Most buildings with a Building Management System are already collecting data that would explain a meaningful share of their energy waste. The system's there to keep plant running - it usually isn't being asked the second question of what its own data reveals about how efficiently that plant is actually running.
HVAC - the biggest single opportunity in most buildings
Heating, ventilation and air conditioning typically represents one of the largest controllable energy loads on a commercial site, and it's also the system most likely to be running against outdated or drifted settings that nobody's revisited since commissioning.
Setpoints that quietly drifted from sensible
Temperature and pressure setpoints get adjusted for a complaint, an event, or a one-off need, and then never get put back. Many BMS platforms can trend setpoints, sensor values, commands, alarms and runtimes - provided those points have actually been configured for logging and the history's been retained - and reviewing that record is usually the fastest way to find setpoints that made sense once and don't any more.
Occupancy - is the building actually being run for the people in it?
A BMS that's genuinely responding to real occupancy - reducing conditioning in unused zones, adjusting for actual working hours rather than a generic schedule - behaves very differently from one running a fixed schedule regardless of who's actually there. The gap between the two is often larger than people expect.
Alarms as a genuine diagnostic, not just noise
A recurring alarm is evidence worth investigating, not a nuisance to be muted - though what it actually reveals varies. Some expose a genuine plant fault or control problem; others turn out to be a bad alarm threshold, a sensor issue, or a communications fault rather than anything operationally wrong. Either way, an alarm that's quietly acknowledged and dismissed repeatedly without anyone establishing which of those it actually is means useful diagnostic information is being discarded. Alarm history, reviewed properly rather than cleared reflexively, is one of the more underused diagnostic tools already sitting inside most BMS platforms.
Runtime - equipment running when nothing needs it to
Runtime logs show exactly when plant actually operated, which is often quite different from when it was supposed to. Equipment left running outside occupied hours, or running continuously when it was designed to cycle, shows up clearly in runtime data - and rarely shows up any other way until someone's specifically looking for it.
Trending - the pattern over time, not the snapshot
A single reading tells you very little. A trend over weeks or months shows drift, seasonal variation, and genuine anomalies far more reliably than any point-in-time check - and most BMS platforms already log enough history to build that picture, even if nobody's routinely pulling it.
Integrating BMS data with half-hourly meter data
A BMS shows what the building's systems were doing. Half-hourly meter data shows what that actually cost, in real electricity terms, at the same moments. Combined, the two datasets confirm - or challenge - assumptions about which loads are actually driving cost, rather than which ones simply feel like they should be.
Most of the diagnostic work here doesn't need new sensors or new equipment - it needs someone to actually look at what the existing system has already been recording. If you want that review done properly, get in touch.