Applying Design Thinking to LXD: A DevLearn Case Study
When preparing to speak last week on applying design thinking techniques to identifying performance issues, I found myself applying those same techniques to designing a learning experience. I wanted to capture my process, iterations and all, to show my work and hopefully help extend the session. I hope this example serves as a general template of how learning experience design (LXD) and design thinking can intermingle.
I’m going to use the stages of the d.school 5-step design thinking process as a framework here, but just know that I could have shoehorned my HCD process into a number of other frameworks or headings.
Empathize
I’ve led some conversations within the L&D world on HCD, and during those conversations I’ve tried to get a sense of what gaps people see in their own understanding. By asking people to fill out an empathy map or simply asking in the chat, I was able to casually build a persona in my head.
Based on this informal research, I expected that the average session attendee was familiar with the basic concept of design thinking (or had at least heard the term thrown around) but wasn’t sure:
- how to get started,
- whether it would be an accepted practice in their organization, or
- whether it even made sense to try it out.
Define
Based on what I knew about the probable session participants, I knew that I wanted to provide a space for practice with a real-life scenario. Of course, what’s a problem without constraints? I had a basic idea of mine at this point.
- The audience could be any size, from 50 to 200.
- The problem statement needed to be described quickly, universally understood, and complex enough to be interesting.
- I couldn’t pull my friends away from their own conference experiences to make them act as facilitators during my session, so I wanted to be able to run the session as one big group rather than the traditional smaller groups.*
*We added this constraint after talking to John Horvatis, Motivf’s Senior partner, who made the point that the holy grail would be to simulate the intimacy of a smaller group regardless of the session size.
Ideate
Knowing the problem, I pretty quickly converged on how best to demonstrate some key concepts (the importance of varied perspectives, the difference between “yes, and” and “no, because/no, but”, and the value of sketching to align thinking.
These align well with the major “buckets” that we use to summarize HCD. Some discussion and feedback from my Motivf colleagues brought the idea from small groups to a large-group exercise. From here, I came up with a plan. I would briefly describe HCD to bring everyone up to a baseline understanding, then present a case study (I picked the decline of Circuit City, a wicked problem that everyone likely had some familiarity with).
- Provide four personas based on a case study, asking each person to take on one persona for the activities.
- Give a broad request based on a surface-level understanding of a problem statement (“design a course”).
- Ask the participants to, individually, come up with How Might We (HMW) statements and submit them using an online whiteboard (MURAL, my favorite).
- Affinitize the HMW statements as a group and ask the participants to work together among their immediate seatmates to reframe the original request with something that addresses the better-understood problems. Those re-framed requests would be added to the same MURAL board.
- Sketch what that might look like and share out if time permits.
Prototype
Perfect! I set up a resources page that would hold the personas and links to the MURAL. I knew what this session would look like, and I was ready.

But wait! As I sat in sessions and thought about my prototype, I started to think about the seating arrangements and the technology I saw people use.
Empathize (Again)
Some people had laptops, although they were mostly people working on answering emails during sessions (in other words, probably not the people who were going to actively participate anyway). Everyone else was holding mobile phones to take notes. I expected tablets, but phones was a surprise. MURAL isn’t that easy to use on a phone.
Define (Again)
I added a new constraint. Now my session goal was to:
Provide a space for a large group of learning professionals to practice applying design thinking techniques to a real-life, quickly understood scenario in a way that provides transparency into the process between all participants and can be accessed on a mobile phone.
Keeping the problem definition in mind was critical. It ensured that I didn’t veer off track, trading important pieces of my goal for convenience. It would have been easier to go back to small groups, but the ability to see into everyone’s process would have meant that some people would leave with insufficient, or even incorrect, practice. I could have simulated the whole thing by asking for volunteers, but I wanted everyone to have an extremely low barrier to participation. If you didn’t leave having written your own HMW statement, I couldn’t call it a success.
Ideate (Again)
I thought about my options. A Google Doc would be visible, but messy. I thought of using Zapier or another tool to connect a Google Form submission to MURAL to post stickies based on submissions. This was, as I learned, a thing that was only possible in my head and no one had created an integration for such a thing.
I knew that Poll Everywhere was a good option (h/t Cara North), but I kept thinking about Step 4:
Affinitize the HMW statements as a group and ask the participants to work together among their immediate seatmates to reframe the original request with something that addresses the better-understood problems. Those re-framed requests would be added to the same MURAL board.
Affinitizing the HMW statements, grouping them by categories in the moment, would be a key to helping people identify trends. I didn’t have an easy way to do this on the fly in Poll Everywhere, so I would have to adjust this piece.
The other thing that mobile phones would affect was the ability of participants to refer to personas while they submitted responses. Two tabs on a laptop is one thing, but multiple tabs on a mobile phone can get unwieldy. I needed to hand out the personas as paper artifacts so they could be referenced as people responded. Otherwise, we risked forgetting about the perspective we were representing.
Prototype (Again)
By this point, it was Thursday night. I presented Friday morning. I ran my idea by my very patient friends one more time (maybe two or three…) and updated my resources page.
Testing
It worked! I started the session with a poll asking what people hoped to get out of the session, and the results helped me shape the presentation on the fly. I was able to quickly get everyone on the same page regarding design thinking, clear up any misconceptions, and start getting audience participation quickly.

How Might We?
After talking about the personas for a bit, I played the part of the stakeholder in a meeting and asked everyone to create a 30-minute training for all of the new employees to refresh them on product knowledge. The participants played the part of the instructional designers assigned that task hoping to, well, get out of doing it.
I boiled the company’s need down to its most basic element: increasing revenue. With that prompt, I asked participants to submit HMW statements to address the issue. The catch was that they were each assigned a persona and they needed to create the HMW statements from that person’s perspective. How would a new employee, or a seasoned employee, or a store manager want to see revenue increased?
We got 116 statements with different ideas on what could be done. These included structural changes (“incentivize coaching”), microlearning opportunities (“make a series of mini videos for each product that can be accessed by mobile”), mentorship programs (“partner new hires with seasoned employees”), and more (“create better signage on products so employees don’t have to remember everything”).

Rather than affinitizing, I had people vote using Poll Everywhere. This had an interesting bonus. Participants, as I expected, started voting before everyone had finished putting in their statements. This was a teaching moment — of course when you see a voting button you want to vote. We want to start quantifying and judging ideas immediately, and I got to call out that impulse and bat it down until voting was open.
Brainstorming
It was clear from the previous step that in-store performance support tools (for employees and customers, which was an interesting addition) and mentorship programs were the most interesting opportunities. Using these HMW statements, we re-framed that original request to build a 30-minute course. We got a wide range of ideas (36 in total):
- Create a desk reference guide
- Create a mentor/mentee partnership
- Create a mock-up for an easy to use performance support app [ed. note, I especially love this one because it’s a request to create a prototype, not the app itself!]
- Create product flip cards
- Create effective signage
- Get Sophia the Robot to explain products
- Deploy walkie talkies
- Create tent cards with product specs
Prototyping
I gave everyone the opportunity to sketch, but at this point we got embroiled in the final Q&A so they mostly got out of drawing. I wished we’d had another hour to flesh out some top ideas, but I left happy with what we’d created in the session together.
Resources
Want to play along at home? You can access all of the resources that I shared in the session here!
Slides
Personas
Larry, Director of Store Operations

