There is a version of retrofit that consists of putting a heat pump on the side of an otherwise untouched house and calling the job done. It is popular because it is visible, grant-friendly and quick. It is also, in most older properties, an expensive way to heat the sky.
Fabric first inverts the order. Reduce what the building loses, then size the system that replaces the remainder. The measures are not exotic and the sequence is not complicated — but the sequence is where most of the value sits, and it is the part most often got wrong.
The argument in one paragraph
A heating system exists to replace heat the building is losing. The rate of that loss is a property of the fabric: insulation, airtightness, glazing, and the junctions between them. If you install the heating first, you size it for the building as it stands — the worst it will ever be. Improve the fabric afterwards and you own a system that is now oversized, which in a heat pump means it cycles, runs inefficiently, and costs more to run than a correctly sized unit would have.
Do it the other way round and every downstream decision gets smaller and cheaper. Less plant, smaller emitters, lower running cost, and a building that stays comfortable for longer when the heating is off.
What "fabric" actually covers
Four elements, in rough order of how much they usually matter.
Roof and loft insulation. Almost always the cheapest heat you will ever save, and the first thing to check. Detail matters more than depth: insulation that blocks the ventilation path at the eaves creates a condensation problem in the roof space.
Walls. The largest surface area in most houses and the biggest single decision in a retrofit. Solid walls can be insulated internally or externally. External keeps internal floor area and treats the junctions more forgivingly, but it changes the appearance and is not always possible on a street elevation. Internal preserves the exterior and costs you room dimensions, and it demands careful detailing at every floor, ceiling and party-wall junction, because each of those is a route around the insulation.
Floors. Suspended timber floors are frequently the largest unaddressed heat loss in a period property, and the least visible once dealt with.
Airtightness. The one people underestimate. Uncontrolled air leakage undermines every other measure — insulation only works on air that stays put. Sealing is cheap relative to its effect, and it is the measure most likely to be skipped because there is nothing to point at afterwards.
Glazing sits slightly apart. It is usually a modest share of total heat loss but a large share of perceived comfort, because a cold single-glazed pane radiates away from you noticeably. Worth doing, rarely the first thing to do.
The ventilation trap
Here is the part that turns a good retrofit into a bad one.
Older houses are ventilated by accident. Draughts around windows, open chimneys, gaps in floorboards — all of it leaks heat, and all of it also carries moisture out of the building. A household generates a great deal of moisture: cooking, washing, drying clothes, breathing. In a leaky house that moisture leaves without anyone thinking about it.
Seal the building without replacing that airflow deliberately and the moisture stays. It finds the coldest surface available and condenses there. The result is damp and mould in a house that never had either, and the retrofit gets the blame.
The fix is not to leave the house draughty. It is to design the ventilation at the same time as the fabric, so that air changes happen by intent rather than by accident. In practice that means continuous mechanical extract, or mechanical ventilation with heat recovery where the airtightness justifies it. Both are unglamorous. Both are the difference between a retrofit that works and one that creates a problem.
If you take one thing from this piece: insulation and ventilation are a single decision, not two.
Then, and only then, the heating
Once demand is known — genuinely known, from an improved building rather than an assumed one — the heating specification follows almost mechanically. A lower-demand building needs less plant. Emitters can run at lower flow temperatures, which is precisely the condition under which a heat pump performs well.
This is why the order produces a better outcome even for people whose primary goal is a heat pump. Fabric first is not an argument against low-carbon heating. It is the thing that makes low-carbon heating work properly.
Solar generation sits at the end of the queue for a similar reason. It is worth doing where the roof orientation, pitch and shading genuinely suit it, and worth declining where they do not. Panels that will not pay back are not a green choice.
Why this is unfashionable
Fabric-first work is largely invisible. Nobody photographs their airtightness membrane. There is no product on the wall to point at, no app, and no obvious moment where the house looks upgraded.
It is also harder to sell, because the benefit is a negative — a cost that does not occur, a boiler that does not run as long. Technology is easier to market because it is a thing you can put in a brochure.
That asymmetry is worth being suspicious of. When a retrofit proposal leads with equipment and treats insulation as an optional extra, it is usually optimising for what is easy to sell rather than what will reduce the bills.
What good looks like
A well-run fabric-first retrofit tends to share a few features.
- A survey and an energy model before anything is specified, so the scope follows the building rather than a template.
- A written scope with the sequence made explicit, and ventilation designed alongside the fabric rather than after it.
- Inspection and photography at each stage, because the things that determine performance disappear behind plaster and cannot be checked later.
- A post-works reassessment against the same model, so the result is measured rather than asserted.
- Someone willing to tell you which measures are not worth doing on your particular property.
That last one is the tell. A specification where every available measure is recommended is a sales document. A specification with something crossed off it, and a reason given, is an assessment.
We have written up how we approach retrofit in more detail, including the standards question and how works are inspected. If you would rather see the outcome than the method, the project write-ups carry the scope and the results for each property.
Written by Greenish Property. General information, not advice — take independent professional advice before acting on anything here.
