Key takeaways
- Most heat pump customers are comparing against an old gas or oil bill: 53,190 grant-funded installs to September 2026 replaced gas and 17,184 replaced oil[1].
- Measure heat out and electricity in, plus flow temperature, starts and defrosts; without those you are guessing.
- Seasonal performance is heat delivered divided by electricity used over a full year, not a figure read off the controller on a mild day.
- Treat a performance complaint as a fault investigation with a written outcome, not a debate about the weather.
- Agree in writing what monitoring data you will see and why before you connect to a customer's system.
Why performance data matters more than it used to
A heat pump customer judges the job by two numbers: how warm the house is and what the electricity bill says. The second is compared, consciously or not, with the old fuel bill. Of the Boiler Upgrade Scheme installations paid out by the end of September 2026, 53,190 displaced gas heating and 17,184 displaced oil[1]. Those customers have years of bills to compare against, and they will.
Source: Department for Energy Security and Net Zero, 2026[1]
Show the data
| Item | Value |
|---|---|
| Displaced gas heating | 53,190 |
| Displaced oil heating | 17,184 |
Without data, a complaint about a high bill turns into an argument about the weather, the tariff and the customer's habits. With data, it becomes a short investigation with an answer. Monitoring also protects you: if the system is performing as designed, you can show it.
What to measure
| Reading | What it tells you | Where it comes from |
|---|---|---|
| Electricity used by the heat pump | The cost side of efficiency | Dedicated electricity sub-meter, or the controller's estimate |
| Heat delivered | The output side of efficiency | Heat meter on the flow and return, or the controller's estimate |
| Flow and return temperatures | Whether the system runs as cool as it was designed to | Controller or monitoring sensors |
| Compressor starts and run hours | Short cycling caused by oversizing, zoning or low flow | Controller log |
| Defrost cycles | Whether defrost is normal for the weather or excessive | Controller log |
| Hot water energy | How much of the bill is hot water and the high-temperature cycle | Controller split or separate meter |
| Room temperatures | Whether comfort matches the design | Room sensors or the customer's thermostat |
Controller estimates are a useful start but they are estimates. Where a customer is likely to query performance, or you want evidence for your design approach, fit a proper electricity sub-meter and a heat meter at installation. It is much cheaper to fit them then than to come back later.
Metering and monitoring options
- Manufacturer app or cloud portal: easy to set up, and often lets you see faults remotely, but the data and the access belong to the manufacturer's platform.
- Dedicated electricity sub-meter: gives an accurate electricity figure for the heat pump alone, separate from the rest of the house.
- Heat meter: measures flow rate and temperature difference to calculate heat delivered. Needs correct placement and commissioning.
- Independent monitoring platform: combines meter data and controller data and can show seasonal efficiency over time; check how data is stored and shared.
- Manual readings at each service: better than nothing, and the minimum for any system on an aftercare plan.
Connectivity is often the weak link. In early 2020, 18% of adults in Great Britain used internet-connected energy or lighting controls, rising to 26% of those aged 35 to 44[2]. Plenty of customers will need you to set up the app with them at handover, and some will not want it at all. Respect that, and agree a manual reading routine instead.
Before you access a customer's system remotely, agree in writing what data you will see, why, and how long you keep it. It is their home and their data.
Working out real efficiency
The figure that matters is seasonal performance: total heat delivered divided by total electricity used over a full year. A reading off the controller on a mild afternoon will flatter the system; one taken during a defrost will do the opposite.
Example (fictional): over twelve months a heat pump delivers 9,000kWh of heat for space heating and hot water and uses 2,900kWh of electricity. Seasonal performance is 9,000 divided by 2,900, about 3.1. The design estimate was 3.4. The gap is worth investigating, and in this Example the controller log shows a flow temperature well above the design figure for most of the winter because the compensation curve was never adjusted after commissioning.
Always compare like with like. If a customer heated two rooms with an oil boiler and now heats the whole house to 21°C, the bill comparison will mislead both of you. Ask how they use the heating now and how they used it before.
Resolving a performance complaint
Listen and log
Record what the customer is unhappy about: comfort, bills, noise or hot water. Note the dates and the weather if they mention it.
Collect the evidence
Ask for recent electricity bills and the old fuel bills, and pull the meter and controller data for the same period.
Compare against the design
Check flow temperatures, room temperatures and seasonal performance against what you designed for.
