Gregg Bernstein has had many careers. Today he leads User Research for Condé Nast—a media company that brings us The New Yorker, Bon Appetit, Vanity Fair and Vogue.
But Gregg’s storied career has taken him other places as well—from Senior Director of User Research at Vox Media, to designing album covers for punk bands, a Design Professor at Georgia State, and most recently the author of the book, Research Practice.
As someone who has led and built research teams across MailChimp, Vox, and Condé Nast, Gregg has had a lot of practice with one of the industry’s favorite buzzwords—democratization. We sat down with him to chat about what democratization even means, how to enable it, and why.
What does research democratization mean?
‘Democratization’ is a word that’s often thrown around, but not often defined. So that’s the first thing we asked Gregg—what does democratization even mean? And he had a great definition: democratization means breaking down the barriers between people and the information that will help them make good decisions.
This can come in multiple flavors and forms, but there are two big ways you can enable democratization in your organization:
- The first step in democratization is enabling the organization to discover the insights and information they need to make the best decisions. Access to interview transcripts, survey results, analytics dashboards that people across product, design, marketing, or any other part of the company can discover and trust. Oftentimes, that gets teams 90% of the way to the answers they need.
- Once teams have access to answers that already exist, they can take the next step of solving for questions that remain unanswered. Gregg is enabling that at Condé Nast today by training teams to run their own surveys and interviews, and to do them with the rigor that research deserves.
Why does research democratization matter?
While researching at MailChimp Gregg began collaborating with the company’s Chief Data Scientist, learning more about survey methodology & discovering discrepancies between the qualitative data he was gathering and the quantitative data Data Science had access to. They soon discovered that they had a symbiotic relationship—Data Science and research findings brought together could paint a picture that neither could alone.
The depth of qualitative insights helped Data Scientists interpret their results and build better models, while Data Science’s findings gave scale to User Researchers—allowing them to make a compelling case and cross-check their learnings. Their relationship was so fruitful that the teams eventually merged!
A similar relationship naturally emerged with the Customer Support team. Support had access to trends on what customers were complaining about—now that’s powerful organically collated data. Gregg’s team began to take advantage of opportunities to cross-pollinate—organically breaking down those silos as each team helped the other uncover the truth about their users.
This is when Gregg first discovered the power of breaking down silos across the organization. By sharing insights across the company, each team was able to make better decisions, faster.
Making research discoverable
Despite all their great work, at some point, Gregg and his team at Vox Media hit a wall: they realized they could never scale their research practice to take up every study that came their way, but they could help teams across the company leverage existing insights & run research.
The team began collating findings from previous studies and training materials on how to run research at one single URL. This was a single place where anybody in the company could go to answer questions about their customers.
Gregg also created transparency through research-specific Slack channels where anyone could ask the questions they were struggling with. This led teams that wouldn’t have otherwise known about research to not just discover answers, but engage in research studies themselves.
But where should these resources live?
Digging into the conversation with Gregg, one thing became clear—his repository of insights and training materials always lived where the company’s resources were naturally housed. If the company used Drive, the research needed to be stored there. If they used Notion, Coda, or Confluence, that’s where the research would end up.
Also, read: UX Research Repositories - A Definitive Guide
Gregg was adamant on this point—no matter where you do your work, on Excel sheets, Miro, or anywhere else—you need your insights to be discoverable where the team is going to organically search for it, or they will simply never be found.
Democratizing research: A tactical guide
While theoretical conversation is great, at the end of the day we need to know how to actually enable democratization. To that end I asked Gregg—given that he’s done this many times over, what steps does he advise research leaders to take when democratizing research in their organizations?
- Step #1: If you’re new to the company, the first thing you need to do is to go on a listening tour. Meet with as many people as possible and ask them what information they use to make decisions.
- Step #2: If you don’t already have access to the information people trust, get access to it and create a centralized space—even if just a spreadsheet—where you can list all the resources that can answer your team’s questions.
- Step #3: Share this resource as widely as possible—this is the easiest way to break down silos and get a win under your belt.
- Step #4: Encourage the makers of these insights—data science teams, researchers, customer support—to share their work in a space that’s accessible to the entire organization. That could be Slack, Drive, Notion, or wherever your company lives.
But even while you’re taking all these great steps, Gregg made sure to stress one final thing—you must have the support of your management. They should help you get you into the right meetings, champion your work with the executive team, and build the right bridges.
Research democratization is not a one person job. "You absolutely need help," says Gregg.
.svg)
