CASE 01 / PLATFORM + MOBILE — SCM GROUP
Maestro Connect
OP 10
Context
SCM Group builds the machines that cut, shape, and finish wood at industrial scale. Maestro Connect is the software layer that sits on top of that hardware — the IoT platform that tells a shop what its machines are doing, when they’ll need attention, and where production is slipping.
The product existed, but it had grown feature-first. It displayed data faithfully and explained almost nothing. Operators were reading raw values and graphs and doing the interpretation in their heads.
OP 20
Problem
The platform was technically complete and practically hard to act on. Every screen was accurate, and none of them said what to do next — so the judgement fell to the operator mid-shift, with a line to keep running.
My job, leading the design team, was to reframe the product around decisions rather than data: what does this person need to know, right now, to keep the shop moving?
OP 30
Process
We worked from the floor up. With service engineers, shop owners, and the SCM product team, we mapped the real questions people brought to the screen: alarms and maintenance, production throughput, machine health over time.
From that we built a design system that could carry the whole platform. A component library, clear data-visualisation rules, and interaction patterns that scaled from a single machine to a full fleet. Automated processes replaced steps people used to reconstruct by hand.
OP 40
Mobile
The people responsible for a shop’s output are rarely sitting at a desk. They’re on the floor, between machines, or off-site entirely. The platform lived on the desktop; the decisions it supported did not. I designed the companion app from a blank canvas — the sole UI/UX designer on the project.
A mobile companion isn’t the desktop shrunk down. The context is glanceable, one-handed, often interrupted — so I built the app around the moments that pull someone toward their phone: a maintenance date approaching, a ticket changing status, production drifting from plan. Each became a fast, self-contained flow, in a visual language faithful to the desktop platform. One system, two contexts.
OP 50
Outcome
Maestro Connect went from a data readout to a working tool — one that woodworking professionals now use across more than 40 countries to run production, plan maintenance, and catch problems before they stop a line. And its core value travels: production, maintenance calendars, and ticket status, unhooked from the control room.
Just as important, the design system outlived the project. The team can keep extending the platform without the interface drifting apart.
What it could have been
The finished screens argue for themselves. These are the forks — what was tried first, and what replaced it.
-
FORK 01
A better dashboard, or a different product?
Rejected
Reorganising the wall of numbers into a clearer wall of numbers. And — honestly — an entire intermediate application, which is what actually sat between the two screens shown above. The before/after is a two-year arc compressed into one wipe.
Shipped
A companion for the production day rather than a readout of it. Once the product helps with each step of a shift, the data stops being the subject and becomes the means to an end. That reframing is what the redesign actually was; the visual change is a consequence of it.
-
FORK 02
What does the platform show you first?
Rejected
The machine list — the obvious opening screen for a fleet product, and the one everyone expects.
Shipped
Not showing it. A list first puts you above the plant, scanning, an omniscient data-checker. The people using this are inside the plant, and what they need on arrival is what is coming at them. I argued this hard and won it — which is the only fight in this project I would call decisive.
-
FORK 03
And what did winning that cost?
Rejected
Modularity — my own instinct, and the one I would normally defend first. A platform that can be assembled per customer is almost always the stronger position.
Shipped
Companion over datasheet, at the price of the modular mantra. Both could not lead, and a product that is configurable into anything is also a product with no opinion about the shift you are standing in. I conceded the thing I usually argue for so the thing that mattered more could hold.
-
FORK 04
Does the platform go to the phone?
Rejected
A responsive cut of the web application. Never seriously entertained: a data-dense platform reflowed onto a phone is not a smaller product, it is an overwhelmed one.
Shipped
A separate companion, purpose-built, doing things the web version does not do at all. That is a strategic decision rather than a layout one — and if modularity is ever won back, this is already its own thing, which makes it a module and a place the product could be sold in its own right.
-
FORK 05
How does a tooling value survive changing units?
Rejected
Converting the number and trusting it to still read. Across 40+ countries the copy changes constantly and occasionally a whole feature does; the assumption that a figure travels intact does not hold.
Shipped
Infographics where the number stops carrying it. Knife sharpness or spindle life is legible in millimetres and much harder to judge in inches — the same value, no longer readable. Drawing the state makes the problem obvious regardless of which unit the reader thinks in.