InnovationSystem
From Data to Action
Turning an industrial energy-monitoring platform into a clear, building-by-building consumption view for a mixed-use complex.
Role
UX, Service Designer
Mapped the devices and assets to a data schema, created low/hi fi mockups.
Results
In production
Visualization live, mapping per-building, floor, and unit consumption to a custom central plant and device data schema.
Client
InnovationSystem
Energy-monitoring company specialised in consumption analytics for energy-intensive facilities.
01. A platform built for factories, not homes
InnovationSystem's platform monitors energy consumption for energy-intensive companies and production plants. This project was an experiment in applying it to a very different context: a mixed-use complex of three buildings. The interface, designed around industrial installations, had to be rethought so that consumption could be read building by building, floor by floor, unit by unit and asset by asset — and compared against the energy drawn by the electrical and thermal circuits, which shifts between summer and winter. This breakdown was defined by talking with the building's administrator and the technician responsible for installing InnovationSystem's monitoring hardware.
My first step was to translate the physical installation into a hierarchical schema: from the complex down through its systems to each specific plant (for example, Building A3). Beneath each node I mapped how every consumption figure is actually calculated — which analyzer, meter or tag it refers to — so the data had a single, unambiguous source.
02. Mapping the existing platform
Before proposing anything new, I documented how the platform's GUI had been organised for this specific mixed-use installation. This was primarily a mapping of the existing screens — screenshots stitched into a flow — with some early sketches layered on top.
Laying out the current interface this way made it clear where the industrial-oriented design got in the way of a reading suited to residents and the administrator, and gave the whole team a shared reference to discuss against.
03. Proposing modular components
From that analysis I put forward two modular component proposals. The first focused on visualising thermal and electrical consumption for each central plant; the second introduced the building → floor → unit → asset view that the client was asking for.
Each proposal sat alongside the client's own requirements, numbered so that every design decision could be traced back to a specific request. I prototyped hi-fi interfaces by using AI code assistants, this reduced each iteration cycle and allowed me to fit in more stakeholder interaction.
04. Bridging the data to Danea Domustudio
The client needed the consumption data to land in Danea Domustudio, the accounting software used to manage the complex. I couldn't change how the platform exported its data — only analyse the shape it came in.
So I studied the platform's output for compatibility against what Danea Domustudio expected, then built a conversion step that reshaped the exported data into the Excel format the accounting software could import. This kept the two systems aligned without touching the platform's core.
05. Into production
The visualization went into production for the mixed-use complex, delivering the building-by-building, floor-by-floor and unit-by-unit readings the client had set out to obtain. It also raised a question of data ownership: the administrator wanted key readings saved outside the platform, so the complex wouldn't depend on a proprietary system to keep its own consumption history. I can't show the final result here, as it contains proprietary technology and data present in the production version.
Running the design against real meters and circuits showed which readings administrators actually rely on to manage the complex, and where the view had to be simpler and more direct.
Conclusions & Lessons
The platform was modular, but at the time it was missing key features — a timeline and historical data among them. Much of the work was adapting a system built for industrial plants to a mixed-use complex, and designing around those gaps.
It also meant building real domain knowledge: understanding how HVAC systems and IoT monitoring actually work — central plants, meters, circuits — so the data schema mirrored the physical installation. And it showed how differently each role reads that data: administrators think in units and bills, technicians in circuits and meters.