Design thinking where it counts
At Mindset Consulting — SAP's first US-based AppHaus Network Partner — I worked as an embedded senior UX practitioner on enterprise digital transformation engagements. My primary account was Cargill, where I served as dedicated UX lead across a multi-year product team. I also led design thinking and UX discovery engagements for Ecolab and Entegris.
Challenge
Cargill's financial instrument master data request process had no UX foundation — users had not been identified, workflows had not been mapped, and no design support existed for the five distinct roles involved in submission and approval.
Approach
As dedicated UX lead on Cargill's enterprise SAP program, I facilitated a full-day design thinking workshop with stakeholders representing five roles: Requestor, FI Approver, Maintainer, Functional Approver, and Viewer. Dot voting surfaced stakeholder priorities. Whiteboard flow mapping aligned the group on interaction logic across roles. I followed with user research, detailed role-specific flows, and interaction prototypes delivered to engineering — all within MyDataQ, Cargill's enterprise master data platform.
Outcomes
Prior to this engagement, no UX documentation existed for the financial instrument master data request workflow. The workshop established that foundation: five distinct user roles identified and documented for the first time, with role-specific interaction flows covering submission, validation, approval, and status tracking across each. This became the UX basis for subsequent prototyping and engineering work. MyDataQ — documented in Mindset's published case study — reduced Cargill's customer data request SLA from three days to three hours and simplified data entry 85% across five apps
Key Learning
In multi-role enterprise workflows, the highest-risk design decisions are not within each role's experience — they are at the handoff points between roles. Optimizing the Requestor flow without understanding what the Maintainer receives downstream creates friction that no UI polish fixes. Mapping all five roles in a single session made those handoffs visible before a single screen was designed.
Growth Opportunity
The FI Master Data module addressed one data domain. MyDataQ's architecture supports multiple data types across Cargill's enterprise. The natural next phase would extend the same UX model — scenario-driven entry, role-appropriate validation, unified approval chain — across additional master data domains.
Challenge
Ecolab's cross-divisional sales agreement management process generated 50,000 contract changes annually across disconnected workflows. Territory reps, account managers, and approval chains were operating without a unified system — creating approval bottlenecks and manual overhead that scaled with the business.
Approach
I joined after discovery was complete, taking ownership of the Frame phase. Working directly with Ecolab stakeholders, I facilitated structured ideation and design studio sessions — guiding cross-functional participants from initial concept sketches through iterative refinement to a consolidated solution framework — and delivered interaction prototypes for the proposed system.
Outcomes
The resulting platform serves 250 users across Ecolab's divisions. Documented in Mindset's published case study: 90% reduction in average agreement approval process length, 24,000 hours saved, 2.75 days eliminated waiting for batch jobs within the first three months.
Key Learning
Enterprise process redesign works best when stakeholder workshops separate organizational ownership questions from user experience questions. The design studio structure gave cross-divisional participants a neutral surface to evaluate concepts on merit rather than organizational politics — producing faster alignment than stakeholder interviews alone would have.
Growth Opportunity
The redesigned workflow addressed agreement creation and approval. The larger opportunity is intelligence — proactively flagging agreements at risk of dispute before they enter the approval chain, rather than managing disputes after they surface.
Challenge
Entegris needed to replace decades-old plant floor processes — paper worksheets, Excel printouts, AS/400 dependencies — with mobile-first SAP Fiori experiences on ruggedized Android tablets across multiple manufacturing plants.
Approach
I led a multi-phase UX engagement alongside a junior designer, treating client delivery and practitioner development as simultaneous tracks. We conducted plant floor user interviews, mapped current-state journeys across putaway, receiving, and cycle counting workflows, synthesized findings into a 2x2 problem prioritization matrix, facilitated shared goal definition with Entegris stakeholders, and generated concept sketches for tablet-based interactions. I prepared the junior designer before each phase, kept the work grounded during facilitation, and debriefed after — every step of the way.
Outcomes
Entegris stakeholders received research-grounded UX artifacts across the full discovery arc: journey maps, problem prioritization, goal alignment, and interaction concepts. The junior designer completed their first end-to-end client engagement with direct mentorship throughout. The broader Entegris program — documented in Mindset's published case study — delivered 24 custom Fiori apps replacing the AS/400 system and earned Most Innovative UX at Mindset's 2023 EXperts Awards.
Key Learning
Developing a junior consultant's facilitation capability works best when mentorship runs parallel to live client work, not after it. When the junior designer had to manage real stakeholder dynamics in real time with immediate coaching available, the learning compressed significantly. The client benefited too — two practitioners in the room produces richer observation and more robust synthesis.
Growth Opportunity
Developing a junior practitioner through a live client engagement works. What it exposed is the need for a more structured mentorship framework — defined coaching moments at each phase, explicit criteria for when to step in versus let the junior designer navigate independently, and a deliberate debrief process after each stakeholder interaction. Running it ad hoc produced good outcomes here. A repeatable model would scale that outcome across a practice.
