Brian S. Utesch, Ph.D. — UX Research Leadership

Brian S. Utesch, Ph.D.
Rigorous methods. Measurable outcomes.
Research that drives decisions.

Co-creator of UMUX-Lite. Inventor of the Recommendation Adoption Score. Three US patents. More than two decades building research teams and methodology at IBM and Cisco, with a consistent pattern: if the research does not change the product, the research did not work.

Ph.D. Human Factors UX Research Leadership AI Workflow Strategy Certified User Experience Professional

Expert in UX research methods:

Usability Testing Surveys Interviews Ethnography Behavioral Metrics Journey Mapping Field Studies Diary Studies Concept Testing Contextual Inquiry A/B Testing Experiments Usability Benchmarking Focus Groups Card Sorting Persona Development Secondary Research Clickstream Analysis Heuristic Evaluation Task Analysis Observation Unmoderated Testing Participatory Design Remote Moderated Testing

What I believe about this work.

Research is a science, a discipline, and a profession

We need to treat UX research with the rigor of a scientific approach, not as a creative exercise that borrows from science when convenient. That means maintaining the unique attributes of systematic inquiry, including replicability, transparency, and evidence-based conclusions, whether the work is qualitative or quantitative.

Balance and converging evidence

We need qualitative and quantitative data. We need behavioral and attitudinal data. We need to conduct generative and evaluative research. Getting out of balance on any of these dimensions will not produce the insights and impact the organization needs. Converging evidence from multiple methods is how research earns the right to influence decisions.

Measurement is king

We have to measure everything we do as researchers, and we always start with the current as-is situation from which all comparisons of value must begin. Without a baseline, there is no way to demonstrate improvement. Without measurement, research is just opinion with a methodology section.

The receipts.

Research leaders talk about connecting insight to impact. These are the numbers behind that claim.

10th → 1st

Industry web ease-of-use rankings

Led ibm.com global redesign research across 8 markets, 7 languages, and 2,500+ participants.

One of the largest instrumented user test programs in IBM history.

17% → 43%

Research recommendation adoption in six months

Created the Recommendation Adoption Score (RAS) at Cisco to hold research accountable end-to-end.

Published by Nielsen Norman Group in February 2026.

6 → 18

Research team scaled at Cisco

Expanded product coverage from ~10 to 25+ tools and lifted researcher proficiency from 2.8 to 4.2 on a 5-point scale.

A structured methods readiness program tied to team and individual OKRs.

283%

Customer engagement growth in one year

Transformed IBM Lotus UXR from qualitative-first, employee-only to large-sample, externally-focused practice.

External participant mix climbed from 27% to over 60% of all research.

56% → 69%

+13 pts in 12 months

Customer satisfaction on IBM PCD eSupport, redesigned with task-based UCD methodology.

Also generated $193K in e-commerce revenue and saved $74K in research costs through in-house execution.

9,419

▲ from 1,390 in two years

Annual user research engagements at IBM Lotus/Collaboration Solutions by 2013.

External participant mix climbed from 27% to over 60% of all research.

250

From zero in ~5 months

Active sellers in IBM's Seller Feedback Community, which Brian founded from scratch.

Created reusable user-centered design research assets deployed across worldwide product squads.

2 Awards

Same program cycle

Sammy Award and Stevie Award for the IBM CHQ MarTech user research program.

Independent external industry recognition for research excellence and organizational impact.

91%

59 of 65 studies reported time savings

Share of Cisco research studies that saved time through AI assistance, with 69% of all studies AI-assisted.

Evidence that AI belongs inside the research workflow, not as a bolt-on. The foundation for the Pathfinder methodology.

21% → 35%

Generative research mix, 2023 to 2025

Shifted the team's work upstream from evaluative testing toward generative research that shapes product direction.

Surpassed the 33% target. Repositioned research from evaluating the "how" to influencing the "what."

Problems solved with evidence.

Original contributions to the field and applied research that started with real problems and ended with measurable change.

UMUX-Lite

Ten Questions Was Costing Us Data

The problem. The System Usability Scale required users to answer 10 questions to measure usability. In fast-moving product environments, that length was causing drop-off and losing data we needed. I wanted a shorter instrument that could do the same job without sacrificing validity.

The solution. I partnered with Jim Lewis at IBM to create a 2-item scale and validated it against real user data, demonstrating high correlation with the SUS (.81), which meant we could continue using SUS norms as the benchmark. We published it at CHI 2013 and IBM Design adopted it as a company-wide standard. It has since been independently validated multiple times by other researchers, and MeasuringU has made minor refinements and now calls it UX-Lite.

