Short answer. Work round the window with a strip of tissue on a windy day and identify the location. Draughts down the sides mean worn staff and parting beads. A draught at the meeting rails means the sashes have dropped out of alignment. A draught at the bottom rail means the sash is not sitting square on the cill. A draught around the outside of the box frame is failed pointing between the frame and the masonry — a builder's job, not a sash job. Only once you know which, does the fix make sense.
Why the gaps matter more than the glass
The instinct with a cold room is to blame the single glazing. The measurements say otherwise. Historic England's guidance on draught proofing windows and doors states that less than a quarter of the heat lost through a typical traditional window escapes by conduction through the glass — the rest is lost as draughts. Its wider advice puts around one-fifth of a home's heat loss through the windows, most of it through air gaps rather than the glazing, and attributes 15–20% of the heat lost in older buildings to draughts generally.
There is a second effect that does not show up in a heat-loss calculation: moving air makes people feel colder than the air temperature warrants, so they turn the heating up and run it longer. That is why draught proofing is bought for comfort rather than as an energy investment. If it is the cold glass surface you are reacting to, that is a different measure — see secondary glazing versus draught proofing.
Finding where the draught is coming from
This takes five minutes per window and it is the most useful thing you can do before getting a quote. Pick a windy day, because a still day will tell you very little.
- Close the window fully and turn the fastener. Close the internal doors to the room.
- Turn off any extract fan and close any trickle vents, so the room is not being deliberately ventilated while you test. Open them again afterwards.
- Tape a strip of light tissue paper to a pencil so it hangs freely. A stick of incense works too if you would rather watch smoke. Do not use a naked flame near curtains or loose paper.
- Hold it about 20mm from the joint and move slowly along each of the six locations below, pausing at each. The tissue lifts or the smoke bends where air is moving. The back of a damp hand is a good second opinion and costs nothing.
- Write down which locations moved the tissue, and take that list to whoever quotes.
The six places to check
| Location | What you will notice | What it means |
|---|---|---|
| Meeting rails (the joint across the middle) | Air at chest height; often the strongest single source | The sashes have dropped, so the splayed rails no longer wedge together. Cords stretched, cords broken, or the wrong weights in the box. |
| The sides of the sashes, full height | A steady draught the whole way up both edges | The staff bead and parting bead are the only things bridging the running clearance, and they are worn, shrunken, painted over or missing. |
| Bottom rail onto the cill | Cold air across the floor, worse in wind | The sash is no longer sitting down square on the cill, or the cill itself is decayed or worn hollow. |
| Top rail against the head of the box | Easily missed; check with the tissue held high | The upper sash has dropped away from the head, usually for the same reason the meeting rails have parted. |
| The weight pockets and around the axle pulleys | Air coming out of the pulley slot itself, or from behind the architrave | The box frame is a hollow cavity, not a sealed one. It connects to the outside and, in some houses, to the wall construction behind. |
| Around the outside of the box frame | Air at the junction of the frame and the wall, behind the architrave or under the internal cill board | Cracked, loose or missing pointing or mastic between the box frame and the masonry opening. Nothing whatsoever to do with the sashes. |
The glass-to-timber putty line is worth a seventh look. Historic England lists cracked, loose or missing putty among the core window defects, and once it has gone the pane is a genuine air path as well as a source of rattle. See rattling sash windows for how to tell loose glazing from a loose sash.
Draughts around the frame, not the sash
This deserves its own section because it is the most commonly misdiagnosed fault on the whole window, and because it is the one that no amount of brush pile will ever cure.
A box sash frame is bedded into a masonry opening and the joint between the two is pointed, or latterly mastic-sealed. Historic England's schedule of window defects lists cracked, loose or missing pointing between the frame and the opening explicitly. Mortar shrinks and falls out; mastic hardens, loses adhesion and lifts away from the timber. Behind an architrave or under an internal cill board, the resulting gap can be wide and completely hidden.
If your tissue moves at the wall junction rather than at the sash, sealing the sashes will change almost nothing and you will reasonably conclude that draught proofing does not work. The fix is repointing or re-sealing the perimeter, which is a builder's job. Ask any firm quoting you whether they check the frame-to-masonry joint as part of the survey — a specialist who checks it, tells you when the sashes are not the problem and prices the perimeter work separately is a better bet than one who quotes for pile without looking.
The right fix for each location
Sequence matters. Historic England's guidance is explicit that windows should be assessed for repairs before any draught proofing is carried out, and notes that straightforward repairs alone can reduce air infiltration and heat loss by up to a third. Sealing over a dropped sash or a rotten cill traps the fault behind new beads.
