The design of legal terms

Lady Justice Mural. Photo by Lotus Johnson. Some rights reserved.

A case study featuring Uber’s November 2016 update

In the UX world, we talk a lot about how important it is to design with the needs of users in mind. We’re passionate advocates for patterns that are recognizable, set clear expectations, and are easy to follow. We’re aghast at companies that place misleading text on buttons to inflate clicks, or that otherwise fool people into taking actions they didn’t intend to take. It feels warm and fuzzy to be in the important role of Protector of the User Experience™.

Design-minded people are quick to rally behind a litany of good UX principles, but many of us are complacent and lazy when it comes to the experience people encounter when they sign up to our services. I’m talking specifically about the legal language people are faced with and then asked to agree to as new users.

We have lots of excuses for these almost uniformly bad experiences:

• No one ever reads terms or legal language, so why bother?
• Lawyers and courts use a certain brand of hard-to-understand language and patterns that we just can’t get away from
• We don’t want all this legal language but certain jurisdictions and laws force our hand and we have to protect the company
• Our intentions are good and we would never do anything to compromise our user’s trust so it’s OK that they don’t understand everything they’re agreeing to
• It’s hard to write this stuff in a way that’s easy to understand and there are higher impact consumer experiences we think are more important

To these things, I say: bullshit!

I’m not a lawyer so I can’t make a detailed legal argument that would stand up in court, but I can point to precedence. There are more and more examples of companies who care enough to make baby steps towards helping people understand what they’re agreeing to by signing up to a service. Two early examples are Pinterest and 500px. They present their terms in two columns, with the full legal text on the left and a shorter, easier-to-understand summary on the right. By spending just a few minutes with either of these legal documents, a user can get a clear understanding of what they are being asked to agree to. It’s not perfect, but at least they’re trying.

From Pinterest’s Terms

There’s also reason to think that the law may be shifting in favor of simpler, easier-to-digest terms. In a 2012 argument for United States vs. Nosal, Chief Judge Kozinski laments how unreasonable terms of service sometimes are:

“Not only are the terms of service vague and generally unknown — unless you look real hard at the small print at the bottom of a webpage — but website owners retain the right to change the terms at any time and without notice … Accordingly, behavior that wasn’t criminal yesterday can become criminal today without an act of Congress, and without any notice whatsoever.”

Confused dog.

In March 2016, the Department of Business Innovation and Skills in the UK began to gather research from constituents to understand how they currently engage with legal language. This is the first step in an effort to create guidelines that will encourage companies to write terms of service with comprehension in mind. In his forward to the call for evidence, then Minister for Skills Nick Boles writes:

“… customers may be missing differences that could have a significant influence on their choice. That is not good for the individual but it is bad for the economy as a whole also — individual choices drive competition between businesses and boost productivity. That is why in this call for evidence we are looking for practical ways to make T&Cs more transparent and accessible for customers.”

Let’s move on from the potential legal and business arguments for more human terms and delve in to the murky waters of our moral imperative as members of the design community.

The community of product designers and UX writers are pretty quick to lament dark patterns, confusing design systems, and the pros and cons of changes to visual language. We anxiously dissect how to give useful design feedback and we wring our hands over how to make our content more conversational (but not too conversational!). Yet very few of us seem to care that we’re complicit in putting legal handcuffs on our users in one of their first interactions. Here’s how it tends to go down:

• A company designs thoughtful marketing campaigns aimed at getting people interested with promises to disrupt the world and make their lives better. Prominent themes usually include community, living your best life, being a better human, saving/getting time.
• People decide to download the app, or sign up.
• But before they can experience the revelation we have on offer, we present them with an objectively complicated, long piece of content that contains a bunch of legally binding statements. We ask them to tap “Agree” to continue.
• And they do, not because they actually agree or understand, but because our industry has desensitized them to this experience so much that they barely even think about it.
• For most people, that’s the end of it but others who experience a meaningful legal problem such as discrimination find themselves constrained by legal devices they didn’t read and didn’t understand.

As a community we’ve accepted that this is OK or at least that it’s acceptable to ignore. This isn’t even a dark pattern — it’s the status quo.

We’re entering a new phase of politics where our right to litigate unfair or discriminatory practices may be more important than ever. Now, especially now, people who claim to care about user experience need to advocate for clear, direct and human legal terms so that people are able to make valid personal choices about consent and so that they understand what data is collected and how it will be used.

Uber

Uber is not unique or even particularly bad when it comes to the length or the nature of the language they use in their most recent terms update (November 21, 2016). They’re just getting my attention because they made a recent change and they’re pretty representative of how most tech companies do this.

Notification email

Uber email: received Novmeber 22, 2016

On November 22, 2016 I received an email from Uber with the subject line “We’ve updated our Terms of Service”. Points to them for a clear subject line and easy-to-read and understand email text.

