Aligning UX Design with User Psychology: An Interview with “Bottlenecks” Author David Evans

Why do some digital experiences engage us and catch on, while others do not?

17 min readSep 28, 2017


In his fascinating new book, Bottlenecks: Aligning UX Design with User Psychology, David Evans explores the psychological constrictions of attention, perception, memory, disposition, motivation, and social influence that determine whether customers or users will be receptive to digital innovations.

David Evans

David is a Senior Manager of Customer Research at Microsoft, where he influences the design and positioning of Office 365, Cortana, and the Office Graph. He also holds a Ph.D. in social psychology.

projekt202 recently spoke with David about his new book and the importance of behavioral design.

What led you to write your new book, Bottlenecks: Aligning UX Design with User Psychology?

Not unlike the researchers at projekt202, I have been involved in custom and ad-hoc research for much of the last 15 years. I really enjoy collecting data sets and answering specific questions, and that sort of inductive approach to answering questions.

I had run my own usability firm called Psychster for about seven of those years. Microsoft has always been my client in one form or another, and I joined full-time about three years ago. Also, while I was doing that, I was teaching at the University of Washington in a couple of different departments — the Masters of Communication, and Human Centered Design — and when I would go to campus, I was thinking about psychology theory.

Bottlenecks, for me, was balancing the inductive — “Let’s go collect a new data set, let’s run a new study” — with the deductive, saying, “If we acknowledge that these psychological processes are at play with our customers, as they use our products, we might be able to make better design decisions even without conducting another study.”

At the very least, it’s conducting a better study with better hypotheses through it, or based on theory. It was my opportunity on the weekends to sit down and remind ourselves, “Hey, let’s not forget about cognitive perceptual and motivational and social psychology here,” because it’s all at play and really fascinating.

You write, “Digital innovations must survive the psychological bottlenecks of attention, perception, memory, disposition, motivation and social influence if they are to proliferate.” For companies to succeed, do you believe there’s a greater need today for them to understand and apply these principles?

I’ve enjoyed other books in the same space, at the intersection of psychology and tech, but look at where we are in the history of tech. Most of the work that has been done here is looking at cognitive psychology, as it relates to websites or other forms of visual, graphic design.

What was important to me was to also tell the story of how things go beyond attention, perception and memory, and into disposition, by which I mean personality, life stage, need fulfillment, and then on into motivational psych.

I think the world, at this point, is rediscovering behaviorist principles of reinforcement and timing, ways that are really relevant.

Lastly, to social psychology, we do a lot of measuring whether people will recommend the products that we build and take to market. I don’t know if we’ve really sat down and said, “Why would they recommend that? What is the social dynamic by which one of those recommendations would actually get acted upon?”

I wanted to tell those stories, but we’re now far enough along in the history of the web that it’s not just that we see what happens when a new platform arises, but now we’re actually aware of multiple platforms. The first one was definitely the World Wide Web and we saw it get populated with websites, then we’ve seen the app platform emerge and really get populated with a bunch of competitors.

I think we can see in the near horizon, in our own lifetimes, that we’re going to have a voice-skills platform. You’re going to have in-car, a computing platform. We are definitely going to have a virtual and augmented reality platform. Really, that same history is going to emerge, whereby, in the early days, a vast number of alternative competitors are going to show up on these platforms, and then we’re going to have a gigantic extinction spasm, so to speak, as some are used and most are ignored, not engaged with.

Is psychology relevant? It’s relevant as we learn how to deal with the competitiveness of all of the different products.

The wonderful thing about working in this space is that it draws from so many different disciplines. It’s really fun to work with folks just getting into UX, project strategy, or project management, or all the different, diverse names it goes under.

In some ways, what I hope for folks who are senior or experienced in this area, is to let you build a more persuasive business case about a certain design by being able to cite theory and psychological science that could be as much as 50 or 60 years old. That gives a framework and a scientific vocabulary for stuff that along the way you’re going to encounter, like people’s memory constraints or signal detection theory.

For projekt202, while we do design work and software across a host of industries, we also do a lot of enterprise software in which there’s emphasis on what you call task-positive paths. That is, they’re really focused on productivity and allowing people to consume information and take action. What kinds of tools would you then recommend for deciding how much emphasis to put on task-negative features, ones that aren’t trying to drive toward a specific, intentional focus?

