You Can't A/B Test The Couch
UX RESEARCH MANAGER
Redbox had data, not insight. I built the UX research function from scratch: budget, methodology, cadence, organizational buy-in that turned it into the decision-making infrastructure that product and engineering actually used. By the time I left, skipping research wasn't an option anymore.
The Research Function Didn't Exist. I Built It.
Redbox didn't have a UX research practice. I built one. As Design Manager, I spent half my time conducting research because decisions made without customer evidence were just expensive guesses. Leadership noticed what changed. When I proposed switching roles to own it full time, they didn't hesitate. I hired a dedicated researcher, took ownership of a $150K annual budget, and turned research from a side activity into the infrastructure that shaped how every product got built.
RESEARCH PRACTICE ARCHITECTURE
$150K
Annual research budget secured and managed annually
4
Distinct research methods institutionalized across product teams
4
Product surfaces covered: kiosk, web, mobile and tablet
RESEARCH PRACTICE ARCHITECTURE
ONGOING
Kiosk environment observation
Analytics Review
Customer support analysis
PER SPRINT
Task-based usability testing
Live streaming with product teams
Watch-along findings report
QUATERLY
In-store intercept studies
Quantitative customer profiling
Quarterly Report Outs
ANNUAL
Customer segmentation refresh
Competitive benchmarking
Research roadmap planning
Kiosk Observation Revealed What Analytics Never Could
Analytics told us where people abandoned. It didn't tell us why. The only way to answer that was to be there -- in the store, watching real rentals happen in real conditions. That's what field research gave us that no dashboard could.
FIELD RESEARCH METHODS
Task Based Usability
Kiosk Environment Studies
Researchers stationed at retail kiosks during peak hours to observe ambient conditions, queue behavior, and real-world usage patterns. Findings surfaced lighting failures, navigation confusion, and time-pressure abandonment that A/B testing couldn't isolate.
Contextual Inquiry
In-Store Ethnographic Studies
Quarterly deep-dives at partner retail locations -- Walgreens, McDonald's, Walmart -- pairing behavioral observation with post-rental interviews. These sessions consistently surfaced the gap between what customers said they wanted and what they actually did at the machine.
Live Usability Sessions Changed How Product Teams Made Decisions
The biggest organizational shift wasn't a research finding, it was a process change. I introduced live-streamed usability sessions so stakeholders could watch real customers struggle in real time. No findings report has the same impact as watching someone fail at a task you designed and discuss solutions after the session ended.
EVALUATIVE RESEARCH METHODS
Blu-ray Upsell Testing
In a structured task analysis study, the Blu-ray upsell revealed multiple issues and a dark pattern. Revenue takes a significant hit if the screen isn't fixed.
Live Streaming With Stakeholder
Live weekly research sessions was a feedback loop to gain early insight on early designs.
One Research Model For One Customer
Testing was covered across four screens. The instinct was to treat each surface as a separate design problem. I made the case that the customer was the same person making a decision about how to spend a Friday night, and that research had to follow the customer, not individual screens. Cross form factor studies mapped behavior across all four surfaces delivering a coherent user experience.
THE STRATEGIC PRINCIPLE
"You can't A/B test the couch. You can't instrument the parking lot. Some of the most important design decisions Redbox made were protected by research that no quant could have surfaced. That's why the function needed to exist."
Outcomes
Organizational Impact, not just study outputs.
-
Built and managed a $150K annual UX research budget from zero -- the first dedicated research function in the company's history
-
Institutionalized research as an organizational behavior: skipping a study required justification, not the other way around
-
Collapsed the assumed persona model from 7 to 2 validated customer segments, sharpening product decisions across all three surfaces
-
Live streaming and Watch-Along reports reduced the time from usability insight to sprint action from weeks to 24 hours
-
Cross form factor research framework prevented three separate product teams from designing for three different imaginary customers
Next
What I would have pushed for next:
-
Formalize a research repository so findings from individual studies could be accessed by all
-
Hire a junior researcher to expand study throughput and free senior time for strategic synthesis
-
Bring research upstream into product strategy, as studies were still largely evaluative; the next phase would be shaping roadmap decisions before wireframes