UMUX-Lite is now one of the most widely cited usability scales in the field, used by research teams who need a valid, low-burden measure that fits inside real product research cycles.

UMUX-LITE: When There's No Time for the SUS (CHI 2013) UMUX-LITE and SUS Correspondence (HCI International 2015)

Recommendation Adoption Score

No Way to Know If Research Was Being Used

The problem. As a research manager, I had no quantifiable way to measure what actions were being taken on the outputs of my team's work. We could count studies completed and recommendations delivered, but whether those recommendations reached the product and the user was invisible. Research organizations were operating on faith, not evidence.

The solution. I developed the Recommendation Adoption Score in collaboration with my colleague Tammi Fitzwater. The RAS tracks each recommendation from delivery through implementation, producing a quantifiable adoption rate that tells us not just how much of our work is being used but where in the system recommendations are breaking down. That signal directs where to invest limited research resources for maximum impact.

RAS drove adoption from 17% to 43% in six months at Cisco. Published as a three-article series by Nielsen Norman Group. I built a free, open app so any research team can start tracking adoption immediately.

Research Recommendation Breakage (NN/g) The Recommendation-Adoption Score (NN/g) Implementing and Tracking the RAS (NN/g)

Try the free RAS app → Built for the research community to make adoption tracking accessible to any team.

Pathfinder AI Workflow Enablement

AI Was Being Deployed Without Understanding the Work

The problem. Organizations were rushing to automate user workflows with AI without first understanding how those workflows function today. The result was AI features bolted onto broken processes, with no baseline to measure improvement against and no mechanism to protect the user's role in the workflow.

The solution. I created Pathfinder as a standardized methodology that bakes user engagement into every stage of workflow automation. The process starts by mapping the current-state workflow through journey maps and service blueprints, then engages with users throughout the design of the new automated workflow rather than only at the end. The result is a new workflow with a measurable baseline, so the expected benefit of automation can be predicted and validated rather than assumed.

At Cisco, 69% of research studies became AI-assisted under this methodology. The generative research mix shifted from 21% to 35%, moving the team upstream from evaluating designs to shaping what gets built.

How Our Team Works in the Age of AI: Introduction to Pathfinder Two Kinds of AI, Two Different Futures for UX Research

IBM Collaboration Solutions

What Does Collaboration Actually Mean?

The problem. IBM was building collaboration tools like Connections without a data-driven understanding of what collaboration looks like in practice. Product decisions were based on assumptions about how people work together rather than evidence of their actual goals, tasks, and role relationships.

The solution. Ran five structured surveys with 2,455 participants to map the goals, tasks, and dependencies that define collaboration across the organization. Identified 17 distinct roles with enough data to build role-specific goal models, each showing which goals mattered most and how they depended on each other. The resulting framework made it possible to personalize the product experience by role rather than shipping one generic interface to everyone.

The General Manager presented the findings at the Lotusphere General Session in Orlando. The goal models became the evidence base for product roadmap decisions across IBM's collaboration portfolio.

ibm.com

Validating Design Alternatives on Live Traffic

The problem. Traditional usability testing could evaluate designs in a lab, but it could not predict how they would perform at scale with real users doing real tasks. Launching a redesigned page and hoping for the best was the default, and rollbacks were expensive.

The solution. Created a live A/B testing methodology that combined implicit behavioral data (click paths, page linger time, goal completion) with explicit attitudinal data (preference surveys, satisfaction intercepts) on live ibm.com traffic. Routed 100,000+ visits per variant by pre-selecting incoming domains, then analyzed first-click patterns and path frequencies to understand not just what users did but why they preferred one design over the other.

The converging behavioral and attitudinal evidence validated design alternatives before committing to a live launch, eliminating the guesswork. The methodology was adopted as a company-wide ibm.com service offering, used across subsequent redesign programs.

IBM Digital Sales

Which Seller Pain Points Matter Most?

The problem. IBM's Digital Sales team was building tools for sellers without a systematic understanding of which problems were most severe or most important to fix. Investment decisions were driven by stakeholder opinion and squeaky-wheel escalation rather than user evidence. With dozens of competing pain points, the team had no defensible way to prioritize.

The solution. Went on-location to sales centers in Seoul, Dublin, Dallas, and Atlanta to interview sellers directly. Built empathy maps from the qualitative data, then distilled 30 discrete seller needs and surveyed sellers at scale to quantify both the current sentiment (how negative is this right now) and the importance (how much does this matter to your work) of each need. Plotted all 30 on a scatterplot of importance versus satisfaction to identify the quadrant where needs were both highly important and deeply unsatisfied.

