Water is coming through concrete. Here's how to work out why.

Before anything else, rate what the water is doing

Two observations decide most of what follows: how fast the water is moving, and whether you can point at a line. Both are free, both take ninety seconds, and between them they eliminate more wrong answers than any tool we own.

Flow rate is the first filter because it dictates the resin family. Nothing about a repair is decided until this is settled, and any contractor who quotes without asking is quoting a product rather than a repair.

The second filter is width. Hold a business card against the crack — its edge is roughly 0.3 mm. A credit card edge is roughly 0.8 mm. If the card edge sits proud of the crack, you are under 0.3 mm and into the territory where microfine cement grout physically cannot enter (as a field rule the opening needs to be three to five times the grout's d95 particle size, which puts the floor at about 0.3 to 0.5 mm) and where a low-viscosity resin is the only thing that will travel. If you can get a fingernail in it, you are past 2 mm and the question changes from what to inject to why it is that wide.

• DRY — no moisture, crack dormant, the concern is strength not water. This is epoxy territory, and only if the movement has genuinely stopped.
• DAMP — staining, a cool patch, no water actually moving. Hydrophilic polyurethane or acrylate gel, both of which bond to damp concrete.
• WEEPING — beads forming on the face, a slow trickle. Hydrophilic if the crack stays wet year-round, hydrophobic if it dries out every winter.
• RUNNING — continuous flow you could catch in a cup. Hydrophobic polyurethane with an accelerator dosed to gel in seconds.
• GUSHING — flow under pressure. Hydrophobic with a very fast gel, and usually a hydraulic cement plug or oakum packing first to knock the flow down before any resin goes near it.

The seven ways water gets through concrete

Every water ingress problem we attend in South East Queensland resolves to one of these seven. Several jobs have two or three of them running at once, which is why the diagnosis matters more than the product.

Read the table left to right and work out which row you are in. If you are in more than one, you are in the row furthest down the list — the broadest path wins, because sealing the narrow ones simply pushes water into the wider one.

Which four injection actually fixes, and what that costs

Paths one to four have something in common: they are defined routes through the element, so resin has something to grip and somewhere to travel. Injection is the correct, permanent answer for all four, provided the element has stopped moving and the water is not being supplied by something that should be fixed upstream.

The figures below are indicative SEQ bands for accessible work, exclusive of GST. They move with element thickness, flow rate, access, and whether render or paint has to be stripped and reinstated — you cannot inject what you cannot see, and exposure is quoted separately or the job loses money.

Below every one of these numbers sits a site attendance minimum of roughly $450 to $900. No legitimate contractor can travel, set up a pump, inject, and return to cut back and make good for less than that, which is why a single short crack prices as a minimum job rather than at a linear-metre rate.

Path seven: when it is not a leak

This is the path most often missed, and it is the reason we ask about timing before we ask about cracks. Three different problems masquerade as water coming through concrete, and each has a different trade attached to it.

Condensation. Damp that appears on cold surfaces in still, humid, poorly ventilated rooms — undercrofts, plant rooms, storerooms, a closed-up garage — with no correlation to rain. Below-ground concrete sits close to ground temperature all year while summer air in South East Queensland is far warmer and carries far more moisture. When that air touches a surface colder than its dew point, water forms on the concrete. It is real water and it will grow mould, but no amount of injection will stop it. The fix is ventilation, dehumidification, insulation, and cutting the internal moisture sources.

A leaking service. A weeping wall above a pressurised water main, a corroded sewer line, or a blocked stormwater pipe is a plumbing job. It is quick to rule out with a pressure test and expensive to get wrong — we have attended walls that had been injected twice before anybody isolated the meter and watched the dial.

Drainage. Blocked or absent agricultural drains (the perforated pipe in gravel behind a retaining wall or beside a footing), rendered-over weep holes, a driveway falling toward the building, a downpipe discharging at the footing, no drainage cell behind a wall. This is the actual cause behind a very large share of enquiries that arrive asking for injection by name. Sealing the structure without restoring drainage raises the water head against it — and on a retaining wall, raises the load on a structure that may never have been designed for saturated backfill.

