dark patterns are negatively affecting the business (and simply aren’t being adequately accounted for), then we need to learn to properly collect data on the matter and communicate it in the language of the business.

Let’s assume that every KPI (Key Performance Indicator) that LinkedIn is tracking on their dark patterns are bottom-line quantitative metrics. However, the majority of existing data on dark patterns is qualitative. We’re speaking two different languages here. If we want to communicate the true negative impact of dark patterns, we need to tie them directly to the quantitative metrics that they’re affecting. Revenue, acquisition, etc. We then need to conclusively demonstrate how, over the course of time, these metrics will receive a net negative impact as a result of the dark pattern. We can’t simply say “dark patterns are bad for the longevity of a business”. We need to demonstrate it using tangible data. For example, say that the dark pattern was put into place to acquire more users and increase resulting revenue. We would need to prove that, while user acquisition and revenue may increase dramatically for a couple quarters, over the course of several years, we will actually end up with less total weekly active users and less total revenue due to the deteriorating effects of dark patterns. It has to be demonstrated that the long term negative effect of the dark pattern is greater than it’s immediate positive effect.

For example, we could track monthly active users and demonstrate the diminishing return of the dark pattern over time

Another creative way to achieve this is through the usage of predictive metrics that merge qualitative data into a quantitative format. Net Promoter Score (measuring the collective experience) and the Single Ease Question (measuring the task-level experience) both have been shown to serve as predictors for revenue and retention change down the line. But perhaps most importantly, these metrics get companies thinking about external feedback that comes from the users (rather than internally collected data) and they connect classically unquantifiable practices (like dark patterns) to bottom-line KPI impact.

Even these numbers have their flaws though, and until we can tie practices like dark patterns directly to eventual negative impacts on the metrics that they’re intended to improve, we’ll never conclusively know whether or not they’re always detrimental to the end results that brands generate. No matter what, we should always take a balanced, unbiased and calculated approach to these matters, remaining open to new data and concepts, even if it indicates that LinkedIn’s evil practices just may work.

I talk about this more on the UX and Growth Podcast