Matching Brickwork and Materials

Matching brickwork is the part of an extension that people assume is a shopping problem and that turns out to be a geometry problem, a mortar problem and a weathering problem, in roughly that order. The brick itself is the easiest of the four to get close on. Everything around it is what decides whether the finished wall reads as part of the house or as something stuck on the back of it.

This page is about how that decision is actually made and what it costs to make it well. It applies as much to a rear extension in stock brick as to a two storey side extension in reds.

The brick you need stopped being made around 1970

British brickmaking went metric at the turn of the 1970s, and the sizes moved.

A standard modern brick is 215mm long, 102.5mm wide and 65mm deep. Add a 10mm joint and it works to a format of 225 by 112.5 by 75, which is where the four courses to 300mm that every bricklayer works to comes from.

Older bricks are bigger. Imperial work is commonly around 8⅝ inches by 4⅛ by 2⅝, which is roughly 219 by 105 by 67, and in plenty of houses and regions the brick is a full three inches deep. Those are small differences on paper. On a wall they are not, because the eye is extremely good at spotting a repeated module changing size. A metric brick laid next to an imperial one looks short and shallow, and no amount of colour matching fixes it.

Imperial sized bricks are still made. They cost more, they come from fewer places, and the lead time is measured in weeks rather than days. That lead time is the thing that derails programmes, because it is discovered at the point the trench is already open.

Gauge, and why the courses will not line through

This is the arithmetic nobody mentions until the scaffold comes down.

Bricklayers set out vertically to a gauge: brick plus joint, marked on a rod. Modern work is four courses to 300mm. A great deal of imperial work is four courses to 12 inches, which is 305mm. Five millimetres per four courses.

Carry that up a wall. Thirty two courses, about two and a half metres, and the two walls are 40mm apart, which is more than half a brick. Every window head, every string course and every sill on the new work sits at a different height from the old. It is the single most common reason an extension looks wrong to somebody who cannot say why.

The fix is not a different brick. It is to set the gauge rod off the existing wall rather than off the standard, and to open or close the bed joints on the new work by a millimetre or two a course so the two arrive together at the points that matter: the damp proof course, the sills, the heads and the eaves. That is bricklaying rather than buying, and it is why the first thing worth doing is counting sixteen courses on the existing wall and dividing.

Do not assume 12 inches either. Gauge varied by region, by period and by whoever was on the job. Measure it.

The mortar is a sixth of what you are matching

Work the areas out and the joints turn out to be a bigger part of the wall than anybody assumes.

A brick face is 215 by 65. Its share of the wall, once the 10mm joints are counted, is 225 by 75. Divide one by the other and the mortar is a little over 17 per cent of everything you are looking at. Get the brick right and the mortar wrong and roughly a sixth of the wall is the wrong colour, in a fine grid, which the eye reads as a change of tone across the whole panel rather than as lines.

Mortar colour comes from the sand, not the cement. A yellow building sand and a red one produce visibly different walls from identical bags of cement, which is why local mortars look local. Changing sand supplier halfway up shows as a horizontal band forever, so the sand is worth buying in one delivery.

Strength matters as much as colour where you are cutting into old work. A soft handmade or stock brick set in lime mortar wants a mortar softer than the brick, so that movement and salts damage the joint rather than the brick face. Put a hard cement mortar against a soft old brick and the failure relocates into the face of the brick, which spalls, and that is not repointable. On the new extension itself a modern mix is right; where the two meet, the mix is a judgement about the old brick.

Consistency is a site discipline rather than a specification. Gauge boxes rather than shovel counts, and the same person mixing, is what stops a wall banding.

The joint changes the wall more than the brick does

Profile is shadow, and shadow is most of what you actually see from ten metres.

A bucket handle joint, tooled to a shallow concave, is the modern default. It is durable and it throws a soft, even shadow. Weather struck, cut back at the top and pressed out at the bottom, sheds water well and gives an older, crisper line, and it is what a lot of period work has. A flush joint reads flattest and shows the brick most. A recessed joint gives a deep dramatic shadow and holds water on the top edge of every brick, which is why it belongs on sheltered elevations and rarely on a British rear wall.

Set a bucket handle joint next to weather struck original work with a perfect brick match, and it still reads as new. Get the profile right with an imperfect brick and it often does not. If you only win one of these arguments, win this one.

Bond, and the courses that give it away

Old solid walls and new cavity walls are laid in different patterns, and the pattern is unmissable.

A solid wall a brick thick is bonded through its thickness, so headers appear: Flemish bond alternates header and stretcher along every course, English bond runs whole courses of each. A modern cavity wall has a single skin of brick tied back to blockwork, so it is laid in stretcher bond and there are no headers at all, because there is nothing for them to bond into.

Butt a stretcher bond extension against a Flemish bond Victorian house and the difference in rhythm is obvious even to somebody who does not know what they are looking at. Where it matters, the answer is snapped headers: bricks cut in half and laid to imitate the bond. It is slow, it wastes brick, and it works.