This is one of the newer psychological processes that I talk about in the book. In a nutshell, neuro-imaging studies using functional MRI have identified two forms of attention. One is task-positive and the other is task-negative. Task-positive is saying, no matter what task you give people — cognitive, mental math, something involving words or semantics, some sort of rational task puzzle — similar areas of the cortex light up.

An equally interesting discovery is that when people are between tasks, they have another pattern of activity in their cortex that lights up. That task-negative was almost harder for people to discover, but, for me, is just as interesting, because when you are not solving a math problem or doing your taxes or navigating with a map, what is your brain doing?

The answer is that it defaults to a case where you’re thinking about new memories, old ideas, going over social situations.

So why is this important? We translated some work that I had done for into a simple question. We popped up a survey as people hit the site, and it said, “Could you tell me about your visit today? Did you have a goal? Or did you have no goal?” There was extra context on that question, but it was very simple.

What we found was that, depending on how people answered, they showed very different patterns of navigation around the site. People with a goal obviously used the search field, because it’s the quickest way for them to meet their goal. That was, at that time, about 75 percent of the people on AllRecipes.

But 25 percent said they had no goal. The interesting thing about them is, even though they were a minority, they were more likely to click on ads, because when you’re in that task-negative mindset, you welcome intrusions. You don’t know what you’re looking for, so you’re open for new thumbnails, new banners, new interstitials.

The takeaway and the reason I was excited to include this in the book is that this mindset — whether you’re task-positive, trying to solve something, or task-negative and mind-wandering — actually determines good business models.

A company’s ROI can be directly impacted by knowing whether users are task-positive or task-negative, correct?

Task-positive people, they’re never going to click on an ad, because they’re using Microsoft Office, right? And they’re doing something, or they’re using Skype and they’re talking. You need to monetize them with some subscriptions. Task-negative people are more likely to be monetized with ads. In fact, if you ever tried to get a task-negative user into a subscription, it’s too much work to even go through the conversion flow.

I think this is an exciting way to look at it, because it determines how you monetize.

This is a great opportunity for marrying together one answer on a pop-up survey, an interstitial survey or an in-product survey — “What is your goal today?” or “Tell me about your visit. Did you even have a goal?” Marry that with the behavioral data that we’ve got increasingly in large volumes on the back-end, and you can really see very interesting, different ways of behaving on that. Again, it was a self-report survey, but then you marry that with the telemetry that you have on how they use the product.

Is that an opportunity, for instance, for creative directors to use a Gestalt perception that lets designers see how all elements can coalesce into a meaningful user experience?

Someone who is trained in these seven perceptual principles, we assume things are related that are close, proximally. We assume things are related in how they work, what they mean, and what they do. If they’re similar in color, shade, size or shape, that’s what these Gestalt principles are. These date back to the 1920s. We assume design elements are affordances. We assume their function based on their orientation on the page.

An expert review, like by a creative director, or even using heuristics, can suss out and fix these elements. If you find them and fix them, what happens is that users are very pre-cognitive. They don’t have to think. Their initial assumptions are correct, instead of incorrect, versus violating all of the Gestalt principles. Then, you put a huge cognitive burden on people.

One of the examples I talk about in the book is back in 2007, when we were downloading Office, before it became Office 365 and an online service. We saw that people had already purchased the product, but then they pushed the wrong button to download. Instead of clicking the download button, they would click some other element on the page, and this was the kind of thing where we found it through an inductive study.

But a creative director who understood these perceptual principles would have said, “That’s a trap and people are going to go down the wrong path,” and they would have caught it.

A lot of points in the book, I really tried to write and rewrite principles such that they could almost be copied and pasted into some sort of document, email, PowerPoint or whatever to put to use, like you’re describing with the creative director who needs this to add to the heuristics that they’re already using in bad design.

This is super powerful stuff and it’s very basic. I brag that when we caught that download problem with Office, we effectively found a multimillion-dollar impedance to the revenue of the business. What usability researcher doesn’t want to find one of those, right?

I really believe that the experience has strong implications for the relationship of the brand, absolutely.

Another bottleneck you discuss in your book deals with the limitation of people’s memory, and how that affects their ability to get work done. Given all the day-to-day distractions and knowing that somebody isn’t working directly and solely in the software at hand, do we need to alter the way we apply that principle, of that threshold of memory?

When you suggest that human psychological hard wiring, our phenotype, is changing due to digital environments, a good scientist is going to say, “Not likely.” It takes far more generations for us to actually see a physiological change.