Check the settings
Compensation curve, schedules, hot water cycle frequency, immersion use and any setbacks the customer has added.
Check the hydraulics
Flow rate, filters, air in the system, zone valves closing on the heat pump and any buffer or low-loss header mixing.
Fix and explain
Put right what is yours to fix, explain what is down to usage or tariff, and write it up for the customer.
Follow up
Look at the data again after two or three weeks of similar weather and share the result.
Your legal baseline is simple: every service contract is treated as including a term that the work is done with reasonable care and skill[3]. A system that cannot reach its design performance because of an installation or commissioning error is yours to put right. A bill that is higher because the household now heats every room for longer is not, but it still deserves a clear explanation.
Common causes of poor performance
- Flow temperature set higher than the design, often to fix one cold room.
- A flat or untouched weather compensation curve.
- Zone valves or thermostatic valves throttling flow so the unit short cycles.
- A buffer tank or low-loss header mixing return water into the flow.
- The immersion heater doing work the heat pump should do, or a high-temperature hot water cycle running every day.
- Emitters undersized in one or two rooms, pushing the whole system hotter.
- Clogged filters or strainers reducing flow after the first few months.
Most of these are settings, not hardware. That is the strongest argument for remote monitoring: you spot them in the first winter, often before the customer does.
Build monitoring into every install
The cheapest time to make a system measurable is the day you fit it. Add the monitoring to your standard specification so it is not an optional extra the customer has to be sold on separately.
- Specify the electricity sub-meter and, where the job justifies it, the heat meter in the design, with their positions on the schematic.
- Record baseline readings at commissioning: flow and return temperatures, flow rate, pressure and controller settings.
- Photograph the controller settings screen by screen and keep the photos in the job file.
- Set up the app or monitoring account with the customer present, and confirm who has access.
- Put a reminder in the diary to review the data after the first cold fortnight.
- Store each year’s readings against the property so the next engineer can see the history.
None of this needs expensive software. A consistent job file and a calendar reminder catch most problems early, and the records become your evidence when a customer, manufacturer or certification body asks how the system has performed.
What to tell the customer at handover
Set expectations before the first bill arrives. Explain that a heat pump runs for longer at a lower temperature than a boiler, so radiators feel warm rather than hot, and that turning the heating off and on in bursts usually makes performance worse. Show them where to see daily electricity use, and tell them which numbers you will look at if they ever call with a concern. Customers who know what normal looks like report problems earlier and argue about them less.
Homes with more than one technology
Bills are harder to read when the heat pump sits alongside solar panels, a battery or a solar diverter. One in three MCS-certified installations in 2026 combined more than one renewable technology in the same home[4]. MCS heat pump notifications were 60% of total UK heat pump sales in 2024[5], so a large share of systems also come with certification records you can use as a baseline. Separate the heat pump's own electricity use with a sub-meter, otherwise a sunny month or a battery setting can look like a heat pump fault.
Feed what you learn back into design and aftercare. Our guides to heat pump design software and aftercare plans show how to turn monitoring into better installs and a steady service list.
Frequently asked questions
How do I measure a heat pump's real efficiency?
Divide the heat delivered by the electricity used over a full year, using a heat meter and a dedicated electricity sub-meter if possible. Short readings on mild days are misleading.
Are controller efficiency figures accurate?
They are estimates. Useful for spotting trends, but for disputes or design feedback fit proper metering at installation.
What usually causes a high heat pump bill?
Most often settings: high flow temperatures, an untouched compensation curve, short cycling, or the immersion doing the work. Undersized emitters in one room can push the whole system hotter.
Who is responsible if a heat pump underperforms?
Work must be done with reasonable care and skill[3], so installation or commissioning errors are the installer's to fix. Usage and tariff changes are not, but explain them clearly.
References
Department for Energy Security and Net Zero (2026). Boiler Upgrade Scheme (BUS) monthly statistics: England and Wales, September 2026 (opens in a new tab) ↑
Office for National Statistics (2020). Internet access - households and individuals, Great Britain: 2020 (opens in a new tab) ↑
legislation.gov.uk (The National Archives) (2015). Consumer Rights Act 2015, section 49 (opens in a new tab) ↑
MCS (Microgeneration Certification Scheme) (2026). Record number of UK households combining renewables to improve energy resilience (20 August 2026) (opens in a new tab) ↑
Heating and Hotwater Industry Council (HHIC) (2024). Bumper year for UK heat pumps (opens in a new tab) ↑