Process
Research program management
Research teams do their best work when everything around the research already works: participants show up, consent is covered, data is handled correctly and the money is accounted for. That's the system I build, usually starting from nothing.
How a study moves
The operations a study passes through, in order. I've built each of these at least once.
Before research
Running research
After research
- 1Intake
- 2Ethics, consent, NDAs
- 3Recruit and incentives
- 4Schedule and run
- 5Results and share-out
- 6Repository
- 7Retention and deletion
- 1Intake
- 2Ethics, consent, NDAs
- 3Recruit and incentives
- 4Schedule and run
- 5Results and share-out
- 6Repositoryback to 1
- 7Retention and deletion
New requests start by checking past research
After step 6, the process returns to step 1: New requests start by checking past research
Intake
A product or engineering team brings a question. Every question comes in through the same door, so research capacity goes where it matters.
Where: Reddit
Ethics, consent and NDAs
Codified research ethics and NDAs written with legal, so every participant is covered before a session starts.
Where: Splunk, Reddit
Recruiting and incentives
A recruiting pipeline with NDA intake and tracking, an owned incentive budget, and timelines and costs set before recruiting begins.
Where: Splunk, Reddit, Credit Karma, Meta
Scheduling and sessions
Scheduling handled for researchers, a lab to run sessions in, and an in-person program covering ethics, recruitment, participant experience, data and building security.
Where: Splunk, Credit Karma
Results and share-out
Findings delivered back to the team that asked, with a share-out so the answer reaches more than one person.
Where: Reddit
Repository and archive
Studies land in a research repository so past work can be found and reused. Revamping Credit Karma's archive cut duplicate research by at least 20% right away.
Where: Credit Karma, Robinhood
Retention and deletion
Retention standards, automated data retrieval and deletion requests, and compliance with CCPA, GDPR and FINRA.
Where: Splunk, Robinhood
Running the budget
Research is expensive. Owning the spend is part of the job.
Budgets
From organizing a new team's first budget to running $14–20M a year, with every purchase tracked.
Where: Splunk, Meta
Vendors and contractors
Contract submission and oversight for 60+ vendors and 30 contractors.
Where: Meta
Tooling negotiations
Selecting, negotiating and onboarding research tools, and renegotiating renewals. A UserTesting renegotiation saved 50%.
Where: Splunk, Reddit, Credit Karma
Making research go further
Opening research to more of the company without lowering the bar.
Every question
One intake form, then triage
Already answered
Covered by past research
Link the existing study and close the request
Research library
Self-serve research
Small, low-risk questions
Teams run it themselves with templates, rules and light oversight
About 50 studies a quarter at Credit Karma
Rapid research
Focused, needs a researcher
Dedicated researchers answer it in one to two weeks
30+ studies in its first quarter at Reddit
Embedded research
Strategic or generative
Goes onto a researcher's roadmap with the product team
Planned with research leads
Rapid research
Fixed cycles for focused questions: two weeks for a typical study, as little as one for some. It ran 30+ studies in its first quarter and cut cycle time about 30%.
Where: Reddit
Democratized platforms
A self-serve platform with clear rules, tracking, tooling and enablement sessions, so non-researchers can answer their own questions. At Credit Karma it ran about 50 studies a quarter.
Where: Credit Karma
AI-assisted operations
Evaluating AI research tools with legal, then using AI-assisted workflows to speed recruiting, clean participant data and make past research easier to find.
Where: Robinhood, Meta
Inclusive research and accessibility
A program to make research more inclusive and accessible, so studies reach a broader range of participants.
Where: Credit Karma
International research
Taking research into France, Germany, Italy, the UK and Singapore, and exploring Latin America.
Where: Robinhood
Results
- research budget managed at Meta
- $14–20M
- research budget managed at Meta
- faster recruitment at Meta, Robinhood and Credit Karma
- 30–40%
- faster recruitment at Meta, Robinhood and Credit Karma
- new international research markets at Robinhood
- 5
- new international research markets at Robinhood
Tools I work with
- UserTesting
- User Interviews
- Qualtrics
- Ethnio
- Dscout
- Airtable
- Tableau
- SQL
- Jira
- Confluence
- Asana
- Salesforce
- Figma
Toolkit
An intake form, a vendor spend tracker, an in-person research checklist and a participant data request workflow, rebuilt with made-up data.