Building a UX practice from scratch - strategy, discovery with leadership and a design system for a multi-product B2B platform.
Wüest Partner is one of Switzerland's leading real-estate data and valuation companies, with a growing ecosystem of digital products for portfolio management, sustainability and data onboarding. As the sole designer across this ecosystem, I owned far more than screens: I led discovery on ambiguous new initiatives, designed three data-heavy platforms end-to-end, built the design system that unifies them, and elaborated the UX vision for the product organization.
How open questions became directions everyone could commit to.
The most valuable work started without a brief: a new module or a new client need. I ran Double Diamond discovery workshops with product managers, domain experts and engineers - and my job was to turn that ambiguity into a direction everyone could commit to.
The toolkit stayed deliberately light: workshops that map jobs-to-be-done and decision points, open questions re-phrased into challenges with a success metric attached, and quick, throwaway prototypes, so leadership could react to something real instead of debating abstractions.
As a audience I want to goal so that I can benefit, as measured by success metric. Currently, I'm unable to do this because problem.
Discovery is woven through every phase of delivery - from the first framing of a problem to a validated solution running in production. Each phase pairs divergent exploration with convergent decision-making.
The operating model puts four roles at one table for product discovery - product manager, product owner, tech lead and design lead - so business, product, technical and user perspectives shape every decision from the start. As design lead, I brought the user's perspective to that table.
Two products show the approach end to end: a fund-planning tool that started as a blank page, and a rent-roll module shaped with nine stakeholders in three months.
Rapid AI prototypes became part of my discovery toolkit. Mapped onto the Double Diamond, they move the first clickable prototype from the end of a weeks-long process to the very beginning - while people stay essential for validation and alignment. Hover a phase for details.
A new product for planning funds and portfolios - from a blank page to an MVP.
Fund Management is a new Wüest Dimensions satellite for planning funds and portfolios. Users model a fund's structure, capital and loans, fees and cashflows, apply rules and scenarios, and see the effect on every portfolio and property. There was no existing tool to redesign - just a wide range of users and more questions than answers.
Select a node, or add an asset with +
Discovery started with the people: who plans what, on which level, and what they need to decide. Mapping their jobs side by side showed where the levels meet - and where one person's output becomes another person's input.
For every level of the product - the executive summary, a portfolio, a single property - I mapped what users need to see, what they need to do, and which decisions they need to make. Mapping it level by level kept the navigation and every screen focused on what matters there.
The discovery board held use cases, a simplified sitemap and a feature parking lot - and a growing list of open questions. I worked through them in Double Diamond workshops with product managers, domain experts and engineers, tested the answers in a workshop with a key client, and refined them over two demo rounds.
The heart of the product is a guided creation flow: set up the fund's structure, define the rules that drive its cashflows, then refine capital structure, loans and fees. A multi-level navigation moves users between portfolio, fund and property without losing context, while empty states and a getting-started screen guide first-time users. The MVP - capital structure, chart of accounts, rules, scenarios and settings - went to engineering with a dedicated handoff page.
Funds, portfolios, SPVs, properties and buildings - with scenarios and project versions on top. Before settling on one navigation, I explored three sidebar concepts: drill-down panels, a level switch on a single list, and an expandable tree scoped to a working project, with toggles that decide which properties count into the forecast.
Fund Management was also where the shared design system started: the team wrote the proposal for one component library across all satellites, and the product was planned as the first to use it. The business case behind it: the cost of one button.
Bringing client data into Wüest Dimensions - with a rent-roll module that turns scattered files into one reliable source of truth.
RED-X is the platform that brings client data into Wüest Dimensions. Its rent-roll module turns rent-roll files into a reliable source of truth for asset, property and valuation managers - three months from the first workshop to high-fidelity design.
Users needed four things: one view of leasing, revenue and risk data; early warnings for terminations, rent increases and renewals; comparisons across the portfolio; and an import they could trust.
I ran a Double Diamond workshop with nine stakeholders to map jobs-to-be-done, decision points and conflicting expectations across three personas. Quick, throwaway prototypes then aligned stakeholders, surfaced constraints early and secured buy-in, and targeted walkthroughs validated the concept with the client.
The concept went straight to high fidelity, assembled from existing design-system components. AG Grid stays the foundation for dense tables; an executive summary pairs KPI tiles with a rent-roll timeline; the area-units view shows every unit's size, rent, tenant and status; and a step-by-step wizard breaks contract creation - including conditions and indexation - into manageable stages.
In parallel, I prototyped AI-assisted parsing that turns PDF and Excel rent rolls into an editable, structured table. It became the shared alignment artifact for AI experts, the Chief Product Officer, the Chief Security Officer and product managers, bridging business, technical and compliance perspectives on a feature nobody had built before.
The module earned strong feedback for its at-a-glance overview of occupancy, renewals, vacancies and revenue - especially the timeline view and the unit-level drill-downs. Because UI and micro-interactions match the rest of the platform, users move between modules without relearning patterns. Next on the roadmap: AI parsing, anomaly and risk signals, scenario suggestions and natural-language queries.
A vision for how design works across the ecosystem - and the shared system that makes it real.
Every design discipline in the product organization ran through one function. Beyond product delivery, I covered strategy, discovery, systems, measurement and alignment - working directly with leadership, product owners and engineering leads.
I elaborated the UX vision and operating model for the product organization. It rests on three pillars: knowing what to build, building it consistently and fast, and proving that it works. Each pillar has one long-term outcome.
The Wüest Dimensions Design System is the single source of truth for every satellite product: 12 color roles, a Roboto type scale, 286 Feather icons and 21 documented component groups - from buttons and alerts to AG Grid tables, KPI boxes and the side navigation. Components ship with all their states and variants defined, so new modules are designed and built from existing parts. Change a component once - the primary button, say - and the update reaches every module that uses it.
A design system only gets built when its value is visible in business terms. With the Fund Management team, I made that case with the smallest possible example: a single button. Built separately in every product, it costs the full price each time. Built once in the system, it costs a little more up front - and very little after that.
Every change has to be made three times.
Every change happens once and reaches every product.
From the third product on, the system pays for itself - and every further product saves another 17 hours. For a single button.
The point was never the button. A shared system moves the conversation from pixels to people:
The system came with a clear vision and a phased plan, so each product could adopt it without pausing delivery.
To make the vision actionable, I mapped out an eight-step roadmap that takes any product area from a usability audit to measured, communicated results - each step with clear methods and a shareable outcome.
To connect design to business outcomes, I built KPI trees: every UX initiative traces up through a metric and a design KPI to a business goal. The principle: nothing ships just because it looks better - it ships because it moves a number that matters. Below, a sample branch.
The strategy is proven in shipped work: three data-heavy platforms, each designed end-to-end and all built on the Wüest Dimensions design system - one visual and interaction language across the entire ecosystem.
The UX vision lays out a clear blueprint for how design works across the product ecosystem: three pillars, an eight-step roadmap and a direct line from business goals to design decisions. Ambiguous initiatives like rent-roll management and AI-assisted parsing moved from open questions to validated concepts. Three platforms share one design language through the Wüest Dimensions system, new modules are assembled from documented components, and a single change propagates across every product - all from a design function of one.
Running design solo at this scale is a lesson in leverage. Screens are the least of it: the design system, discovery with leadership and a shared language of business KPIs are what let one designer serve an entire product organization. The skill that mattered most was comfort with ambiguity - walking into a room with C-level stakeholders and no brief, and walking out with a direction everyone could commit to.