UX design in the real world – Q&A with UX expert Larry Marine

October 23, 2017

How the psychology of UX design affects us in the real world, from Valentine’s Day to consumer air travel to the United States Air Force

Cognitive Scientist and UX Research and Design consultant, Larry Marine knows the impact of UX in the world around us. With almost 30 years of experience in the field of UX, Larry has had the opportunity to impart user need-oriented design approaches to clients as well as students in the US Air Force Academy, where he taught Behavioral Science undergrads about the importance of UX in complex systems.

This week Justinmind caught up with Larry to pick his brain about all things UX, from his own UX ‘Ah-Ha’ moment and fascinating user-centered design success stories featuring many a prototype, to the psychology of UX design and some top tips on analytics tool design. Read on for all the best bits in our interview!

Hi Larry! To kick this off, let’s talk about your experience in the field of UX. How did you get started? When and what was your ‘Ah-Ha’ moment?

I was a computer systems technician for the Navy and saw how most folks struggled with technology, in general, and computers, specifically. I started college with a double major in Psychology and Computer Science to find ways to make technology more approachable for the common user. Then I found out about Dr. Norman’s new Cognitive Science program at UC San Diego and transferred into it. Cognitive Science is a blend of several computer sciences and Cognitive Psychology and focuses on redesigning technology for the users. This is where I developed my well-known quote: It’s far easier to adapt the technology to the user than to force the user to adapt to the technology.

An early Ah-Ha moment came when I was in college standing in line at a cash machine. I waited about 6 feet directly back from a guy using the machine. I thought, why did I choose to stand right here? Certainly to offer some privacy, but why at this very spot? Then I noticed a crack in the sidewalk. I observed this spot, and then many others and realized that humans subconsciously use structure in the environment to organize behaviors. That crack, or planter, or awning was never intended to guide our behaviors, but we naturally used it to our advantage. That’s when I realized that we can selectively create stimuli in the world to incite desired user behaviors. We can control people’s actions by what we show them.


What’s an average day like for you as a UX Research and Design consultant?

“An unfortunate side-effect of being a Cognitive Scientist is being constantly reminded of how poorly designed our world is.”

I don’t have an average day. I could be doing any number of things, such as researching Coast Guard rescue operations in Norway, usability testing a medical device in San Francisco or conducting a website UX design seminar in Argentina.

I find time to read the latest discussions on UX design and even respond with a bit of my own (mis)perceptions. I find that writing improves my understanding of new ideas better than just reading about them. It also helps identify new blog topics.

I also try to find time to provide feedback on designs. I offer a free design review for anyone that asks. It’s a great way to keep up with design trends and business models.

When you accidentally smack your thumb with a hammer, you are reminded of how much you use that thumb for about a week. As a Cognitive Scientist, I’ve developed a cognitive sore thumb and suffer the mental burdens of bad design every day.

Download Justinmind today and integrate your UX design and research processes

Download free

What’s your preferred UX research process?

Given that most of my work involves identifying new, disruptive designs, I rely heavily on observational research. As they say, what users do and say are two completely different things. If you need new ideas, watch users, don’t ask them.

I think most researchers only practice a couple of methods and rely on them exclusively, regardless of the situation. An incorrect research method yields inaccurate data, resulting in inadequate designs. You end up solving the wrong problem, very well.

I’m currently writing a book on user research methods that helps identify which of the various methods to use given a project’s objectives. For instance, if I were focusing on polishing an existing website, usability testing would be an appropriate research method, but usability testing is terrible at finding new disruptive design opportunities.

Learn about usability testing with Justinmind here.

Tell us one of your user-centered design success stories.

The executives at Proflowers.com wanted a design that specifically did not look or act like any other florist website. They suggested a number of competitor sites we could test, but instead, we conducted observations at actual flower shops. We quickly noticed that most of the shoppers were men buying flowers for women. What do men know about flowers? Roses, and that’s pretty much it. So why did every website at the time require shoppers to build a bouquet? Because, that’s how florists thought of the problem, that shoppers wanted bouquets.