The details are the other giveaway. A soldier course over the openings, a projecting plinth at the base, a dentil or corbelled eaves course, segmental arches with proper voussoirs, a band of tile creasing. Reproducing one or two of those does more for the read of an extension than the brick colour, because they are the marks of the period rather than of the clay.

The near miss is the failure

Match it properly or change it deliberately. What fails is the attempt that lands in between.

The eye judges a surface against whatever is next to it. A brick that is close but not right invites the comparison and then loses it, every time somebody looks at the back of the house. A brick that is obviously different, or a cladding, or a render, stops the comparison from being made at all, because it is clearly a later addition and honest about it.

Planning officers usually take the same view, which surprises people. An addition that reads as an addition is easier to assess than a pastiche that nearly works, and in conservation areas a subordinate, clearly modern rear extension is often preferred to an imitation. What the officer is actually assessing in a designated area is worth reading before choosing materials.

Right on day one, wrong in year five

You are trying to match a wall that has had ninety years of weather, and you are buying a brick that has had none.

The existing wall carries soot, rain staining, algae on the shaded side and lime run under the sills. The brick in your hand is the colour that wall was when it was built. Match the sample to the weathered wall and the new work is a shade too dark on the day it is finished and will keep getting darker. Match it to the wall’s original colour and you accept a few years of visible difference while the two converge.

The mortar moves faster than the brick. New mortar lightens for a year or more as it carbonates, and efflorescence, the white bloom of salts brought out by drying, is normal in the first winter and usually weathers off. Judging a match in February on a wall built in November is judging it too early.

Two things not to do. Cleaning the old wall to bring it back to the new one, with acid or a pressure washer, opens the face of the brick and accelerates exactly the decay you were trying to hide. And a near match is sometimes corrected afterwards by tinting the new brickwork, which does work and is a real trade, but tinted brick ages on a different curve to the wall beside it.

Where the new wall meets the old

Two walls on two different foundations will move independently, whatever you do at the joint.

Toothing in, cutting alternate bricks out of the existing wall and bonding the new work into the gaps, gives an invisible junction and is the traditional answer. It also assumes the two walls move together. They do not: the new foundation is fresh, probably deeper, and on shrinkable Essex clay the two will settle and respond to seasonal movement on their own schedules. A toothed joint that moves cracks in a stepped diagonal through the new brickwork, in the most visible place there is.

The usual answer now is a straight vertical joint with a proprietary wall starter fixed to the existing wall, ties bedded into the new courses, and a sealed movement joint at the face. It is a deliberate line rather than a hidden one, and it stays straight instead of cracking. Where the wall is short and the ground is reliable, toothing can still be the right call; it is a decision, not a default.

Two things have to line up at that junction whatever the joint. The new damp proof course must connect to the existing one and to the floor membrane, with no break. And the cavity tray over any abutment has to discharge, with weep holes below it. Those are the details that decide whether the junction is dry, and neither is visible afterwards.

The sample panel, and the condition behind it

There is a planning condition about materials, and it is looser than people think.

Permitted development under Class A carries a condition that the materials used in exterior work be of a similar appearance to those used on the exterior of the existing house. Similar, not identical, and judged against the house as it stands today rather than as originally built. A conservatory is excepted from that condition altogether. The full set of Class A conditions is worth checking, because materials is only one of them.

Where a full application is made, particularly in a conservation area, the permission often arrives with a condition requiring samples or a sample panel to be approved before work starts. That is a real programme item. Build a square metre, lay it in the mortar you intend to use, tool the joint the way you intend to tool it, and get it signed off. Discharging that condition late holds the bricklayer off site, and it is entirely avoidable.

If you are not matching

The alternatives age differently and it is worth knowing how. Render is the cheapest visual reset and the most maintenance, and it wants a sound background and a bellcast at the base or it will stain and crack at every junction. Fibre cement or timber boarding above a brick plinth is the treatment that dominates contemporary rear extensions for a reason: the plinth ties the new work to the house at the level where the two actually touch, and the change of material happens above eye level. A deliberately dark brick does the same job in one material.

Whatever you choose, the junction is what sells it. Where the eaves line runs, where the new brick stops, and where the roof meets the old wall are what a passer-by reads without noticing they are reading it. The roof form drives most of those lines, which is why materials and roof are one conversation rather than two.

What to put in the specification

Name the brick and its size format, and say whether it is imperial or metric. State the bond. State the mortar mix, the sand, and the joint profile, and say who is approving the sample panel. Ask for the lead time in writing before the trench is dug. A quotation that answers those has been thought about. One that says “brick to match existing” has moved the risk onto you and left the decision to whoever is holding the trowel in eleven weeks’ time.

We set the gauge rod against the existing wall on the first morning of brickwork, because that is the last cheap moment to find out the two walls are not going to agree.

Permitted development conditions change from time to time and how they apply depends on your property and its designation. Treat this as an explanation of how the decision is made rather than as confirmation of what you may build.

Last reviewed August 2026. Planning and Building Regulations change, and your property may differ.