CX research: How design teams embrace human insight

At UserTesting, we believe everyone can make better, more confident decisions with access to human insight, no matter their job title, however, anyone in the business of doing CX research knows that’s easier said than done.

How to empower non-researchers to leverage CX research 

It helps to have a research team that can develop programs. Even if teams spend a lot of time working closely with the CX research team throughout the development process, they can’t help everyone with every project—nor do they want to. Often, there’s a desire to responsibly enable more people across all disciplines in an organization to conduct customer experience research on their own.

After all, product managers, designers, and marketers who want to be close to their prospects and customers should have access to insight at the speed of business today. They don’t want to submit a request to the research team and get a report back weeks later. They want to hear directly from customers themselves. This is why it’s good to create different support systems to help your colleagues:

• Know which research method to use
• Access existing data
• Recruit participants
• Get feedback
• Analyze the results

Related reading: How CX programs avoid getting defunded

An example of CX researchers and design working together

Recently, we went into a project knowing that it would be an iterative and ongoing process, so we used that to our advantage. We also knew that the designer would mostly conduct usability testing on static images or lightweight clickable prototypes. The general structure of the study—goals, participants, and tasks—would be similar each time.

We decided to pair a UX researcher with a designer, having them work super closely together on the first few projects, and then slowly transitioning more and more of the work from the researcher to the designer.

Initially, the researcher spent more effort, but as the designer learned how easy it is to get user feedback on every new design, they gradually took over and did the bulk of the work themself.

Related reading: Generative vs. evaluation research

The researcher created a test plan template that was filled out together with the designer:

• What are you hoping to learn? Example: “Is it clear that “...” leads to more options, and is it a good replacement for “Action dropdown?”
• Who do we want feedback from? Example: Existing active users and new users who’ve never used UserTesting.

Based on this, the researcher filled out the test plan with specific screeners, tasks, and questions. This test plan forms the basis for all future studies and is updated and tweaked based on the focus of the study.

For the first few studies, the researcher recruited participants, conducted the sessions, and discussed the results, but made sure to invite the designer to every session. As the designer gets the hang of it, they start to take over some of the sessions. During the training period, the researcher stays close to the analysis for the longest. This knowledge transfer is crucial, as they’d write their summaries and present them to each other. Knowing how to make sense of everything you’ve heard and seen, and making decisions based on that, is the hardest part of conducting customer experience research.

Remember that these were quick and dirty usability studies, with typically 2-3 participants each round and very focused goals. There were many opportunities for the designer to learn how to do this themself. They got good at testing specific components and flows. This type of research lends itself especially well to unmoderated testing as well. The template created was also set up as a template in UserTesting so that the designer could launch a new study whenever he needed to. When we research our product using our contributor network, we typically whitelabel the design and remove all references to “UserTesting.”

Scaling a researcher’s time

The training took 4-5 hours per week of a researcher’s time for two months. We’ve now settled into a partnership model towards the right of the graph above. The researcher maybe spends an hour total every month working with the designer. 

What design does themselves:

• Repurpose the previous test plan
• Conduct or launch sessions
• Make decisions based on findings

What research helps with:

• Choose methodology, if not the typical prototype test
• Review test plan if new
• Recruitment, if not from our contributor network
• Debrief on session and discussion of findings when requested

It’s worth the extra effort upfront

Design teams are a central part of the product development process, so the process must work well and scale in a way that enables more decisions to be made with customer insight. 

By giving researchers time to train others, we reap the benefits in the long term. Now, we don’t add any new patterns or updates before testing it.