How I Conducted My First Remote KJ Activity

Trello is a versatile tool.

Last year during a project, I learned to conduct a KJ activity with a remote team. Since then, I’ve led other KJ meetings with great success and have continued to iterate on my technique.

You May Be Wondering

What is Trello?

Trello is a collaboration tool for both co-located and remote teams. Many teams use Trello to manage their projects and tasks.

What is the KJ Method?

Also called an affinity diagram, the KJ method is a technique for reaching consensus or determining priorities as a team. It is a way to gather team members’ input and prioritize it objectively.

“KJ” stands for Jiro Kawakita, the creator of the technique (the initials are in reverse order because the Japanese use their last names first).

Project Background

My last team project at Center Centre involved learning a lot about developers and how they learn tools like frameworks and toolkits.

ATS (Applied Technical Systems) commissioned the project in conjunction with their client, a U.S. federal agency. The goal was to create a prototype of a developer documentation site. We had to create the prototype from scratch because the product was still in development.

I had many roles during this project. My first role was to plan our very first user research study. The goal of the study was to learn about the problems developers face when evaluating and learning documentation sites.

I know from experience that more heads are better than one, so I set out to plan an activity with my team to find out what we needed to know.

The Challenge

Our client, whom we collaborated with remotely, wanted to join us for the activity. The activity needed to be remote and accessible so our client could contribute.

The Solution

A fellow teammate and classmate, Randal Flamm, facilitated an in-person KJ using Trello at Center Centre just weeks before.

I decided to use Trello because of his success.

The Activity

Step 1: Planning

I crafted an agenda, scheduled 30 minutes for the brainstorming activity meeting, and invited all participants (my team and our client).

Step 2: Creating the board

I created a Trello board (below) and gave it five sections (or “lists”):

• Questions
• Screener
• User Interview
• Both?
• Other research outside of interviews

Trello KJ: “What we want to learn from developers”

First, we had to determine what we wanted to learn about developers. I asked participants to write down questions they wanted to ask them in the first list, “Questions.”

After the ideation phase of the activity, I asked participants to sort the questions into the other four categories:

• “Screener” for questions that would help us learn about the user research interview candidates.
• “User Interview” for questions we should ask during the user research interview.
• “Both” for questions we could ask either in the screener or during the user research interview.
• “Other research outside of interviews” if team members came up with ideas for ways to do user research outside of interviews.

Step 3: Discussion

After sorting our ideas, we voted on what we thought were the most important things for us to learn during the user interviews. I gave everyone an equal number of votes.

The votes framed the discussion at the end of the activity.

We talked about the meaning and reasons behind the most voted for ideas. Starting the discussion in this way let us talk about things we knew we needed to learn in the study.

Results

The KJ activity took only 30 minutes. In that time, I was able to gather question ideas, create a shared understanding of the problem amongst my team, and learn about a secondary user.

Using the questions my teammates created in the KJ, I created open-ended questions for the study’s moderator/interview guide.

We also learned about our assumptions during the KJ. For example, before the activity, we assumed our only users were developers. Well, we discovered we had a second audience: decision makers. These decision makers were usually managers or product owners. They contributed to the developers’ decision about what tools to use or purchase.

Reflection

During the voting part of the activity, I instructed participants to pick a sticker from the menu and use it to vote. This sounds simple enough, right?

The problem I ran into was that multiple people used the same sticker. Because of this, people couldn’t keep track of what they had voted for or how many times they had voted.

The simplicity actually made things confusing for the participants.

The next time I tried a KJ via a Trello board I created a legend. I gave everyone their own, unique sticker. It worked much better. And the team I facilitated the activity for had fun using the stickers!

Trello board for Collider’s social media strategy KJ

I Continue to Use this Technique

I facilitated another KJ a few months later for a startup in Chattanooga called Collider. The KJ required multiple ideation/brainstorming rounds. Collider only had a few post-it notes — not enough for the entire team — and they did not have enough free wall space for the whole team to use.

I decided to use Trello again for that meeting, and it was a great decision.

The Collider team preparing for the activity

The meeting went as planned, and participants were able to fully participate and engage in the discussion. The activity even ended a few minutes early. A few of the team members thanked me afterward and told me I was great at running meetings. This was very exciting to hear!

Through experimentation and iteration, I’ve learned how to lead effective group activities when one or more team members are remote. I look forward to applying what I’ve learned in future meetings and learning more so I can continue to improve.

Want to read more on how to do a KJ activity yourself? Check out this article by Jared Spool:

The KJ-Technique: A Group Process for Establishing Priorities