METHOD
Interviews · Usability test
PARTICIPANTS
7 recent-transaction consumers
CONTEXT
B2C consumer product
PHASES
2
01
THE PROBLEM
A consumer-facing app was being designed to help people track the home buying and selling process — progress, documents, and the actions required of them along the way.
Two things were unproven.
The need: would consumers actually find value in a tool that tracked their transaction, stored their documents, and surfaced their required actions?
The execution: did the proposed interaction design support how people expected to navigate — or would it get in their way?
02
MY ROLE
Sole researcher. I designed both phases, recruited participants who had recently completed a transaction, wrote the interview and usability protocols, moderated all sessions, analyzed the behavioral and attitudinal data, and reported findings to the product team.
03
RESEARCH QUESTIONS
Would consumers value an app that helped them track their transaction, store documents, and manage required actions?
Did the proposed interaction design support how they expected to navigate that experience?
04
APPROACH, AND WHY THIS METHOD
The two questions needed two different kinds of evidence, so I sequenced a generative phase into an evaluative one.
PHASE 1 · GENERATIVE
Interviews
Sessions with consumers who had completed a transaction within the last year, exploring whether a journey tracker would have been valuable, and what information they needed most.
PHASE 2 · EVALUATIVE
Usability test
Task-based sessions on the proposed design, measuring time-on-task, first-click accuracy, and ease of completion on a Likert scale — focused on whether people understood where to find information and what was interactive.
WHY THIS ORDER
Need before execution
There was no point evaluating an interface for a product people didn't want. Establishing the need first meant the usability findings landed on a foundation worth improving.
Recent-transaction sample
Recruiting people who had just been through the process meant they could speak to real friction and real information gaps from memory, not hypotheticals.
Behavioral + attitudinal metrics
Time-on-task and first-click accuracy captured what people did; the Likert ease ratings captured how it felt. Together they separated "slow but confident" from "fast but lost."
05
HOW I ANALYZED IT
Interviews were coded thematically to establish what consumers valued and where they currently lost visibility. The usability data combined first-click analysis (did people head to the right place first?), time-on-task, and ease ratings — then I looked for tasks where the behavioral and attitudinal signals disagreed, because those gaps are where the sharpest design problems hide.
Illustrative reconstruction. Buyers and sellers framed it differently, but the underlying want was identical — autonomy and visibility over their own transaction.
06
WHAT I FOUND
Consumers wanted autonomy
On both sides of a transaction, people wanted to reduce how often they had to contact their agent for information that should already be available to them.
Buyers wanted to track progress and know what was next. Sellers wanted real-time offer activity and time on market. Different vantage points, one shared need: visibility and control over their own process.
Usability surfaced three issues
01
Agent contact information was hard to find when users needed it.
02
Completed tasks were difficult to locate, so people lost their sense of progress.
03
Without a consistent design system, users were unsure what was actually interactive.
That third issue was the important one. It wasn't a single screen's problem — it pointed to a product-maturity gap beyond any individual design decision.
Inconsistent affordances meant users couldn't build a reliable model of what they could tap, click, or act on — which undercut the very autonomy the product was meant to deliver.
07
RECOMMENDATION
The recommendations tracked directly to the three issues: surface agent contact where and when users reach for it, make task state legible so completed and outstanding actions are easy to locate, and establish a consistent design system so interactive elements read as interactive. The first two were screen-level fixes; the third was a signal to leadership about foundational investment the product needed before polish would matter.
08
IMPACT
The research demonstrated clear consumer demand for a tool that gave people visibility and control over their own transaction, and the application was green-lit following the study.
The business later took the consumer product in a different strategic direction, toward competing in the broader home-search market. The research had done its job: it validated the need.
What the product ultimately became was a separate business decision. Being clear about that boundary — between what research established and what strategy later chose — is part of representing the work honestly
09
WHAT I'D DO DIFFERENTLY
Sample size
Seven participants was enough to surface strong qualitative signal and clear usability issues, but a larger evaluative round would have let me quantify the task metrics with more confidence.
A second usability round
Testing a revised design after the first round would have confirmed the fixes worked, rather than leaving the recommendations unvalidated.
Prototype fidelity
Some of the "what's interactive?" confusion was partly a fidelity artifact. Separating true design problems from prototype limitations more cleanly would have sharpened the findings.