Distributed research is now a primary way (if not the #1 way) that organizations collect user signal. 

In UserInterviews “State of User Research 2025,” 71% of respondents said their organization has people who do research (PwDR).

Every team needs feedback, and these days, they’re not afraid to get it themselves. With AI, anyone can draft, run, and analyze their own study. Being a gatekeeper won’t stop them from talking to customers directly.

But as we know, a lot gets lost when everyone talks to users without any coordination. 

One call with one vocal customer isn't a pattern across a hundred quiet ones. A five-minute conversation captures whatever someone remembers to say, not what they actually did. With more information than ever, the result is fractured signal, scattered everywhere. 

None of the data points are wrong, exactly. It just rarely adds up to something a team can act on, and it’s harder for someone less experienced to distinguish a one-off from a larger issue.

When anyone can collect data, the real value of the researcher isn’t running every study; it's knowing how to make sense of it all and ultimately steer the team in the right direction. 

We keep landing on bad solutions

Every time this comes up, the traditional tools we've had have forced us into a binary. Chaos, or crackdown.

Chaos looks like no limits: anyone can talk to anyone, run anything, and ship findings before they've asked the right people the right questions. 

Crackdown looks like the opposite: heavy training requirements, complicated systems, and a research function that feels precious, restrictive, and slow. 

Both are understandable paths. Neither one works.

At the end of the day, researchers often feel like they’re babysitting, while PMs and designers feel like they’re being blocked and slowed down

Every product team I've talked to wants the same outcome: more good decisions, made faster, by people close to the problem. But what previously held us back was inflexibility. We relied on tools that couldn't flex to fit the moment, so teams had to pick a lane and live with its downsides.

AI opens up a door we couldn't access before

For the first time, we can build flexibility with guardrails, guidance without a training requirement, and deep customization without a heavy setup lift. Those combinations simply weren't possible before. Now they are, thanks to AI.

What does this look like in practice? 

• Expertise lives in the tool, not just in someone's head. Methodology, sampling, and incentive choices come with built-in guidance, so the person asking the question doesn't have to be a research expert to ask it well. This expertise is programmed by an expert researcher, not just AI guessing.
• Policy gets built into the workflow, not bolted on after the fact. The right participants and the right use of budget and tools should be the default path, not a rule someone has to remember to enforce.
• Access is seamless and in-context. No one should have to learn a complex new tool to get an answer. The work should happen where teams already are.
• A person is still in the loop, actively. Think quality oversight the way a strong engineering team thinks about code review: someone experienced can look at a finding and say, "yes, that's a good one," before it goes anywhere important.
• Leadership gets a bird's-eye view. Leaders have to be able to see what's happening across the whole org, not just their own corner of it.

That’s how we get guidance, flexibility, speed, and quality, all at once.

How to actually implement it

For this to work, teams need the right infrastructure. That’s what we’ve been building for the past 8 months.

In Dscout AI Studio, AI-guided templates put expertise into the tool itself. Researchers lock in screener logic, study flow, brand standards, and compliance requirements before a single collaborator opens the tool—expertise programmed by researchers with more than a decade of experience, not inferred by AI on the fly.

From there, collaborators just chat with AI to build a study or find an answer in their results. Or they ask AI to run the study for them. No new tool to learn, no research degree required—just guidance built into the workflow, exactly where policy needs to live.

Research teams stay in the loop, as the directors (not the babysitters). They set review checkpoints where a second look matters, cap incentive budgets, add guardrails, enforce company policy, and see what's happening across every study in the org, not just a small corner of it.

This work helps ensure that research and the voice of the customer actually stay at the center of the organization, even as teams build faster and faster.

Research teams at leading brands like LinkedIn, Vodafone, Peloton, and more have already expanded access to insights across their product org using this structure. We’ll be sharing their stories in the coming months.

The past asked us to choose between access and rigor. We don’t have to choose anymore.