B2B SaaS Food Logistics Product Design Responsive UX/UI

Supply
Pro

Type Solo Product Design Case
Role Product Designer & UI/UX
Scope UX Research · Product Strategy · UI Design
Duration 4 Months
Tools Perplexity · Claude · Figma · Miro · Figma Make · Claude Code
Status In Handoff

Supply Pro was born out of a problem I personally encountered over 17 years as a head pastry chef and chef in high-end restaurants. Procurement in the restaurant industry is chaotic and labour-intensive. However, the real issue is not just a lack of communication and documentation, but also a lack of anticipation.

Professional kitchen brigade in service — the environment Supply Pro was designed for
The Problem

Supplier issues
discovered almost too late.

The problem is the agony, the fear of not knowing whether the supplier will arrive on time with the necessary products and whether they match what was ordered. At that moment, service is already at risk. There is very little time left to react.

Research

Deep Qualitative
Research

Interviews

10 Top Executive Chefs with +10 years experience and biggest Fruit & Vegetable suppliers in Switzerland

Countries

Chefs from Germany, Uruguay, France, Switzerland and Spain

Directly Affected

of the chefs surveyed are directly affected by late deliveries

"

"An early warning system for delays or supplier problems would be a good solution to this problem, as it would allow us to respond sooner and salvage the service."

Portrait of Daniel Garcia Serra Daniel Garcia Serra · Executive Chef at the Yellowstone Club, Big Sky, Montana, United States
"

"Then you often have to run around to make sure they deliver the product on time and in good condition, or you have to work with other suppliers, but it's always a headache when that happens to you."

Portrait of Marco Silva Lindblom Marco Silva Lindblom · Executive Chef at the England Embassy, Montevideo, Uruguay
"

"If the product doesn't meet the required standard, I immediately feel frustrated and concerned about the operational side of things. As a chef, I am responsible for the final quality."

Portrait of Edoardo De Simone Edoardo De Simone · Executive Chef at Clinique La Prairie, Clarens, Suisse
How Might We

The key
design question.

The user persona and the ten interviews converged into a single question. This became the filter for every product and design decision that followed.

How might we help chefs in gastronomy detect supply disruptions early so they can react before service is affected?

Process & Key Decisions

Three decisions that
shaped the product

The interviews became the foundation for every product decision. Hearing the same operational tensions across multiple kitchens helped separate assumptions from real workflow problems.

Back to the User
Scope Before Features
Competitive analysis initially pulled me away from the core problem. I became distracted by feature-heavy products and started designing beyond what chefs had actually described in interviews. The research helped realign the MVP around one specific moment: the gap between order placement and delivery arrival.
The Niche Defines the Design
Domain Knowledge as Design Input
Kitchens under time pressure and high stress do not need an impressive interface. They need one that works. This shaped every design decision: no decorative elements, no complex flows, no hidden actions. Functional beats decorative. Seventeen years as a head pastry chef and cook in Michelin-starred restaurants was not a nice addition. It was the decisive advantage.
UX Writing Is Not a Detail
Language as Interface
The first usability test showed that several terms were unclear or contradictory. Testers hesitated not because the interface was poorly structured, but because they were unsure what an action meant. A chef does not read, they scan. This forced a complete overhaul of the UX writing and visual hierarchy. The F-pattern layout stayed. The language was rebuilt from scratch.
Testing & Iteration

What the test
actually revealed

Supply Pro usability testing session
What Was Tested
The test goal was clear: is the early warning system understandable and easy to use for chefs? Five people from the gastronomy environment tested the prototype. The scenario was realistic: two simultaneous problem alerts — a quality issue and a delay — to be resolved within three minutes.
What Worked
The F-pattern layout held. Testers scanned information intuitively from top left to right. No tester had difficulty locating the relevant information. The visual hierarchy guided attention without instruction.
What Failed
The UX writing. Several terms were unclear or contradictory. Testers hesitated not because the interface was poorly structured, but because they were unsure what an action meant. That is a different problem with significant consequences.
What Changed
Complete overhaul of the UX writing. Every term was reviewed: is it unambiguous? Does it trigger the right expectation? The typographic scale was adjusted and a clearer visual hierarchy through color was introduced. The layout stayed. The language was rebuilt.
"

"Even though it was in another language, you could still figure out what to do. In fact, it's super clear — it's not at all hard to understand how it works. There's nothing superfluous; it simply provides the necessary information."