The high-importance, low-satisfaction quadrant became the prioritized investment roadmap for the Digital Sales product team. This upstream research directly fed the creation of the Seller Feedback Community (0 to 250 active sellers in five months) and the attitudinal KPI framework that tracked product health longitudinally.

Depth by DrillBit Labs

What Does the Market Actually Want from a UX Researcher?

The problem. The field has always guessed at what skills, methods, and experiences researchers need. Career advice is based on personal opinion and small-sample observation rather than evidence. Practitioners trying to develop professionally, and leaders trying to build career ladders, have no objective benchmark for what the market is requesting at each level of seniority.

The solution. I started this work with an analysis of 100 job postings, then partnered with Lawton Pybus and Thomas Stokes to scale it. We built a data pipeline to collect, clean, and code thousands of UX research job descriptions, producing the field's first data-driven career ladder with benchmarked qualifications, responsibilities, methods, and compensation at every seniority level. The analysis now covers nearly 3,000 postings and continues as a living dataset, tracking how the market evolves including the emergence of AI as a hiring factor.

The data gives practitioners an objective guide to what they need to demonstrate in their current role and in the role they aspire to, replacing guesswork with evidence. Published through Depth by DrillBit Labs and UX Collective, with an interactive market snapshot tool at uxr-job-market-jul-2026.netlify.app.

Two decades. Two companies. One consistent standard.

Progressively senior technical and leadership roles focused on building research as a business-critical function, not a support service.

Head of Intelligent Experience Research

Cisco · Research Triangle Park, NC · 2022 – Present

Invented RAS and built the tracking infrastructure that drove adoption from 17% to 43% in six months. Scaled the team from 6 to 18, expanded product coverage from ~10 to 25+ tools, and launched a structured methods readiness program tied to OKRs. Developed the Pathfinder AI workflow enablement methodology grounded in current-state workflow research.

STSM & Design Research Practice Lead, Manager

IBM · Research Triangle Park, NC · 2019 – 2022

Led the user research program across the full IBM CHQ MarTech stack, establishing shared KPIs aligned to organizational OKRs. Recognized with the Sammy Award and the Stevie Award in the same program cycle.

STSM & Design Research Practice Lead

IBM · Research Triangle Park, NC · 2016 – 2019

Primary voice of the customer for all IBM digital sales tools globally. Founded the Seller Feedback Community, growing it from zero to 250 active sellers in approximately five months. Established the attitudinal KPI framework and intercept survey infrastructure across IBM's Cognitive Sales Advisor.

Senior Scientist & Senior Technical Staff Member

IBM · Research Triangle Park, NC · 2004 – 2016

Co-created UMUX-Lite at CHI 2013. Led ibm.com user research across 8 international markets on a $1M+ portfolio. Transformed Lotus UXR into a large-scale quantitative practice. Built the predictive brand perception model from 13,000 task arrays. Recognized as ibm.com Innovation Champion, received the GM's Innovator Award, and appointed to the IBM TechGen program.

Advisory Engineer, Personal Computing Division

IBM · Research Triangle Park, NC · 2001 – 2004

Redesigned IBM PCD eSupport using task-based UCD. Customer satisfaction improved from 56% to 69%, generated $193K in e-commerce revenue, and saved $74K in research costs. Won first place in Customer Insight at TechConnect 2003.

Invited to share the work beyond Cisco.

Thought leadership that led to speaking engagements at major organizations, plus hosting the field's recognized measurement authorities at the Cisco Insights conference.

Organizations invited Brian to present

External Speaking Engagements

Invited to present on UX research practice, measurement, and team development.

Truist

Travelers Insurance

Deloitte

Lenovo

Great Question

TriUXPA

387

Registrants, 51 speakers, 69 NPS

Cisco Insights Conference

Created the first cross-Cisco research conference to break down best-practice silos across research teams company-wide.

Built from idea (Aug 2022) to execution (May 2023) with 3 co-chairs and 14 volunteers

24 sessions across 11 short talks, 9 long talks, and 4 panels/workshops

92% agreed it improved their understanding of research practice across Cisco

92% said it made them excited about the future of UX research at Cisco

0 → 43

External eminence examples in one year

Team Eminence Built

Grew a publishing and speaking culture across the research team.

43 external talks, presentations, and publications in a single year

23 internal talks, presentations, and blogs

12 external articles authored personally in 2025

Brian S. Utesch, Ph.D.
Rigorous methods. Measurable outcomes.
Research that drives decisions.