When Research Became a Point of View: How UX Researchers Can Use AI to Build Influence

For a while, every conversation about AI in research came back to the same thing: speed. How fast can we recruit? How quickly can we synthesize? How many interviews can we turn around before the sprint ends?

Those questions still matter. But for research teams that have moved past the experimentation phase, a more interesting question has taken over: what do you do with all that time you just got back?

That's what Figma's Christa Simon and Peggy Zhang got into during their AMA with Rally, and their answers are worth sitting with. (You can watch the full AMA here.)

The Recommendation Is The Deliverable

"Before, people were asking, 'Where do I find a report related to a topic I care about?' Now people are asking, 'What is the most important insight that surfaced from that?'" — Peggy Zhang

A lot of research organizations were built around a familiar rhythm: gather evidence, present findings, and let stakeholders decide what happens next. That model made sense when decisions moved slowly enough for everyone to interpret findings on their own timeline.

Product teams aren't operating that way anymore. AI is accelerating development cycles, and the researchers who are gaining ground are the ones who have gotten comfortable leading with a conclusion and supporting it with evidence, rather than presenting evidence and waiting for someone else to draw the line.

That's a different kind of ownership than most research functions were designed for, and it's worth naming clearly: it's an opinion. A trained, evidence-backed, rigorously considered opinion, but an opinion. Researchers often have visibility into patterns that no individual stakeholder can see, sitting at the intersection of multiple teams, product areas, and initiatives. Asking leaders to connect those dots themselves means asking them to do the hardest and most valuable part of a researcher's job.

A recommendation gives the conversation somewhere to start. Stakeholders can agree, push back, or refine it. What matters is that things move.

Relationships Are Still The Job

"The job of a researcher is like 80% relationship building. I don't think that changes, or gets replaced by AI." — Christa Simon

Here's the part that doesn't show up in the AI productivity discourse as often as it should: research has always lived inside a web of relationships, and that's truer now more than ever.

Researchers move horizontally across an organization in a way most functions don't. They see recurring customer needs across products, recurring friction across workflows, recurring assumptions across teams. That perspective is truly hard to replicate, and it doesn't get automated away when synthesis gets faster.

The shift is in how you think about it. Less about delivering research, more about architecting how insights travel: across teams, over time, and into the moment a decision is actually being made. That's increasingly what defines what a research function is worth.

AI Raises The Bar On Judgment, Not Just Output

"AI doesn't replace research judgment; it raises the bar on it." — Christa Simon

Stakeholders now have faster access to research than ever before. They can search repositories, summarize findings, and ask questions directly of the accumulated data. Which sounds great, until you realize it means the room you're walking into already has opinions formed from partial information.

Separating signal from noise has always been a core research skill, and now it's the most visible one. Why does this finding matter more than that one? How does this observation connect to what we already know? What's the difference between something interesting and something consequential? Those are judgment calls, and they're the ones that compound into influence over time.

The faster the tools get, the more that judgment becomes the thing people are coming to you for.

Infrastructure Is What Makes Influence Scale

"Initially, I thought we're going to have all these AI capabilities, it's going to be so good, we're going to save so much time, and maybe we just don't have anything else left to do. But what surprised me is there's so much still left to do." — Peggy Zhang

Individual productivity gains are real and worth celebrating. Organizational impact is a different challenge.

The research teams seeing the most meaningful returns from AI are often the ones that have already invested in shared foundations: common workflows, accessible repositories, consistent standards, clear pathways for knowledge sharing. Those systems create compounding value because they make research reusable, and reusable research is the thing that actually scales.

Here's the part that catches teams off guard: when AI makes research faster, more research happens. (Look up the Jevons Paradox if you want a rabbit hole.) The question becomes whether your organization can absorb, distribute, and act on the volume of insight that AI now makes possible. Without infrastructure, speed creates more outputs. With it, speed creates more reach.

Research Still Owns The Hardest Part

Research has always been about helping organizations understand reality well enough to make better decisions. AI changes the mechanics of that work without changing the responsibility.

If AI handles more of the scheduling, coordination, synthesis, and retrieval, what's left is the part that was always the most important anyway: building trust, connecting ideas, and helping product teams understand what actually matters.

The research functions that build lasting influence won't be the ones producing the most insights. They'll be the ones that can turn insights into conviction, conviction into alignment, and alignment into better decisions.

That's always been the job; It's just more visible now.

Want to dig deeper into how AI has shifted organizational influence into a core research skill? Watch the full Figma AMA.

If you're ready to build the research infrastructure that makes AI acceleration actually stick, request your personalized Rally demo today.