Working memory has this limited capacity. Information takes up different amounts, but it is very interesting to me that we have this sense that working memory is shrinking. The truth of the matter is even more interesting than that perception.

The truth is, first of all, working memory capacity is not shrinking. We are trying to move more information through that bottleneck than we ever have before. We’re feeling a little bit tight, pinched, not a lot of elbow room, but that’s just really the nature of the competitiveness.

It brings to mind a line from your book: “Our nervous systems are radically bottlenecked, and the retinae of our eyes are only the first of several constrictions.”

Mary Meeker’s internet report came out and, for me, the most striking slide in there was that the number of hours on the internet with a smartphone did not replace the number of hours we’re on the internet with a PC. It added to it. We’ve gone from something like four hours a day online to about eight when you include being online with a smartphone.

That’s what I mean when I say there’s vastly more content that we’re asking people to attend to, and it’s even increasing in the last 10 years by a factor of 2.

With so much more content, how are people processing and retaining it? Has that changed?

What is going on is that when people feel that they have an artificial memory aid, in order to be able to access the memory later, they actually do less work to encode a memory richly than they would if they don’t think they have a means of accessing it later. For the longest time, we only saw this in seniors. In seniors who have a life partner, someone who has been married to them for 50 years, they end up saying, “Oh, I don’t have to worry about it, because I know my wife well.” So, people do less work to encode a memory.

But nowadays, what we’re realizing is that people have the same type of reaction when they think, “Oh, I can search for it later on a search engine, so I don’t really have to remember the name of that movie, because I know I’ll be able to access it this way later.” That is a documented effect. People work less to encode a memory richly.

Does that also affect our attention span?

I don’t think our attention span ability — which would be the number of seconds we’re looking at the same device or piece of content — I don’t think what we’re able to do is changing. I think what we are doing is changing.

I’ve seen some really fascinating results from Nielsen. When you set someone in a room with nothing but a television, then, on average, their gaze is at the television for a minute or two before it goes somewhere else. But the moment you give them a tablet in their lap, now all of a sudden, the amount of time on average before their gaze shifts goes down to something like 7 or 10 seconds. You add in a phone and a friend, and, in that sense, the attention span of humans sitting in living rooms looking at TVs has gone down, but is circumstantial based on the other things they have to attend to.

I don’t believe their physiological ability to pay attention to something is changing. But we are shifting around, we’re juggling more content, we’re trying to determine what’s meaningful and what’s not.

What does this mean for a practitioner?

If you were working to minimize the cognitive load on people through your designs, imagine how much more you should be minimizing if you realize that it’s not just your device, your website, your app or your content that they’re interacting with, but it’s all the others in the same room. I believe that we should be continually designing around, minimizing, that cognitive effort people have to do.

What we were just talking about with Gestalt principles is one of the first and most effective ways to do that. Don’t make people think, just because you put the wrong color, or the wrong left or right orientation of a button. Keep people’s working memory capacity for the important stuff.

There are so many aspects that influence a person’s ability to maintain attention or memory. It leads to another bottleneck you write about, one of disposition. Would you recommend techniques such as contextual inquiry and direct user observation for understanding users’ needs to address that bottleneck?

Disposition is the word I’m using to refer to personality as well as your needs. Maslow’s hierarchy of needs, I am now a big fan of, and motivation and social influence.

Let’s start with Maslow’s needs. We’re familiar that, fundamentally, we have to fulfill physiological and security needs, and, after that, we turn to belongingness, then we start concerning ourselves with status, and finally, we focus on fulfillment, and fun and actualization.

This is a theory that a lot of folks in psychology have had a lot of fun with, to criticize. But I found myself, in the last few years, performing Big Data psychographic segmentation on a bunch of the activities, or a bunch of the stated wants and needs of users for a variety of different sites.

This is an advanced analytic technique, where you ask people, “What do you need?” You give them 150 options, then you use factor and cluster analysis, you find the fundamental needs fulfilled by product and, you know what, I kept finding Maslow’s same five needs, over and over. I found it working on cooking websites. I found it working on travel websites. I would go to talks on the fundamental needs fulfilled by mobile devices, by smartphones: Same five. When I came over to Microsoft, it’s like the same five. I came to be a believer in understanding what fulfilling the needs means, in that inductive way.

So why is it a bottleneck?

