Summary
“Research democratization” has become a rallying cry in many organizations — but also a source of tension. Should everyone be able to run studies, or should research stay tightly centralized? The truth is, both extremes create waste, confusion, and risk. In this talk, I’ll share a practical framework for when to democratize and when to centralize research. You’ll learn how to weigh risk, reversibility, and audience impact; how to put guardrails around democratized research; and how AI can enable broader participation without chaos. You’ll leave with a decision matrix and a “democratization charter” you can adapt to your org, so the right people are on the right horses for the right courses.
Key Insights
-
•
Democratization of research exists on a spectrum and must be customized to each organization's context, risk tolerance, and regulatory environment.
-
•
Not all research tasks should be democratized; high-risk or unclear problems require specialist researcher involvement.
-
•
AI tools are enabling broader roles for PMs and designers, allowing faster prototyping and product documentation, blurring specialist boundaries.
-
•
Effective democratization requires elegant guardrails to protect participant safety, data privacy, and research quality.
-
•
Democratization can massively increase organizational research capacity, as Flight Center’s example shows, tripling volume while maintaining quality.
-
•
Training and certification are critical to enable non-researchers to run low-risk studies effectively and ethically.
-
•
Bias awareness and anti-bias training are essential components of democratization to maintain the integrity of insights.
-
•
A democratization charter outlining who can conduct research, with which audiences and methods, helps set clear expectations and governance.
-
•
Research maturity assessment helps determine an organization's readiness for democratization and identifies needed upskilling.
-
•
Research quality control currently relies on script review and coaching but is evolving toward AI-supported automatic quality feedback.
Notable Quotes
"Democratization is here to stay, and it’s a positive force for good in product development."
"The democratization spectrum ranges from everyone accessing feedback to a free-for-all, but I don’t advocate for a free-for-all."
"We need elegant guardrails to keep the business safe, keep participants safe, and make democratization ethical."
"Flight Center went from 10 researchers running research to over 130 people trained to safely do interviews."
"When done right, democratization multiplies the impact of research rather than diluting it."
"The goal is to lower the floor for who can access research while raising the ceiling of what’s possible."
"It should be as easy to run research as brushing your teeth."
"AI will soon enable us to automatically flag leading questions and coach researchers for quality."
"Not every problem needs research; sometimes it’s fine to ship quickly and learn after."
"Training non-researchers in bias awareness and methodology is essential to prevent poor data and insights."
Or choose a question:
More Videos
"Design is not really about problem solving; it’s about cultural imagination."
Dan HillDesigning for the infrastructures of everyday life
June 4, 2024
"Simplicity rarely loses to complexity in the public square—it’s about meeting people where they are."
Steve ChaparroBringing Into Alignment Brand, Culture and Space
August 13, 2020
"Only 38% of people actually looked at the company review information, even though it was crucial."
Jeff Ephraim Bander Ariane Rahn Philipp ReiterEye Tracking Gamechanger: Why Smartphone Eye Tracking will Revolutionize Your UX Research
March 11, 2022
"If you’re like us, you’d love to have 100% of the UX talent every project needs, but your team isn’t large enough."
Patrick CommarfordDesign Staffing for Impact
January 8, 2024
"If you don’t have any work, make some spec work just to show your skills."
Dave Hoffer Joanne WeaverUX Job Search AMA #2 with Joanne Weaver and Dave Hoffer
April 3, 2025
"I was no longer a point person in recruiting or shaping the team, but that was okay."
Edward CuppsThe Principal Path: Journeying from Management to Individual Contributor
June 11, 2021
"Our edge isn’t in managing what exists; it’s in architecting new meaning at scale where process becomes infrastructure."
Aletheia DelivreNew Shapes and Emerging Identities for Design Ops
September 11, 2025
"Overlap between design leadership and ops is context based; pragmatic collaboration gets things moving faster."
Tim ParmeeChanging Our Design Pressure Points
October 2, 2023
"A design program manager is like air traffic control making sure all the planes land in the right place."
Brennan HartichCommunicating and Establishing DesignOps as a New Function
November 7, 2018