Meeting rails
Alignment first, seal second. The sashes come out, all four cords are renewed, the weights are checked and adjusted, and the sashes are eased so the rails come back into close contact when the window is rebuilt. Only then is brush pile routed into the meeting rail — usually the top rail of the lower sash. Fitting a seal to rails that are still out of line simply makes the window stiff without making it tight.
The sides of the sashes
Both beads are replaced with versions carrying brush pile in a carrier: a new parting bead in the groove in the pulley stile for the upper sash, a new staff bead for the lower. Historic England is unambiguous that wiper seals are the only way of sealing the sides and meeting rails of sliding sash windows, which is why pile rather than foam is correct here. It is also why a self-adhesive strip stuck to a worn bead rarely satisfies anyone: the gap varies in width along its length and the bead itself is the problem.
Bottom rail and cill, top rail and head
These two positions differ in kind: the sash closes against the surface rather than sliding along it. Historic England's guidance allows compression seals at the bottom and top rails of a sash window, and firms genuinely split on this — some fit a hard-wearing compressible seal at the head and cill while using brush pile only at the meeting rail and beads, others use pile at all five positions for consistency and low drag. Both are defensible, so it is a fair question to put to your installer rather than a right-and-wrong. Any decay in the cill is spliced or replaced first.
Weight pockets and pulleys
Very few standard specifications include this and it is worth asking about. The box is a void connected to the outside and it can behave like a small chimney. Lining or sealing the box interior during the overhaul is normally priced as a separate item — figures of roughly £50–£100 a window are published for internal window-box lining. If your tissue lifts at the pulley slot, raise it before you accept a quote.
Around the frame
Repoint or re-seal the perimeter. Priced separately by everyone, because it is not sash work.
If you are doing it yourself
Self-adhesive brush pile is the only stick-on product genuinely suited to a sliding sash. The Energy Saving Trust states plainly that foam strips do not work well on sliding sash windows; Historic England notes that foam has a short life; Historic Environment Scotland says brushes are generally thought to work better and last longer. Our DIY guide covers what each option costs and where it stops. If the building is listed, Historic England treats simple self-adhesive strips as not requiring consent while naming grooving as a method that may require it — the detail is here.
How much you should not seal
An old building needs air moving through it, and this is the part of the subject the trade is least keen to discuss. Historic England's guidance is quantified and worth quoting properly:
- Rooms used for living and sleeping need about 0.4 air changes per hour to keep the air fresh.
- For older buildings in reasonable condition, between 0.4 and 0.8 ac/h is recommended. Because air infiltration in many older buildings exceeds this, draught proofing is normally beneficial — the point is to get down to a sensible rate, not to zero.
- Take special care where there is an open fire or any combustion appliance. Seek specialist advice before sealing any room containing a gas or oil burning appliance.
- Never draught-proof a roof void or a sub-floor void. They are designed to be cross-ventilated. Damp cellars may not benefit either.
- Older buildings need to breathe; ventilation is what allows moisture to evaporate away through permeable finishes.
In practice that means keeping a controllable route for air — a vent you can open, an extract fan in the kitchen and bathroom, or a sash stop fitted above the lower rail so the window can be left open a fixed amount securely. It does not mean leaving the windows leaking.
What to expect afterwards
Realistic expectations, taken from measurement rather than marketing. Historic England reports that research has shown draught proofing can reduce air leakage from windows by 33–50%, and cites testing in which a sash window in good condition achieved as much as an 86% reduction in air infiltration — in the underlying Historic Scotland research the tested window went from 3.5 to 0.5 m³/h at the reference condition. Those are air leakage figures, not total household heat loss figures, and anyone who blurs the two is worth questioning.
What will not change is the temperature of the glass. In the same testing the window's U-value went from 4.5 to 4.2 W/m²K, which the researchers noted was not statistically significant against a measurement uncertainty of ±5.5%. Roughly 72% of the heat lost through the window still goes through the glazing. Draught proofing stops air moving; it does not insulate. If you want the glass figure to move, the same research measured secondary glazing cutting heat loss through the glazing by 63%, timber shutters by 51% and heavy curtains by 14%.
On money, the Energy Saving Trust currently publishes around £55 a year in Great Britain for draught proofing windows, doors and floors across a whole home. That is a generic whole-house figure, not the payback on a house full of overhauled box sashes at several hundred pounds a window — see our cost breakdown and what the service includes. Sash draught proofing is bought for comfort, rattle, dust and keeping original windows in service; the energy saving contributes but is not the business case.