Case Study 03 | Research Operations

Evolving a strategic research practice.

At Capital One Auto Navigator, a three person research team supported a growing portfolio of products and partners. Research demand was rising, but intake was still being handled case by case, making it hard to separate high-value strategic work from lower-priority requests. I simplified the prioritization model to four criteria, made tradeoffs visible, and used demand data to help the practice move from reactive support to strategic partnership. The new system contributed to a 25% increase in strategic requests and helped secure approval for an additional researcher.

Senior User Research ManagerCapital One Auto NavigatorPortfolio planning

Research operations visual showing a shift from tactical intake to strategic planning based on business impact, customer value and capacity fit. — The prioritization model shifted intake from a reactive queue toward more strategic planning.

25%

year-over-year increase in strategic research requests after the model was adopted

1 headcount

additional full-time researcher request approved after capacity became visible

Overview

Research demand was outgrowing the team's operating model

The challenge was not only too many requests. It was the absence of a simple, shared way to discuss value, urgency and capacity across the product portfolio. By turning prioritization into a transparent operating model, I helped leadership see both the demand pattern and the staffing implications.

Make strategic tradeoffs visible

I created a four factor model and recurring prioritization process that made the research portfolio easier for Product, Design and leadership to understand.

Use demand data to shape the practice

I used the prioritization data itself to show where the team was spending time, how demand was changing and why additional research capacity was needed.

Challenge

Research demand was growing faster than team capacity

Three researchers supported 39 designers, while ad-hoc requests made it difficult to distinguish strategic work from overlapping studies or work that partners could self-serve with research support.

Turning Point

The first model was too complex to use

First attempt

A richer nine-factor model captured nuance but created too much effort for regular use.

Adaptation

I reduced the framework to four clearer inputs and prioritized adoption over theoretical precision.

The simplified four factor scoring framework made research demand easier to evaluate and discuss.
View full size

Results

A simpler model made demand and tradeoffs visible

The model gave leadership a transparent reason to sequence, deprioritize or redirect work. In November alone, six requests were deprioritized because they lacked sufficient business impact or overlapped prior evidence.

Quarterly dashboard showing demand, priorities and completions month to month.
View full size

Capacity Decision

The model showed a staffing problem, not just a queue problem

Once the queue was visible, the team could show what additional capacity would unlock: deeper embedded work, proactive research, more foundational studies and less late-stage risk.

Headcount request framed around the value additional research capacity would unlock.
View full size

Result

One additional full-time researcher request was approved.

PreviousAvoiding feature overload through mixed method research NextDriving consistent persona adoption with AI