Aerospace / Enterprise product design
Pratt & WhitneyEngineering clarity.
Without losing depth.
Redesigning a 20-year-old engine test data platform around the people who depend on it.
Explore the case study01 / The challenge
A powerful tool.
An aging experience.
Pratt & Whitney’s field engine test data analysis software had been in use for more than two decades. System analysts, production engineers, and aero thermal engineers relied on it to find test logs, inspect datapoints, and make sense of complex engine data.
The challenge was not to make expert work look simple. It was to make the next decision clearer while preserving the technical control that gives the platform its value. A cosmetic refresh would leave the underlying workflow untouched.
- My responsibility
- Lead UX: discovery, workflow mapping, interaction design, design system, usability testing, and implementation reviews.
- Collaboration
- Pratt & Whitney engineering and product stakeholders.
- Engagement
- Multi-phase redesign of a data-heavy enterprise platform.
20+ years
of legacy product complexity
3 audiences
with different routes into the same data
1 system
for a consistent engineering experience
02 / The approach
One dataset.
Three ways of thinking.
I used interviews and workflow observation to understand how each audience moved through a test log. Rather than treating “the engineer” as a single persona, I mapped the starting point, information needs, and decisions of each group.
System analysts
Find the right evidence
Metadata and parametric search are the entry point. The design needs to express a precise query without burying its logic.
Production engineers
Scan and move through work
Project and testlog throughput depend on a workspace that supports rapid comparison and makes records easy to return to.
Aero thermal engineers
Go deeper into the data
Datapoint-level investigation needs context, selection controls, and access to technical applications—not a stripped-back dashboard.
The design principle
Progressive depth, not a compromise between experts.
I aligned the redesign around streamlined workflows, productivity, and collaboration. The organizing idea was to keep the common path readable, then let users move into richer information when their task demanded it.
03 / The design
From finding a project
to investigating a datapoint.
I organized the experience around a connected sequence: return to a project, build a query, review the results, and investigate a test log. Each step retains the context needed for the next.
- 01Find a project
- 02Define a search
- 03Review testlogs
- 04Inspect datapoints
Decision 01 / Orientation
Two views.
One familiar workspace.
Projects are the user’s point of return. I made project metadata visible before a record is opened and brought recent searches into the same workspace.
The list view supports side-by-side comparison; the card view provides a more contained summary for browsing. Keeping both grounded in the same information avoids forcing every task into one presentation.
The trade-off: preserve density for comparison, but offer a different reading pattern—not fewer capabilities.
Decision 02 / Retrieval
Make the query
as readable as the result.
I rebuilt search around metadata and parametric queries—the language engineers already use. Separating fields, operators, and values exposes the logic instead of hiding it behind a generic keyword box.
Recent searches give complex queries a reusable starting point. Results then preserve that context, so users can understand which records they are reviewing before moving into a test log.
Designing beyond the happy path
A long-running query is still an experience.
The supplied search design includes an explicit processing state for large result sets. Acknowledging the request and explaining the delay helps distinguish “working” from “nothing happened.” This state deserves the same attention as a populated table.
Decision 03 / Investigation
Keep context beside
the work it informs.
The testlog workspace connects file selection, datapoints, technical applications, messages, and test information. This is where simplification through removal would be most damaging: the surrounding detail is part of the investigation.
I used hierarchy to separate orientation, selection, and action. Basic and advanced datapoint views support different depths of analysis, while shared patterns keep those experiences connected.
- Orient: keep the project and testlog path visible.
- Select: expose data types, filters, and row selection near the dataset.
- Act: keep plotting, reduction, profiling, and editing associated with the selected data.
04 / Validation & delivery
Test the journey.
Not just the screen.
I connected the screens into a prototype to make navigation, search, and investigation reviewable as a complete experience. The original Figma file shows the breadth of those relationships—not just a collection of finished layouts.
Usability testing informed refinements to labeling, density, and comparison flows. I worked with engineering through implementation reviews to carry the tested direction into delivery.
Shared patterns
A common design system keeps tables, filters, navigation, and states consistent across workflows.
Iterative refinement
Review terminology and information density against the tasks of each expert audience, rather than a generic notion of simplicity.
Implementation continuity
Design reviews connect intent to engineering decisions, keeping the interaction model intact through handoff.
05 / The outcome
A modern foundation
for expert work.
The redesign established a modern replacement for a legacy aerospace tool: one design system serving three expert workflows, search structured around engineering queries, and a layered approach to datapoint analysis.
My contribution connected discovery, interaction design, validation, and implementation reviews. The goal was continuity—not a polished concept that becomes a different product at handoff.
Impact is described qualitatively. No verified post-launch productivity or adoption figures are available for publication here.
- Metadata and parametric search aligned with engineers’ mental models
- Basic and advanced views that preserve expert control
- Consistent patterns across projects, testlogs, and datapoints
- Design intent maintained through engineering reviews
What I took forward
“In expert tools, clarity comes from organizing complexity—not pretending it isn’t there.”
Working alongside analysts and aero thermal engineers taught me that experts don’t want less information—they want it in the right order. Layering a scannable basic view over a deep advanced view let three very different groups share one product without anyone compromising. I now bring that progressive-depth approach to every data-heavy tool: earn trust with the essentials first, then make depth one deliberate step away.