Think of it this way: Someone will not be interested in their ability to do social sharing on a smartphone — you know, take a picture, upload it to Facebook — if, at that moment, they are stuck on their roof during a hurricane and worried about the battery life before they can call 911 and get rescued by a helicopter. I know that’s a really extreme example, but this is why it’s a bottleneck. This is what Maslow proved. It was called pre-potency. We don’t concern ourselves with the higher-order needs until the lower-order needs are met.

For example, on AllRecipes, they would ask people who were using the site because they were diabetics and they all of a sudden had this diagnosis that they had to respond to with diet. It was a life-or-death matter. You ask them, “Would you like to share recipes with other chefs? Would you like wine pairings?” And the answer was, “No, absolutely not.” I really came to view this as a bottleneck.

Should you use contextual inquiry, which is where, if I can kind of boil it down, you interview and you ask questions, whereas in user observation, you make it a point not to interrupt them at all, just watch? I think that we’re trying to identify needs. A lot of times, you do have to ask people what their subjective experience is.

Which method have you used more often?

In my career, I’ve done a lot more contextual inquiry than I’ve done out-and-out observation. For example, we did a lot of retail research one year. We couldn’t interrupt a shopper, because the moment you did, their behavior was not natural. I think clients look to you to find the right method for their question, and that’s what we do. That’s the expertise we bring.

In an age of Big Data, we should never get to the point where we think we’ll no longer need to ask about people’s inner, subjective experience, because their needs, or their personality, just cannot be revealed without it.

You have a great example of this in play, so to speak?

I have to tell one anecdote that points this out so well. It involved a video game test on an adult-shooter game. They were testing the features by which you could play other members on the web. They brought in a participant and got him going, playing against others out on the web, but they didn’t have the audio set up. All they could do is see a screen. What was happening was that he would go out and, in two seconds, he would get killed in the game, then he’d go back out and get killed again. He was so bad that the other people he was playing against stopped shooting him, and they would just beat him to death with their hands. They were killing him in increasingly embarrassing ways. The game testers decided to stop the test, because, “Oh, this is just a complete fail. This guy has got to be miserable in there. Somebody go in there and save him, he’s the wrong person. We shouldn’t have recruited him. Let’s stop the test.” They walked into the testing room and he had tears on his cheeks, but he was laughing. He was cracking up, because he was saying, “I’ve got to get this game. That was the most fun I’ve had in so long. I’m terrible at this. They were owning me completely.”

Just watching his behavior, you wouldn’t have any idea that subjectively he was having a great time. It’s one of those examples where behavior reveals a lot, but it doesn’t reveal everything.

You’ve mentioned that the word “aligning” in your book’s title is really important to you. Why is that?

There is a large community of very thoughtful people who are working on the idea of habits — on good habits, but also, in some ways, habit-forming technology.

Any time you are working with the intersection of psychology and tech, you have to have a very high ethical bar. There are a lot of great communities that are articulating what those ethical standards are. I think that we’re very excited about what’s going to emerge on that.

A North Star in that is the word “aligning.” It’s the idea that your users come to you as human beings with needs. The book talks a lot about their goals. It talks about the time they spend with your product is time that they’re not spending on other things. Is your product helping them toward their own goals? The developmental section talks about their existential questions, for example. To the extent that we’re aligning interfaces to be more easily used by everyone — people with disabilities, aging populations — but also aligning with the standard psychological principles that we all show in perception and memory. What we’re doing is making interfaces better for people.

But you think the aligning goes even further than that?

It’s like aligning with what your users want to become. You create a product because it’s your aspiration, your passion or your dream to make something and bring it to market, and have it be a success. Your users have all of the same aspirations and dreams.

They’ve got to be very careful about what content they attend to and how they spend their time, and I think that’s why we have bottleneck, to make sure we stay focused on what matters to us.

In the end, that’s where I would lead people, is to say, “Work on aligning with people, instead of overwhelming or distracting.” In that way, you’ll always find yourself ethically on the right side of things, and also on the user side of things.

I wrote this book to be read by both, and understood by both, and then, ultimately, get more alignment.

Your book provides several principles that people can use in the design process, to objectively look at these aspects of human experience when interacting with software. It brings to mind something that projekt202 Chief Experience Officer Peter Eckert says: “Form follows empathy.” The more we have these kinds of things front-and-center in the design process allows us to empathize with our users and create better digital solutions.

Empathy is absolutely key. Your clients are lucky to have you with that orientation. All good comes from empathy, absolutely.




projekt202 is the leader in applying experience strategy and observational insights to the design and development of mobile, cloud, web and workplace software.