Research Repositories
This video is featured in the UX metrics & analytics playlist.
Summary
Key Insights
-
•
Research repositories vary widely, from libraries of reports to rich knowledge bases integrating multiple data sources.
-
•
One major challenge is motivating researchers and non-researchers to consistently populate repositories with relevant, contextualized data.
-
•
Adding contextual metadata—like methodology, market, and date—is crucial for meaningful repository content.
-
•
The term ‘research repository’ is evolving into ‘customer knowledge base’ reflecting integration across organizational knowledge.
-
•
Using templates to structure research entries simplifies data input and improves AI’s ability to parse and summarize content.
-
•
AI agents can help maintain context when searching repositories and flag when no relevant answers exist, reducing hallucinations.
-
•
Human oversight remains essential to verify the relevance and accuracy of AI-generated search results, especially with dated data.
-
•
Integrating AI bots with communication tools like Slack increases repository adoption by enabling on-demand access to research insights.
-
•
Context retention by AI is necessary to prevent misleading conclusions, such as mixing up different customer segments or outdated information.
-
•
Scoring research impact and linking recommendations to outcomes in repositories can improve organizational learning but is still emerging.
Notable Quotes
"Personally, if I was gonna go into some aspect of UX research, I would jump in with both feet into research repositories."
"There are so many different variations of repositories, from research libraries to databases cataloging atomic insights."
"Research knowledge by itself is not enough. You need to add other sources like user voice, surveys, and customer feedback."
"You cannot just put this 5% of users did not want this without knowing why or where this insight came from."
"AI agents should understand the entire context and, if no suitable result exists, come back honestly without lying."
"We made it mandatory to have a one-pager summary for every research project to make insights easy to manage."
"To motivate people to populate the repository, be annoyingly persistent and sometimes create the template for them."
"Connecting the repository to Slack and building an AI bot to answer questions fascinates users and drives usage."
"We have been good with not having fake or hallucinated answers so far, and retaining context is key."
"Human oversight is crucial because no single metric like date alone can determine if research is outdated or still relevant."
Or choose a question:
More Videos
"For every single component, we start with its purpose and emergent need to ground design in user experience."
Dave Malouf Amy ThibodeauPanel: Design Systems and Documentation
November 7, 2017
"With AI tools, you have to come back and reprompt, it’s not a one-shot operation."
John DonmoyerShipping your code generation experiments to production
June 11, 2025
"An AI-driven design tool should help me by lowering the redundant work, reducing tedious tasks and help me focus on the experience and strategy."
Clara Kliman-SilverUX Futures: The Role of Artificial Intelligence in Design
June 7, 2023
"Culture plus empathy equals pride—and pride powers the most successful organizations."
Mark TempletonCreating a Legacy: the ultimate experience
June 9, 2017
"Signals are something that’s happening at the horizon, percolating at the fringes of society or in a narrow subset of the population."
Sarah GallimoreInspire Progress with Artifacts from the Future
November 18, 2022
"You may need to operate the prototype on behalf of users whose assistive tech can’t click or interact with clickable prototypes."
Sam ProulxPrototype Reviews, People With Disabilities, and You
December 8, 2021
"Our monthly voice of the customer reports we call Stories in Dovetail; we share them with the team every month."
Anna Nguyen Emily BroganWhy Our Voice of the Customer is Better Than Yours
March 10, 2022
"How do we continuously provide value throughout the product development cycle without overindexing on just one type of research?"
Anna Poznyakov Richa PrajapatiGet The Most Out Of Stakeholder Collaboration—and Maximize Your Research Impact
March 12, 2021
"Sometimes being a leader means making decisions that feel like high stakes gambling with incomplete data."
Nalini P. KotamrajuTwo Jobs in One: Being a “Leader who is a Researcher” and a “Researcher who is a Leader"
March 10, 2021