Listening Before You Intervene

Surveys, usability tests, A/B tests: they're validation instruments. They tell you whether your hypothesis holds, and they're excellent when you've correctly identified the problem. The trouble is that most research is commissioned before you know what the problem actually is, and the instruments themselves foreclose the questions you haven't thought to ask yet.

Listening tours are designed to surface the questions you should have been asking. Structurally, a tour is a series of open-ended conversations with people at multiple positions within a system: practitioners, administrators, people in adjacent roles, people who've moved through the system and left it, and end users themselves. Each conversation is loosely structured around a few open questions, and the method's discipline comes not from the questions themselves but from who you talk to and how you synthesize what you hear. The goal is to understand the system's internal logic, the way the people inside it actually run it, before you try to change anything.

Listening tours are most useful at the beginning of an intervention — when you're entering a new system, when the data doesn't explain the gap between design and operation, or when the people who will eventually carry out a change need to be part of building the understanding that motivates it.

They're a diagnostic for organizational and systemic dysfunction — the kind that doesn't show up in analytics or satisfaction scores because the people inside the system have adapted around it without naming it. Feature-level user research and discovery sprints assume you already know the problem space. A listening tour assumes you don't.

A client, an information design studio, was experiencing friction they couldn't name. Output was fine, client satisfaction was fine, but something was off in how the team worked together. Leadership had a theory about process, another about workload, and a third about team chemistry, and they'd been rotating through half-interventions for about a year.

Rather than entering with a recommendation, I ran a listening tour: 45-minute conversations with each person on the team plus three past clients, structured around two or three open questions about what the work felt like at its best and where energy went to waste.

The synthesis revealed something none of the theories had named. The studio's implicit hierarchy didn't match its stated values. Senior designers were systematically protecting junior team members from direct client relationships — handling the hard conversations, absorbing ambiguous feedback, running interference before briefs reached the people doing the work. Internally this was framed as mentorship. The junior designers experienced it as exclusion — a signal that they weren't trusted, that they were being managed rather than developed.

That finding shaped a completely different intervention than anyone had initially proposed. Underneath the theories about process and workload sat an unacknowledged belief about who was ready for what, and the tour made it possible to name that belief without anyone having to be the one who said it first.

Before designing primary research for a major airline's accessibility services — a multi-surface system touching booking, ground operations, in-flight experience, and post-trip resolution — I ran a pre-research listening tour with disability justice advocates to understand the political and systemic context before designing a single interview guide.

These conversations were intelligence gathering. I wanted to understand what advocates and community organizers already knew: where previous accessibility research efforts had failed to produce change, what the community's actual priorities were versus what organizations typically assumed they were, and where the system had historically broken down in ways that didn't appear in service data. The conversations were explicitly framed as informing my research design rather than generating research findings. That framing mattered; it invited a different kind of honesty.

What I learned shaped everything about the subsequent study: who I recruited, what I chose not to ask (because it had already been asked to no effect), and what the research needed to learn in order to be useful rather than merely interesting.

The sampling strategy in a listening tour is organized around position in the system rather than demographic profile or user segment. A front-line service worker, a policy administrator, a long-time participant, someone who recently stopped engaging, an adjacent advocate — these people have more than different experiences of the same system. They have structurally different views of it, because each position in a system only makes certain things visible.

They don't produce the kind of evidence that justifies a specific design decision. They're not evaluative, and they're not meant to be.

The output of a well-run listening tour is a more precise problem statement and a better-designed subsequent study. That's its value: it resets the research agenda before you've locked in the wrong one. The discipline is accepting that the method's deliverable is understanding.