Does Anticipatory Design Assume Too Much?

In my previous life working in finance, I worked long hours, had a tortuous commute, and high stress. I did everything I could to simplify my routine, including taking a page from Steve Jobs’ playbook by wearing essentially the same thing to the office every day. I had about 10 pairs of the same black pants, and a few different colored jackets and sweaters. Everything matched with each other, so when I dragged myself out of bed at 4 AM every day, I didn’t have to think too hard about what I was putting on. While technology has made our lives easier in many ways, it’s also given us a dizzying number of options to choose from. Just try searching for a new mattress online, for example, and you’ll see what I mean. The choices are nearly endless, and can easily burn a person out pretty quickly. So when I started hearing folks in the UX community talking about “anticipatory design” as the next big thing in experience design, I was intrigued. The concept, first coined by HUGE CEO, Aaron Shapiro, promotes designing products and experiences to anticipate what customers are likely to want to do next by using the data available. I immediately imagined having my favorite sushi joint ready with my takeout order when I was just a few blocks away—just because I was in the neighborhood.

A new take on a common practice

It sounds like magic, but the concept isn’t all that new. After all, isn’t part of the foundation of a great experience based on anticipating your customer’s needs? We already see anticipatory design being used with services like Netflix and Amazon. Both companies base recommendations on prior purchases and preferences to personalize your experience, and presumably, provide more accurate or relevant suggestions. But, according to Shapiro, the problem with these types of services is that they don’t necessarily make our lives easier by offering us more choices. In fact, the plethora of options we’re now given on a daily basis is just exacerbating decision fatigue. And that’s where Shapiro thinks a more modern, assumptive style of anticipatory design can help. As he notes in an article for FastCompany,

Anticipatory design is fundamentally different: decisions are made and executed on behalf of the user. The goal is not to help the user make a decision, but to create an ecosystem where a decision is never made—it happens automatically and without user input. - Aaron Shapiro

Devices like the Nest Learning Thermostat may be closer to what Shapiro means when it comes to removing unnecessary choice or decisions. The thermostat actually learns from your behavior every time you change the temperature in your home and when you’re home or away, and adjust its programming to match your anticipated needs.That got me wondering about the possible dangers of making these kinds of assumptions “on behalf of the user.” You know what they saying about making assumptions, right? Doing this on behalf of users felt like dangerous territory to me, so I wanted to take a look at a few potential landmines for assumptive design, and how we, as champions of great experiences, can utilize anticipatory design for good while avoiding the pitfalls of assuming we know what our customers want. (Reminder: You are not your user.)

Privacy

One of the examples Shapiro gives suggests an app using the information in your calendar to make flight reservations or pre-arrange for car service after a meeting. While that sounds great in theory, what’s required for all that to work is full access to all your personal information. To make assumptions, companies (and the systems they use to learn about you) need every bit of intel they can get ahold of. That means relinquishing a large portion of your personal privacy. And no matter how secure a company thinks it is, your customers will always be wary of a breach in their privacy. That means if you’re asking for massive amounts of highly personal information, you’ll need to assure your customers that their information is safe with you—and do everything you can to make sure that happens. To effectively implement anticipatory design in your product, you’ll need a strong focus on adhering to—and exceeding—best practices of consumer privacy and security. If that trust is never established—or worse, broken—your customers will never have the confidence to give you the level of access you need to their personal details, that will enable you to craft a fully-anticipatory experience.

Choice

Many years ago I went on a date with a guy I only knew casually through friends. We got along well, and I always got the sense he appreciated my independent nature. Until we went out for dinner. He had chosen one of his favorite nice restaurants, and from the moment I sat at the bar, every decision was promptly made by him. I usually prefer wine, but the last time he saw me I was drinking a beer. Instead of stopping to confirm what I wanted, he (wrongly) assumed I wanted a beer. And when it came to dinner, he didn’t even ask what I liked, he just ordered his “usual” for both of us. While this is more of an example of a bad date, it also illustrates how frustrating it can be when others make decisions without your input—or with little or no information on your preferences. With anticipatory design, you run the risk of taking what little information customers give you (see privacy above) and making educated guesses based on that alone. While there’s only so much you can do with the information you have, there are ways to help customers feel like they have a say in the choices made for them—without adding to the burden of decision fatigue. For example, if I’m a frequent traveler that often books a single, king bed when I get a hotel, an app may understandably assume (ahem, anticipate) that’s what I want. But what if I was planning a girl’s weekend to Vegas, or maybe I have children I’m bringing along on vacation this time? There’s no way any app or service will know that. But they can confirm that assumption with me. A simple confirmation, like, “You typically reserve a single, king-sized bed, so we’ve selected that for your reservation. If you’d like something different, you can change it here.” Making your customers’ lives easier by eliminating the choices they need to make is a great idea, as long as you don’t remove their option for choice completely. No one likes to have their order taken without any input.

Dark patterns

If you know something about your customers then you naturally will want to use that information to help better their experience. But beware of the temptation to resort to dark patterns in the process. Dark patterns are designs used to influence people to unknowingly take actions they may not have intended, like adding a warranty to a purchase, or signing up for a recurring subscription. Just because you understand the psychology and patterns of your customers that doesn’t mean you should use that information solely to your own benefit. If you’re using anticipatory design, be sure to constantly ask how your design is ultimately serving your customers’ needs, while being empathetic to their journey.

Anticipate without assuming

It’s a fine line, but it is possible to anticipate without assuming. By using whatever data your customers are willing to provide—and treating it with care—and then providing helpful suggestions rather than declarations, you can help assure your customers will feel an ease on the load of choices they made that day. All without making them feel like you’re making all their decisions without their input.