But on closer examination, there are a few problems:

• The email begins with a reference to Uber’s international expansion and ties that positive growth to the update of their terms. This is dishonest given that the changes to the terms only apply to people using Uber in the USA.
• The lede of this change should be the update to their arbitration agreement. If you click through to the full terms, there’s some bold text near the top of the page that emphasizes the section on arbitration, so clearly it’s legally important for Uber to show that they’ve tried to draw attention to it. Yet in the email this topic is buried as the third of four bullet points.

• According to Uber’s email, the terms are effective as of November 21st and after that date if I use the app I’m confirming that I’ve read and accepted the new rules. How is this timeline possible when Uber only sent the email on the 22nd, the day after the new terms took effect?
• Lastly, Uber embeds links to the full terms and to their help text under the word “here”. “Here” is not descriptive link text and is a shoddy design pattern that’s been thoroughly lambasted and discounted everywhere (for about 16 years!). It’s not accessible, informative or actionable. It’s unlikely to get people to click through to actually read the terms. (See how hard it is to figure out what you’re clicking on when I tell you to go here, here, and here to read my sources on why this is bad?)

Terms

Uber’s terms look like what a lot of companies do. That’s why they’re a great example of why most consumers have no idea what they’re agreeing to when they sign up to use a service.

• The language of the terms is hard to understand and cynical. If we think of this as a designed part of the user experience, which we should, it’s set up to obfuscate information so that the average person is unlikely to know what they’re agreeing to. Simpler word choices, or even visually organizing the page for scanning would make an immediate improvement.
• The rules set out in these terms are applied to existing customers by default. Instead of asking people to consent to significant changes, consent is automatically applied. Why? Because it’s less overhead for the company and puts the entire responsibility on their customers (this is standard, by the way). If you don’t want to accept the changes to Uber’s terms, you need to read down to the fifth paragraph where it says that to opt out you need to provide notice to Uber in writing within 30 days of the date the new terms are published. If you fail to do this, you’re legally bound by these new rules.
• The framework for these terms is dishonest. Any company that presents terminology like this is well aware that most of their customer are going to be unwilling and in many cases unable to read and understand this kind of language. Yet the entire premise of the terms is this: “By entering into this agreement, you expressly acknowledge that you have read and understand all of the terms of this agreement and have taken time to consider the consequences of this important decision.” In this case “entering into this agreement” means doing nothing but failing to write Uber a letter within the next 30 days. Given that Uber probably monitors analytics on its emails, it knows full well how few people likely click through to read the terms. It’s almost certainly also looked at things like the reading score level of the document and knows it to be over a grade 12 reading comprehension level. That’s almost double the literacy level of the average American.
• There’s no way to easily compare the prior version of Uber’s terms with this new version to understand what’s changed. We know from the disclaimer at the top of the new terms that there have been changes to the arbitration agreement portion but it’s hard to identify exactly what’s changed. This section of the terms is important because it sets out how people are allowed to go after Uber to seek redress if they believe Uber has engaged in illegal practices. We would need to go back and forth between the old version and the new version to really dig into what has changes so we can “fully consider the consequences of this important decision.”
• Looking purely at the updated terms, it’s clear (and unsurprising) that Uber wants to avoid being sued by its passengers. By using Uber under these new terms, you’re agreeing that instead of suing Uber for wrongdoing, you only have the option to go through binding arbitration. This means that instead of going before a judge and jury, potentially in collaboration with other people who share similar concerns (in a class action lawsuit), your only option would be to have an independent arbitrator review your paperwork and make a judgement. You don’t even have a right to a hearing unless your claim is for over $10,000 and you don’t have the right to have your case heard in conjunction with other similar complaints. Finally, the decision of the arbitrator is final. There’s no appeal and no other body to reach out to if you feel the process is unjust.
• One especially sneaky clause is that unless you write to Uber with your objection to the terms within 30 days, you’re bound by the new rules even if you’re engaged in an existing lawsuit against Uber. The terms say, “This will preclude you from bringing any class, collective, or representative action against Uber, and also preclude you from participating in or recovering relief under any current or future class, collective, consolidated, or representative action …” (emphasis mine). All the people who are currently suing Uber need to write to Uber within 30 days to disavow these terms or their lawsuits will be “precluded from participating or recovering relief” from Uber.

I could go on. The point isn’t to pick on Uber because they are no better or worse than most other tech companies. And that’s the important bit:

As UX folks, we need to pay more attention to this stuff because it matters.

Companies, even great companies that provide valuable and worthwhile services exist first and foremost to make a profit. No matter how warm and fuzzy their marketing material is, increasing the value of their company for shareholders is literally their top priority. This is fine, but it’s important for consumers to gain back some equity in that relationship and it’s our job as designer practitioners to advocate hard for that.

The best way to judge how much a company cares isn’t in their pithy release notes, quirky animations, hazy-filtered promo videos, or cute graphics. It’s in how much the company tries to clearly explain or obstruct your ability to make important legal decisions about what you give it and its role in your life. Companies can ask for what they want but they should be legally and morally compelled to present their request to consumers in ways that ensures clear understanding and consent.

As design professionals, we need to own up to our role and responsibility in this.