When PEAK went live at the London headquarters of a global consumer brand in late March 2026, the building's electricity use was tracking just 3% below the previous year, ordinary business-as-usual variation. Eleven weeks later, it was using 14% less than the same window a year earlier, measured on the building's own meters. And the gap was still widening.
At a glance
- 14% reduction in electricity use, 101,521 kWh, in the first 11 weeks against the same period a year earlier
- Around £27,900 saved in under 11 weeks
- 23% down year on year in the most recent fortnight
- Around £126,000 a year projected if the 14% reduction holds
- 38 PEAK actions raised and worked by the site team in the first 12 weeks
Challenge
The building used 3,277,476 kWh of electricity in 2025, roughly £901,000 a year at 27.5p per kWh. At that scale, small inefficiencies add up quickly, but they are also easy to lose in the noise of a busy headquarters. The building is run by an integrated facilities management provider, with a separate controls contractor looking after the BMS.
The question was not whether waste existed. It was where it was hiding, and whether the people responsible for the building could find it and fix it fast enough to make a difference.
What PEAK found
The dominant pattern was plant running when nobody needed it. Within weeks of go-live, PEAK had raised overnight and weekend operation actions across fan coil units on floor after floor, both main air handling units, supply and exhaust fans, and DX cooling units. Sixteen of the first 38 actions flagged equipment running outside occupied hours.
PEAK also surfaced reliability issues the site team needed to know about, including a return air fan that had stopped operating on a main AHU and several unit failures raised as P1 critical actions, along with broken CO2 and airflow sensors on VAV boxes.
How much energy does after-hours plant operation waste?
Out-of-hours operation is the most common fault in commercial buildings, found in 78% of the buildings CIM monitors. Individually most instances are small, often worth a few hundred dollars a year, but they are spread across dozens of units, so together they add up to a material share of a building's bill. Here, it was the single biggest pattern PEAK found.
The cause is almost never mechanical. It is usually a schedule changed for a reason that no longer applies, an optimum start routine that was never tuned, or a unit left in hand after a callout. We break down the signature and the cost range in our analysis of the most common HVAC faults in commercial buildings.
How the loop closed
What made the difference was ownership. Each action in PEAK was assigned to a named engineer in the FM team or the controls contractor, investigated, and closed out with comments on the record. CIM's customer success team ran on-site training with the engineers on the ground, which built the trust needed for the team to work from PEAK day to day rather than treat it as another dashboard.
Closing an action did not end the watch. PEAK kept running the same rules after close-out, so a schedule that crept back or a unit returned to hand would be flagged again rather than quietly undoing the saving.
A batch of overnight operation actions was closed out in mid-May, and the fortnight that followed, from 29 May to 11 June, was the strongest yet: 31,567 kWh saved, down 23% year on year and worth around £8,700 in two weeks.
Impact
Between go-live on 24 March and 11 June, the building used 649,215 kWh against 750,736 kWh for the same window in 2025. That is 101,521 kWh less, a 14% reduction worth around £27,900 at the building's tariff. Before PEAK went live, consumption had been running only 3% below the previous year, so the step change coincides with the platform and the work done through it.
If the 14% reduction holds for a full year, it projects to around 458,800 kWh and £126,000 a year. These figures come from a straight year-on-year comparison of metered consumption rather than a modelled estimate. A formal whole-facility calculation under IPMVP Option C would also adjust the baseline for weather and occupancy. For more on how savings are proven, see our guide to measurement and verification of energy savings.
Why this one matters
Overnight operation is rarely one big fault. It is dozens of small ones spread across a building, each easy to miss and each quietly adding to the bill. Fault detection and diagnostics finds them, but the savings only arrive when someone takes ownership of each action and sees it through. Here, the FM team and controls contractor did exactly that, with PEAK keeping every action visible until it was closed. It is the same principle behind independent building monitoring: the platform raises the issue, the people on site fix it, and the record shows it was done.
For another example from a London site, see how sceptical engineers confirmed a £25,000 fault in 98 minutes.
Results
- 14% reduction in electricity use in the first 11 weeks, 101,521 kWh against the same period a year earlier
- Around £27,900 saved, measured on the building's own meters
- Most recent fortnight down 23% year on year, with savings accelerating
- Overnight and weekend plant operation identified across fan coil units, AHUs, fans and DX units
- 38 actions raised and worked through PEAK by the FM team and controls contractor in the first 12 weeks
- Projected saving of around £126,000 a year if the reduction holds
