From one home to a provider programme

Scale begins with a consistent question

A group can buy equipment at scale more easily than it can establish that the same equipment is right for every home. Its estate may contain converted houses, purpose-built homes, recent acquisitions and services in leased accommodation. Resident needs, plant condition and energy contracts differ. A useful programme begins with a common decision process that can accommodate those differences. The aim is to choose the right intervention for each service and learn across the estate.

Start with a site register that people can maintain. Record the physical address, service identity, property owner, operating entity, energy accounts and responsible manager. Link a campus's separate buildings and supplies without assuming that one registration corresponds to one meter or one property. An acquisition should enter the register with its uncertainty visible. Missing records are a work item, rather than permission to fill a field with an apparently representative sector average.

Build the evidence in layers

The first layer is operational evidence: consumption, contract terms, occupancy context, building condition, maintenance and comfort concerns. The second is the technical survey and its assumptions. The third is external context, such as certificates, property research and public energy statistics. These layers answer different questions. Keep their dates and boundaries visible so a convenient external figure does not displace the evidence of the actual service.

Public datasets help locate and classify evidence, but they are not a free map of every home's energy expenditure. England's CQC data identifies regulated services and locations, with known update and migration limitations. Energy certificates describe assessed building characteristics and can be superseded. Neither establishes who pays a particular bill. ^{[1] [2]} A join between them needs an address or property match, a recorded confidence level and review of ambiguous campuses, extensions and changes of operator.

Use archetypes to organise investigation

Group similar homes by features that affect the decision: building form, main fuel, existing plant, tenure and service operation. Floor area and occupied beds can help explain scale, but should not become the sole basis for ordering investment. A home with unreliable hot water or serious comfort concerns may need attention before a larger home with a higher annual bill. Condition and continuity risk belong at the start of the prioritisation process.

An archetype is a useful commissioning aid. It might allow a group to ask comparable questions about several older heating systems or several roofs approaching repair. It does not remove the site visit or the need to understand local electrical capacity, fabric, access and resident needs. Record where one home differs from its apparent peers. The programme should make those exceptions easier to explain, rather than reward a manager for concealing them in a standard template.

Read the benchmarks without creating a false target

Knight Frank's 2025 trading report offers utility cost comparisons across its sampled care estate for 2024/25. Those observations are historical, sample-based and influenced by the homes included. ^[3] They can prompt useful questions about a provider's own costs. They cannot establish that every converted home should reach the cost of a purpose-built comparator, or that a proposed installation will close a particular gap. Electricity, heating and wider utilities also need a consistent boundary before comparisons are meaningful.

DESNZ's September 2026 non-domestic data release similarly requires care with dates and categories. Its latest consumption observations concern 2024, and broad building categories should not be relabelled as care-home measurements. ^[4] Our practical approach is to show external comparisons alongside their limitations, then return the investment decision to the site's own reconciled evidence. A benchmark can identify an investigation. It should not create a savings promise in a board paper.

Choose a pilot that can teach the group something

A pilot should have a clear question and enough baseline evidence to answer it. A lighting controls trial might test an ancillary area with an understood operating pattern. A heating project might address a known replacement need in a home where the technical assessment and management support are ready. Choose the first project for its relevance and ability to generate reliable learning, not simply because a supplier can install there quickly.

Define the result before commissioning. For controls, that includes use, staff override and suitability of the environment. For solar, it includes generation, on-site use, imports, export and retained costs. For heating, it includes energy input, comfort, hot-water operation, maintenance and whole-project expenditure. Identify the evidence the group can collect without adding unreasonable work to care routines. Some technical information should come through the equipment or maintenance process rather than through staff manually copying readings.

Procure common standards with local authority

A group can standardise the information requested from bidders, the treatment of exclusions, the handover record and the process for approving changes. It can ask for a complete cost boundary and a named maintenance route. These are useful economies of scale because they improve decision quality across different technologies. A uniform equipment specification is a separate choice that needs evidence about where it fits.

The local manager needs a meaningful role in works planning and acceptance. Central approval should not leave the home negotiating interruptions after installation dates have been agreed. Set conditions for resident communication, access, noise, temporary arrangements and escalation. Confirm who can pause work if service conditions change. The group programme should strengthen those arrangements by making them expected, funded parts of delivery at every site.

Turn learning into the next investment decision

Review the first projects at thirty days, ninety days and across a full operating year. Separate initial faults from longer-term performance. Check whether the original assumptions survived actual use and whether the records are good enough to explain the result. If a measure works only in one operating pattern, make that limitation part of the replication brief. A qualified result is more useful than an early success announcement that others cannot reproduce.

The portfolio report should show completed actions, unresolved risks, measured outcomes and the next decision for each home. It should also show the capacity needed to maintain new systems. An estate can accumulate more complexity than its maintenance arrangements can support. A well run programme leaves each home with equipment and records it can use, while the centre learns where to invest next. That is a stronger foundation for releasing resources to care than a single group-wide percentage target without a defensible baseline.

For the next leadership meeting

Two questions to resolve.

1. Would the proposed pilot answer a question that matters elsewhere in our estate?
2. Which apparent comparisons are being distorted by meter boundaries, tenure, occupancy or old data?