The net zero conversation is missing half the picture - the electrical half

Walk into any development meeting where sustainability is on the agenda and within five minutes, someone mentions heat pumps and someone else talks about the building fabric. What almost nobody talks about is the electrical design.

That’s the MEP lag. And it’s quietly become one of the costliest misconceptions in sustainable design.

This matters now because the UK’s emerging definition of “net zero” is moving from marketing language to verifiable proof, and electrical design choices are a major part of what gets measured.

Version 1 of the UK Net Zero Carbon Buildings Standard (UKNZCBS), published in March 2026, changes the terms of the conversation. For the first time in the UK, there is a single, independently verified framework that sets out what a net zero building must demonstrate in practice (across defined scopes and reporting) rather than what it claims on paper. For project teams, this shifts “net zero” from an aspiration to a test: performance must be measured, reported and evidenced against a defined benchmark, and design decisions need an auditable rationale from the outset. A significant proportion of what determines that outcome sits squarely within the electrical engineer’s scope, yet most project teams barely interrogate it.

We are used to making decisions based on considerations of CAPEX versus OPEX. In practice this means weighing the upfront, installed cost of an electrical solution (materials, labour and programme) against the life of the building (energy losses, heat rejection/cooling load, maintenance and replacement). If we only optimise for cost at day one, whole life cost and whole life carbon don’t properly enter the room, even when there is no programme impact and no functional compromise. The simple fix is to make it explicit: agree a study period and compare options on a like for like whole life basis.

There are plenty of ways to reduce carbon in building services, but from an electrical perspective three areas are consistently worth exploring early: 

• The embodied carbon locked into containment, cabling and distribution equipment.
• Whether the building’s control strategy actually reduces in use consumption (rather than just it being modelled well during the design process).
• The electrical architecture choices that shift losses around the system (including emerging approaches like Power over Ethernet).

The carbon nobody counts

A large commercial building contains thousands of metres of cable. Steel containment runs through every floor. Add switchgear, distribution boards and control systems, all specified once and then embedded for decades.

Research by Carbon Leadership Forum noted that the impact of MEP systems on a buildings embodied carbon can account for 15%-50% through the buildings lifespan. This rises to over 70% for a commercial building undergoing a retrofit. This is why carbon needs to be considered early. By the time a project reaches technical sign‑off, the opportunity to influence it is dramatically reduced.

For example, you can specify two different containment approaches, both compliant, and end up with very different embodied carbon outcomes. This isn’t about recommending a brand, it’s about making the carbon visible early.

Option A: wire mesh basket containment (lighter steel content per metre).
Option B: perforated steel cable tray (more material, different support strategy).
 

How to compare: ask for the EPDs for what’s actually being proposed and compare like-for-like (same duty, same standards). The numbers can move a lot depending on material route and recycled content.
 

Outcome: in most like-for-like comparisons, the basket option comes out with a significant reduction in embodied carbon because there’s simply less steel per metre.
 

Takeaway: if we don’t interrogate electrical materials and methods early, we lock in avoidable embodied carbon before the project even reaches technical sign-off.

Smart buildings that actually are

Many Building Energy Management System (BEMS) installations underperform compared to their design capability. Default schedules are set at commissioning and rarely revisited; sensors are installed, data is gathered, and nothing changes. In my experience, “install and forget” remains commonplace, and it isn’t often challenged at design stage.

Under the new standard, the direction of travel is clear: demonstrated in-use performance rather than modelled projections. 

On a commercial office scheme currently on site, we pushed this further than usual. Small power circuits (for desks, monitors and typical workstations) are placed under BMS control and linked directly to lighting occupancy sensors. Out of hours, non‑essential loads shut down automatically. If a floor is occupied, power remains live. When it clears, power cuts.

The result is a building that uses energy when and where it’s needed, not one that merely looks efficient on paper.

It isn’t free. Additional contactors and controls infrastructure are required. But the point is that it’s not “install and forget” technology, it creates a default behaviour (auto shut-down out of hours) that keeps working even if nobody touches it again. It’s visible and adjustable within the BMS (trends/alarms/schedules) so facility managers can tune hours, add exceptions and respond to complaints without a contractor call out.

Takeaway: sustainability isn’t free. If we want real world impact, we have to invest in controls that translate “smart” into in use performance that facilities can see, adjust and prove in operation. Otherwise, it won’t stand up to the direction set by the new standard, so the right question has to be asked at the start.

PoE: efficiency, losses and where they show up

Another emerging area worth interrogating is Power over Ethernet (PoE). It’s often positioned as a neat way to simplify distribution for low power loads (lighting, sensors, access control and small devices), but it also shifts where losses occur. Mains power is converted to DC in the PoE switch or midspan, then often converted again inside the end device. Those AC-DC and DC-DC stages introduce real efficiency penalties and heat that can be invisible in early energy conversations, particularly when multiplied across hundreds or thousands of ports. With the newer PoE-focused content now appearing in BS 7671, it’s a timely reminder that “low power” doesn’t automatically mean “low loss”, and that the most efficient architecture depends on device type, cable lengths, utilisation and how many conversions you’re stacking in series.

Recommendation: treat PoE as an engineered choice, not a default. Use it where it removes significant material, space or commissioning complexity, but require a simple loss/heat check (switch efficiency, end device conversion, cable length and diversity) and confirm who will own configuration and monitoring in operation.

Takeaway: PoE can reduce kit and containment in the right applications, but it can also hide conversion and cable losses. Don’t assume it’s lower carbon by default; demand a simple, evidenced efficiency and heat check before you commit.

Ask this first

Before any other sustainability conversation on a project, one question must cut through the noise:

“Has the whole‑life carbon impact of the electrical installation been quantified - and did it actually change what was specified?”

Not calculated and filed.
Changed.

Mechanical systems will always dominate the net zero narrative. But electrical engineering governs the other half of the equation: how power flows, how it’s managed, how consumption is controlled, and how performance is measured and proven.

The assets that retain value as carbon regulation tightens will be the ones whose electrical performance is measured, managed and evidenced, not assumed.