ESOS Phase 4 and demand flexibility - what to say and what not to promise
Two things are coming up in conversations with large organisations about their buildings: the next round of the Energy Savings Opportunity Scheme, and demand flexibility - being paid to use less electricity when the grid is under strain. Both involve the BMS. One is a solid source of work. The other is worth preparing for, but easy to oversell.
This article is for integrators who work with large corporate occupiers: retail chains, hotel groups, logistics firms, manufacturers and multi-site service businesses. It covers how ESOS Phase 4 creates a pipeline of controls work, what demand flexibility really pays today, and how to talk about both without promising what the evidence does not support.
ESOS Phase 4 - the dates
ESOS requires large undertakings to assess their energy use and identify savings opportunities in four-year phases. The dates for Phase 4, as reported by third-party summaries of the Environment Agency's guidance, are:
- 31 December 2026 - the qualification date: whether an organisation is in scope is judged on that day.
- 5 December 2027 - the deadline to notify compliance through the Environment Agency's portal.
- 5 December 2028 - the deadline for the action plan, with progress updates in the years after.
Sources also report that the plan to refocus ESOS on net zero has been pushed back to Phase 5, and that a Phase 3 progress update falls due on 5 December 2026 (Energy Advice Hub; GOV.UK ESOS guidance). Sources disagree on some Phase 3 dates, and amendments to the regulations were in progress, so confirm every date against the current Environment Agency guidance before putting it in front of a client. Who counts as a large undertaking is set out in the regulations and the guidance - check it rather than assuming.
Why ESOS is a pipeline for controls work
Every ESOS assessment produces a list of energy saving opportunities, and in buildings-heavy organisations controls feature heavily: time schedules, setpoints, plant sequencing, out-of-hours running, ventilation rates and metering. Through 2026 and 2027, those reports will land on the desks of energy and estates managers at large retail, hotel and logistics groups. Then someone has to do the work - and from Phase 4 onwards, organisations are expected to plan what they will act on and report progress.
For an integrator, that suggests a simple approach:
1. Ask to see the last ESOS report. Phase 3 reports already exist for most qualifying organisations. The controls recommendations are a ready-made scope. 2. Turn recommendations into work packages. An ESOS assessor's recommendation is usually brief. Each one needs a site survey, a method, a price and a way to measure the result. 3. Offer to help with the action plan. Measured results from controls work are useful evidence when an organisation reports its progress. 4. Measure everything. An ESOS saving that cannot be shown later is an awkward conversation at the next phase.
What the recommendations look like on site
Assessment reports often say something like "optimise the BMS" or "review AHU control". On site, that turns into specific faults. One of the most common in air systems is the duct static pressure setpoint: set too high, the fans waste energy; cut too low without checking the far end of the run, the last boxes starve. Static pressure reset that responds to what the boxes actually need avoids both.
A large North American dataset found supply air temperature and static pressure setpoints among the faults that are both common and persistent in air handling units (Crowe et al. 2023). Findings like these, written up with trends before and after, turn a one-line ESOS recommendation into a measured result.
What to say about savings
When an ESOS report quotes a saving, or a client asks what optimisation will save, keep to published ranges with their sources. The largest dataset of recommissioning projects found a median whole-building saving of about 6%, with half of projects between roughly 3% and 12% - a North American figure (Crowe et al. 2020). Some buildings do far better and some far worse. No equivalent UK measured dataset exists.
| Recommissioning existing buildings - 446 projects (measured) | 6.4% (IQR 3.4-12.4%) |
|---|---|
| Fault detection and diagnostics (FDD) - after two years (measured) | 9% |
| Potential found in BMS trend data - 151 buildings (potential) | 12% |
| Optimised controls across the whole stock (modelled) | 29% |
Medians, not promises - individual results ran from close to zero to over 50%.
The standard wording that holds up: published studies found these ranges; this building may sit above or below them; we will measure before and after and show the method. What the studies really say covers the evidence in full.
Demand flexibility - real, but small so far
The National Energy System Operator's Demand Flexibility Service pays participants to reduce electricity demand at times of grid stress. NESO describes it as the most accessible entry point for industrial and commercial sites and their aggregators, because it needs no real-time control or complex telemetry (NESO). Ofgem approved an evolved design on 25 March 2026, making the service bi-directional from April (Ofgem decision), and NESO is aiming for an extra 750 MW of industrial and commercial flexibility by 2030.
But the money so far is modest. In 2025, 12 active providers had bids accepted worth £183.63k in all, for 2.52 GWh of turn-down - against 35.4 GWh offered (industry flexibility report, Q4 2025). Most of what was offered was never called.
| Offered by providers | 35.4 GWh |
|---|---|
| Accepted by the system operator | 2.52 GWh |
The accepted bids were worth £183.63k in all, across 12 active providers. Flexibility readiness is worth offering - income is not worth promising.
Offer readiness, not income
That points to a clear position. Do not tell a client their BMS will earn flexibility income - nobody can promise that, and the 2025 figures suggest it would be small. Do offer flexibility readiness, which has value in its own right:
- Know the flexible loads. Which plant can be turned down or shifted for an hour without harming comfort, safety or a process - pre-cooling, pre-heating, fan speed limits, non-critical ventilation, hot water storage.
- Write the strategies. A documented, tested "turn-down" mode in the BMS, with limits and an automatic return, ready if the client or an aggregator wants to use it.
- Meter it. Half-hourly metering and sub-metering good enough to show what was turned down, which also serves ESOS and any savings report.
- Point to support. NESO's Power Responsive programme offers one-to-one support for organisations considering flexibility (NESO Power Responsive).
Done this way, flexibility becomes a sensible extension of good controls rather than a sales promise.
Shaping the offer
An offer for large corporate occupiers built on this might include an ESOS recommendations review, work packages priced per site, measured before-and-after reporting suitable for the action plan, and a flexibility readiness study with tested turn-down modes. For landlords, Optimisation for landlords covers MEES 2031 and in-use ratings, and Retail, hotels and warehouses covers those building types.
Related tools and training: the energy savings estimator, M&V option picker, carpet plot and load profile, and findings to proposal tools are in tecTools, and the Proving and pricing BMS service course is in tecTraining.
Sources mentioned: GOV.UK ESOS guidance and third-party summaries of the Phase 4 dates (Energy Advice Hub) - confirm against current Environment Agency guidance; Crowe et al. 2020 and 2023 (LBNL, North American); NESO on the Demand Flexibility Service and Power Responsive; Ofgem DFS decision (March 2026); 2025 service figures via an industry quarterly flexibility report. No flexibility income is implied for any site.