Smart Buildings & IoT: A Pragmatic Owner's Roadmap

Start with the outcomes — energy, occupancy, IAQ, asset uptime — and let those drive the sensor and network strategy. The mistake we see most often is buying a 'smart building platform' before knowing which questions the building should answer. Outcome-first, platform-second is the only sequence that pays back.

Energy outcomes need submetering at the panel and at major equipment, integration into the EMS, and a M&V protocol that compares against the baseline. None of that requires a 'smart building'; it requires sub-meters and a discipline. Most so-called smart buildings are 60% of the way there because they skipped the M&V part.

Occupancy outcomes need either people-counter sensors or anonymized Wi-Fi/Bluetooth counting. Both work; both have privacy implications. The right answer depends on the tenant agreement and the regulatory regime. We default to the lowest-privacy-impact option that still produces the data the operator needs.

IAQ outcomes need CO₂, PM2.5, TVOC, and temperature/humidity at occupancy density appropriate to the space type. Office: one sensor per 1,000 sq ft. Healthcare: per-room. Education: per-classroom. Industrial: tied to process risk. Density matters more than sensor brand.

Network architecture choices: avoid lock-in. Open protocols (BACnet/SC, Haystack 4, FIN 5, native MQTT) plus a platform with documented APIs is the floor. Closed platforms ('our cloud, our app, our data export fee') will look attractive at procurement and dreadful at refinancing.

Twin platforms are only useful if a human or workflow consumes the data weekly. A twin nobody opens is the most expensive screensaver in the building. Before selecting a twin platform, name the role (energy manager, FM lead, asset manager) and the weekly cadence they will actually run.

An owner's first six months: instrument the panel and major equipment, name the outcome owners, pick three KPIs the building will commit to, and run weekly reviews. Sophistication is a year-two question.