Portrait of Valentin Chappuis Valentin Chappuis · Pastry Chef, Vegan Restaurant Aujourd'hui Demain, Paris, France
Before & After

From generic labels
to clear language

Supply Pro — Before: generic status labels

Before: first usability test

Before the test

  • Generic status terms: ambiguous labels that created hesitation at decision points
  • Inconsistent labels: the same action named differently across screens
  • Unclear CTAs: testers could not determine the consequence of an action before taking it
  • Flat visual hierarchy: primary and secondary actions carried equal visual weight
Supply Pro — After: clear action language

After: revised interface

After the iteration

  • Clear status badges: each state communicates exactly what it means with no interpretation needed
  • Unambiguous action labels: consistent terminology across every screen and state
  • Visible hierarchy: primary actions are visually dominant, secondary actions recede
  • Scannable language: every label written for a chef with three seconds and high stress
The Solution

Three screens.
One clear flow.

Supply Pro focuses on a single operational moment: the gap between placing an order and receiving a delivery. From the first tap to a confirmed decision in under three minutes. Supply Pro is designed for shared kitchen use. While fully responsive, the iPad is the primary device, always accessible, always visible to the whole team.

Early Warning: Alert Screen

The Early Warning Screen uses a table layout designed for high-pressure kitchen environments. Critical information is easy and fast to scan. The status badge system communicates urgency at a glance. The chef reviews the alert and makes one decision: accept a suggested alternative or dismiss it.

Supply Pro — Early Warning System screen
Alternative Product Selection

When a supplier is unavailable, the system automatically suggests products from an alternative supplier. The chef selects only what is urgently needed. Alternatives are organized alphabetically to reduce search time in a high-pressure environment. Checkboxes allow quick selection of one or multiple products without switching screens. Price, delivery time, and availability are provided by the alternative supplier.

Supply Pro — Alternative Product Selection screen
Dashboard: Order Overview

The dashboard gives the chef a complete picture of daily supply operations in one view. Four summary cards surface the most critical numbers instantly. Active alerts are positioned top right in orange to ensure urgency is never missed. Below, the Recent Orders table uses the same status badge system as the Early Warning Screen, creating visual consistency across the product.

Supply Pro Dashboard — Order Overview screen
Prototype

See it in
motion.

AI in the Design Process

One tool per
phase of the process

AI was not a single tool. It was a system of specialized tools — one matched to each phase of the Design Thinking process.

AI workflow across the design process
KPIs

Measuring what
actually matters

These indicators validate that Supply Pro is solving the right problem, and demonstrate how quickly it is doing so.

01
Time to Action
Measure Time from alert received to chef action taken
Target Under 2 minutes per alert
Signal Confirms the interface reduces decision friction under pressure
Results Six tests = 100% time-to-action rate
02
Task Completion Rate
Measure Time from alert to confirm alternative order or ignoring alert
Target Under 5 minutes end-to-end
Signal Validates the full flow — detect, decide, resolve — without manual steps
Results Six tests = 100% task completion rate
Product Outlook

The next three
versions

The long-term goal: Supply Pro becomes the operational memory of a professional kitchen.

V1.5
GPS Tracking
Feature Real-time delivery location visible to the chef
Value Eliminates the last blind spot between order and arrival
Signal Requested explicitly in 6 of 10 research interviews
V2
Supplier Portal
Feature Suppliers confirm actively rather than the system reacting passively
Value Closes the loop between kitchen and supply chain
Signal Identified through supplier interviews as a mutual need
V3
Photo Recognition at Goods Receipt
Feature AI detects quality problems directly at unpacking
Value Documents issues instantly with no manual input
Signal Addresses the last step not covered by the current flow
Lessons Learned

What this project
taught me

Tight Scope Is a Strategic Decision
The decision to focus on one single moment saved the product. A tight scope is not a weakness. It is the only way to build an honest MVP that delivers real value without overcomplicating execution.
Domain Knowledge Sharpens Methods
Seventeen years in professional kitchens was not a nice addition to the research. It was the reason the interviews went deep and the design decisions held. Domain knowledge does not replace UX methods. It sharpens them.
Test with Lo-Fi Earlier
The first usability test came too late in the process. Two iteration rounds could have been saved. Low-fidelity testing surfaces language and flow problems before visual decisions lock them in.