Problem: The product team had a list of 13 potential features for the Schwab Advisor Platform and needed to know which would matter most to financial advisors before committing design and development resources.
Outcome: A MaxDiff survey and a follow-up usability study narrowed the focus to three priority features, gave the product owner evidence of what advisors actually wanted, and shaped the design direction before anything moved into development.
Read the full case study
Project Brief
The Schwab Advisor Platform had 13 potential new features and limited capacity to build them. Before committing design and development resources, the product team needed to know which features mattered most to the financial advisors who would use them, and then how the most valuable ones should fit into an advisor's workflow. I designed and ran a two-part study to answer both questions: a quantitative ranking, followed by a qualitative check on design direction.
My Role
Survey design, including feature definitions, MaxDiff structure, and screening criteria
Launching the survey and monitoring response quality
Analysis and synthesis of the quantitative results
Planning, moderating, and synthesizing the usability study
Presenting combined findings and recommendations to product and design partners
Approach
Inputs: Reviewed behavioral analytics and findings from prior usability studies to shape the feature list and the survey design
Quantitative: Ran a MaxDiff survey with 200+ financial advisors to prioritize the 13 features. I built the structure from scratch by adapting a UserTesting stack-ranking survey, so advisors repeatedly chose the most-wanted and least-wanted feature from small sets, with every feature compared against the others. MaxDiff forces tradeoffs between features, rather than asking advisors to rate each one in isolation
Analysis: Used Hierarchical Bayes estimation to generate individual-level utility scores for each feature, then converted them into relative importance scores to rank all 13 features
Triangulation: Compared the results against existing behavioral data to check where stated preferences and actual usage aligned
Qualitative: Followed with a low-fidelity usability study with advisors to evaluate design direction for the top features
Synthesis: Combined the ranking and the usability findings into one readout for product and design
The quantitative ranking decided which features to design; the qualitative study shaped how they should work.
Illustrative MaxDiff task
Of these four features, which matters most and which matters least to your work?
MOST
LEAST
Feature A
Feature B
Feature C
Feature D
Recreated, generic example (not actual survey content): advisors choose one most and one least important feature from a small set, repeated across many sets so every feature can be ranked.
Outcomes
Three priority features. The MaxDiff results narrowed the team's focus from 13 candidate features to three.
Evidence that challenged assumptions. The ranking gave the product owner evidence of what advisors wanted, which ran counter to what other stakeholders expected the priorities to be.
Workflow fit before development. The usability study showed how advisors expected those features to fit into their workflow, which shaped the design direction before anything moved into development.
What I'd Do Differently
Reflection
The results ran counter to what some stakeholders expected, which is exactly when evidence needs to be trusted by everyone, not just the product owner. If I ran this again, I'd bring the stakeholders most likely to disagree into the feature definitions earlier, so the ranking carried their trust as well as the product owner's.