Barbie's dream UX research repository

The gist

Yeah sure Dovetail offers a "repository" but our researchers felt it was suffocating. They had very little confidence they could get their stakeholders in there and leveraging their insights. So I built them something custom.

Especially proud of
  • Freeing up our researchers time
  • Making significant doc usability improvements by keeping up new Coda features (ask me about the filtering experience)
My role

Coda doc architect and handy-man.


1 month to go from 0-1

Trapped knowledge

Our UX research team needed a way to let non-researchers browse through their growing collection of key findings and recommendations.

Until our repo, this information was only accessible in static presentation decks and the minds of our research team. Researchers have better things to do then respond to dozens of Slack requests saying "can you send me the link to the studies that have key findings about [topic]?" And the volume of such messages increased drastically during quarterly OKR planning season. During that time, researchers would spend a significant amount of their time helping gather findings. And it was all up to their individual memories... "oh, I think we have a study on that but I didn't lead it, so-and-so did, let me message them..."

Four slack messages, all asking the UX research team for help locating findings and recommendations the team has produced.

This was inefficient, and unreliable. And the research team knew this. They felt the time-suck, and were burdened knowing this time could be spent running studies for new insights, or doing meta analysis on existing findings. But they had to keep keeping on, the business was dependent on their trapped knowledge.

A perfect opportunity for me, the operations gal, to help with a new solution.

Self-serve research, anyone?

We needed a proper research repository. But not the one baked into our qualitative analysis tool, Dovetail—it was too opinionated in how we organize insights. The team wanted the flexibility to organize their fidnings in a way that suited both them and product managers, designers, and engineers. Dovetail's solution wasn't up to scratch.

Enter Coda. At its core, our Research Repo Doc was run by just 3 data tables:

  • Key findings
  • Recommendations
  • Studies

These would be interlinked with lookups. And through the magic of structured data, we could associate a finding to a certain page of the product, component, or user flow. Here's a walkthrough of me using the Coda research repo with dummy data:

Not only did this meet our self-serve solution, we could also show the business the rate at which research was producing insights, meeting with users, and time spent steeped in raw data.

UX Research efficiency, quantified

Screenshot of the UX research efficiency dashboard. It's displaying dummy data from Q1 of 2023. 8 Initiatives completed, for an average of 2.67 initiatives each month. This is down 27% compared to last quarter, which saw 11 initiatives completed. 80 key findings and 84 recommendations were produced for an average of 5.9 key findings and recommendations each week. This is down 18% from last quarter's 194 combined findings and recommendations. Researchers met with 40 participants for an average of .6 participants per day. This is down 60% compared to last quarter's 105 participants. 28.5 hours of feedback was analyzed, which is an average of 0.06 hours each week. This is down 26% from last quarter's 38.8 hours of feedback.

Every study we ran was tracked as row in our "Studies" table.

  • Lead researcher (People)
  • Knowledgeable contributors (People)
  • Participants met with (Number)
  • Time spent with participants (Time - Duration)
  • Count of insights produced (Number)

Mix in "months", "quarters", and "years" and we can begin to measure UXR output over time. And contextualize it! For instance, maybe Q2 saw a 50% decrease in studies ran compared to Q1, but that was because the research team was conducting the annual benchmark study (a big effort with a massive payoff).

With this information, the team could tell a richer story of their contributions to the business.

Facilitating researchers' growth

It only took a couple extra columns to start helping researchers understand each other's expertise. Up until the repo, folks would just hear in passing or absorb through exposure things like, "so-and-so is really knowledgeable when it comes to tree tests." Now we had a way to track it!

  • Which researchers have led usability studies?
  • Of that group, who led one most recently?
  • Who hasn't led a usability study yet?

With this the team had a way to understand who was expert in what, and easily get guidance and support from them as they had their go using a new method. This is a massive asset for new hires who don't yet know each teammate and their bodies of work. It's also beneficial for designers who want to get practice leading research initiatives. This allows them to find an appropriate partner.

The 2.0 I one day hope to make

  • Task generator for researchers. Every study has a predictable set of tasks that the lead researcher needs to execute and delegate to stakeholders. Luckily, the researchers also use Coda for their task tracking, which gives us the opportunity to have the repo doc and task doc "talk"! Every new study could push auto-generated tasks out (yet another time saver).
  • Hook into Quantive and ProductBoard for stronger connections to OKRs and upcoming feature work.
  • Researchers will often spin up a Coda doc for a single study to house their research plan and organize the effort with other stakeholders. Being able to push data from those docs back to the repo would save researchers tons of time. Coda's newer page embed and writeable cross-doc syncs unlock this capability.
  • Use Coda's AI feature to help audit the growing collection of findings and recommendations. More specifically, looking for similar recommendations and findings across studies. This can give the UX team leverage when advocating UX enhancements and drive roadmap prioritization decisions.

There's more to this story

🔒 But it's reserved for interviews 🔒

Invite me for an interview