In reality, men didn’t ask for any specific type of flower. They told the clerk they needed a bouquet because they forgot their wife’s birthday. Men didn’t care which flowers they bought, they just wanted to avoid buying the wrong ones for that occasion. So, our design approach focused on organizing bouquets by occasion rather than flower type. This eliminated the requirement of knowing which bouquet was right for which occasion.

The site was launched for Valentine’s Day, 1998. A few weeks later, the Proflowers execs called and wanted their money back, saying the site failed miserably. We reviewed their site and noticed that the developers had taken some ‘liberties’ with our design. They justified it by saying that our design didn’t look anything like the other sites and they ‘improved’ our design for us. The execs told them to follow our designs and rebuild it before Mother’s Day.

They have used that same design model ever since and Proflowers has been a top 10 conversion rate site every month for 20 years. (They never asked for their money back, BTW.) This story also proves that good user-centered design is more successful than the average design.

Download Justinmind today and boost your user-centered design process

Download free

When does prototyping come into the UX design process for you?

My design projects are separated into two phases: 1) Accurately identifying the problem and 2) Solving it.

Once we’ve completed our research and the priority matrix, we get started on prototyping. The research identifies the problems; the matrix prioritizes which ones to solve. I developed our custom priority matrix to quantitatively balance the needs and capabilities of the company with the objectives of the users. This matrix identifies new, unmet user needs, focuses our design efforts on the MVP items, and helps us avoid wasting time on lower priority features.

I start by sketching a few ideas on paper and then create more complete wireframes. I’ll add some color or graphics, but I avoid making it too pretty since that often distracts the users when I test the design. Once I feel the design achieves the desired metrics, I pass it off to the visual designers and developers. We collaborate on the design to make sure everything fits the visual language and the technology framework and conduct any more testing, as needed. I’ve found that low-fidelity testing identifies about 95% of the problems and more testing isn’t usually necessary.

We loved your article about improving the UX of Airline Bump policies. Would you say the psychology of UX is something that can be formally taught, or is it a ‘you’ve either got it or you don’t type situation?

[bctt tweet=”The psychology of users is based on understanding how the brain works, how it responds to stimuli.” username=”just_in_mind”]

It really is both, and you won’t know if you have it until it’s been taught to you. There are user behaviors that defy common expectations and, without formal training, you won’t know about these. For instance, a lot of designs rely on some type of instructions or training to solve interaction problems. Any design that relies on instructions is inherently flawed. Instructions are a crutch, not a solution.

You’ve taught how to apply UX design to real-world problems to the Behavioral Science students in the US Air Force Academy. What business do usability and human-centered design principles have in the military?

These young officers will be responsible for setting requirements for new and better systems in the future. Military training has been highly regarded for its success in managing user behaviors. Until now. We’ve reached a point where technology is too complex to rely on training as a solution. Poorly designed technology becomes a cognitive burden rather than a solution, increasing the potential for human error and diverting attention away from the problem at hand, disrupting situational awareness. New solutions must leverage human capabilities and avoid our limitations.

Moreover, this research helps identify new approaches to solving old problems. So much of today’s defense industry is still solving old problems, the old way. They design new technologies that merely automate current frustrations.

You’re big on analytics tool design. For you, what’s essential to successful dashboard design?

Typical website analytics tools and dashboards are useless failures. They provide data and information, but lack any insight, and certainly don’t incite any actions. Website analytics are like footprints in the sand, they tell you where someone went, but not who or why. Moreover, the analytics don’t tell you what’s wrong or how to fix it. If they were any good at all, there wouldn’t be so many lackluster website designs.

If a dashboard doesn’t incite an appropriate action, then it’s just eye-candy. It looks good without providing any real value. Most designers create dashboards to use the data they have rather than identifying what data they really need to incite actions. The key to achieving this is to work backwards. Begin at the end and identify what actions the users need to take. Then determine what insights define those actions, what information leads to this insights, and what data you will need. You’ll find that you don’t have the right data and must design a way to capture it in order to create an actionable dashboard.

Want tips on creating a social media dashboard? Find them here!

Emily Grace Adiseshiah
Emily is Marketing Content Editor at Justinmind