Designing for Better Decision Making
Designers are increasingly tasked with supporting users in making complex and important decisions from business management software to choosing health care plans. But aside from designing generally intuitive systems, there are few formalized structures or best practices for designing decision making experiences. That’s where decision science, which also informs behavioral economics, can give us some insight. In this article, I’m going to explain how we can use a few key principles from years of decision science research to design for better decision making.
Decision scientists study how people make choices.
Decision Science happened when psychologists and economists started talking, and realized that maybe the reason that their traditional economic models weren’t working was because people don’t behave in the formulaic ways that their computational models suggested. Because we’re only human, and we’re way too weird for all that. Richard Thaler just won a Nobel Peace Prize in Economics, and also wrote Nudge, which you should definitely read.
So what have decision science nerds found out that designers can hijack?
We make 95% of their decisions using mental shortcuts or rules of thumb.
We make hundreds of tiny decisions everyday, from choosing when to wake up, what to eat, where to park and what Netflix to kill our evening watching. While we like to imagine that each of these decisions is well thought out, in truth we rely on lots of shortcuts so that our brains don’t have to go on overpower calculating pros and cons on every tiny decision we could make.Daniel Kahneman first coined these mental shortcuts “Heuristics,” as in rules of thumb that dictate our choices. Heuristics aren’t necessarily bad, but they also aren’t rational. As designers, understanding heuristics helps us see the patterns in our users behaviors.
Heuristic 1: What I had last time, please
For a brain that wants to avoid thinking, defaults are the holy grail of shortcuts- why bother making a decision if you can avoid it all together? Research shows that defaults are the most powerful tool that designers have, studies in everything from default insurance policies, organ donation and saving accounts. As a designer, a default is often as simple as a pre-filled option on a form. Yet with this power comes extra responsibility so that nudges don’t turn into dark patterns.
How to use it:
- If there are no negative consequences to a default option, such as encouraging people to take care of themselves, donate to charity or sign up for savings accounts, go for it!
- If you’re selling something, even if it’s really cool, use defaults thoughtfully. Recognize that things can get out of control, and quickly.
- Adding something extra to a shopping cart with a checkbox isn’t just across the line, it’s illegal in the UK.
Heuristic 2: Out of sight, out of mind
Are there more gun deaths in the United States or in Southeast Asia? I can think of at least 5 major mass shootings in the US, and none in Southeast Asia, making me think that the US wins. But hold up- Southeast Asia is much bigger, and pretty dangerous, and when you fact check it has way more gun deaths. The point is that our beliefs are skewed towards what we’re exposed to. Huge for fake news. Decision Scientists call this the availability heuristic, a mental shortcut that relies on immediate examples that come to a given person’s mind when evaluating a specific topic, concept, method or decision. The availability operates on the notion that if something can be recalled, it must be important, or at least more important than alternative solutions which are not as readily recalled. Still don’t buy it? With the craze of fake news and bias in the media, NPR recently did a piece about why the public perception of crime exceeds that reality.
How to use it:
- If you want people to stay informed, you have to fight fake news with real statistics.
- When collecting any user data, don’t ask them to compare things that they don’t know about, or worse, know varying amounts about.
- Be on the defence for discrimination and bias, which can end up in our designs. Just because you don’t know a diverse pool of people doesn’t mean that they don’t exist. Keep personas diverse and based on a wealth of knowledge from actual other humans, not yourself.
- Design data visualizations that show the whole picture, not just the anomalies.
Heuristic 3: Adjust from the anchor
When shopping for say, a new TV, a totally rationale human would make a decision based on what they need and want. And if your budget is $100–300, than you should really only look at TVs in that budget. So why is it that somehow just glancing at the $1000 TV makes all the difference?
My dad always loses to this one- when he buys a new device he spends hours contemplating it, before consistently choosing the second cheapest. He doesn’t really understand what the best decision would be, so he is really just choosing what level of sacrifice he wants to make.
How to use it:
- Use smart but ethically- avoid dark UX patterns like tricking people into spending more money than they really want to.
- Short term people may be happy to be tricked into buying a slightly nicer TV, but you have to live up to it, or risk long term brand issues.
- Unrealistic or unreasonable anchors can scare people away. Know your users enough to give them anchors that they can work with.
Heuristic 4: The Top Percentile
Hate to be the one to break it to all the elementary school teachers, but we can’t all be the “best”- the math just doesn’t work like that. Yet multiple studies show that on average, people are remarkably overconfident, and really bad at knowing where they stand in relation to others. Evolutionarily, this might make a lot of sense- if you’re not a little overconfident you might get sucked into self-doubt.
How to use it:
Because we’re so awful at guessing how well we’re doing in comparison to others, it can be very powerful to provide actual data on it. A few cool examples include:
- Reducing energy use through social influence
- Classroom leaderboards (although personally I’m skeptical of the emotional consequences on this one)
- Encourage people to slow down and think when making a big decision that could have negative consequences by communicating that smart people (like them) have screwed this one up.
Heuristic 5: Emotions “change” the facts
If you’re a human, you have a thing called emotions, and they run the show. We have to recognize that our users are impacted by their emotions all the time. And to an extent, we can control them even with little things. Marketing people have known this for a very long time, which is why they show feel-good commercials rather than ones that make you feel too uncomfortable.
How to use it:
- Understand the emotional state of your user through their actions. With the increased availability of chatbots and natural language data mining, we can understand how our users are feeling more than ever. But we also have to use this data responsibly.
- When designing a user journey, consider what you know about their emotional journey, and design around this without taking advantage of users. Yes, it’s easier to take advantage of people when they are upset, stressed and sad, but it won’t help you build a relationship with them in the large term.
Got more questions than answers? Me too. I’ll be digging deep into some more case studies to figure out how to use Behavioral Science to design things better. Comment if you have examples, suggestions or just want to say hello!