The 48-hour plastic sheet test, run properly

This is the single most useful thing you can do before anyone attends, and it costs about four dollars. It separates condensation from water coming through the element, which is the fork that decides whether you need a waterproofing contractor at all.

The principle is simple. Tape a sheet of plastic tightly over the concrete and leave it long enough for moisture to declare itself. If moisture collects on the room side of the plastic, humid air is condensing on a cold surface. If moisture collects on the concrete side, underneath the sheet, water is moving through the element. If both sides are wet, you have both problems, which in South East Queensland between November and March is common.

1. Cut three sheets, 500 x 500 mmClear plastic works best because you can read the result without lifting anything — a cut-open freezer bag or a piece of builder's film is ideal. Three sheets, not one: the first goes on the damp patch itself, the second on apparently dry concrete about a metre away, and the third on an internal wall well away from the problem as a control.
2. Tape all four edges completelyCloth or duct tape, edge to edge, with no gaps. A sheet taped on three sides tells you nothing because room air gets underneath it. Press the plastic flat against the concrete so there is no air pocket.
3. Write the date and time on each sheetUse a marker on the plastic itself. Photograph all three at the start. Note the weather — this test is far easier to read across 48 dry hours than through a storm, because rain confounds the result by supplying both moisture sources at once.
4. Leave it 48 hours, undisturbedDo not lift a corner to check at 12 hours. Keep the room in its normal condition — do not run a dehumidifier, open a window that is usually shut, or heat a space that is usually cold, because all three change the answer.
5. Read it without removing itMoisture beads on the room-facing surface means condensation. Moisture trapped between the plastic and the concrete, or a dark wet halo on the concrete under the sheet, means moisture is coming through the element. A darkened patch under the sheet on the control wall as well means you have a broader moisture issue than a single leak path.
6. Photograph the result and keep itWhoever attends will want to see the three sheets. It is thirty seconds of evidence that saves an hour of speculation and, on a strata job, it is the first document in a defect file.

The crack you can see is often not where the water gets in

This is the finding that separates a real investigation from a look-and-quote, and it is counter-intuitive enough that it costs people real money every year.

Water entering a below-ground structure does not travel in a straight line. It arrives at the outside face, finds the interface between the membrane and the concrete — or between the structural slab and the topping, or between the wall and its render — and travels sideways along that interface until it reaches the weakest point through the element. It then exits there. On a suspended car park deck with a separate topping slab, water routinely enters at one point and appears metres away. On a basement wall, the visible crack inside is very often not the entry at all; it is simply the easiest exit.

The practical consequence is that mapping matters more than injecting. If you seal only the exit, you have not intercepted the water, you have moved it — and the next exit may be behind a wall lining, into a lift shaft, or into somebody else's lot. This is exactly why the kicker joint, the line where the wall meets the floor slab, is the real path in most leaking basements even when a crack a metre up the wall is what everyone is pointing at. Chase the joint, not the crack.

Why South East Queensland leaks the way it does

Four local conditions shape almost every job we attend, and none of them appear on the American pages that currently dominate search results for these questions.

The wet season, November to March. South East Queensland does not get evenly distributed rain; it gets months of it concentrated into a summer window, and the enquiry curve for active leaks follows that window rather than sitting flat across the year. A crack that weeps politely in August will run in February. This matters for resin selection: a wall that leaks all summer and dries out completely by June is a wall where hydrophilic resin can desiccate, shrink, and let the water back in, which is why a seasonal leak is usually a hydrophobic job even though it is only damp on the day we look at it.

Reactive clay. Much of the region sits on clay soils that swell when wet and shrink when dry, and that annual cycle moves footings, slabs and retaining walls. It is a principal reason cracks in SEQ are active rather than dormant, and it is the reason we fit crack monitors and read them across 4 to 12 weeks — a full moisture and temperature cycle — rather than injecting on the first visit whenever movement is in question. A rigid repair in a moving crack simply cracks again a few millimetres away, which looks like our failure and is very hard to explain afterwards.

Post-2022 flood voidage. The February 2022 flooding moved a great deal of soil. Where floodwater or a subsequent leaking service has washed fines out from under a slab or behind a wall, you get voids — and a void behind the element will swallow effectively unlimited resin without ever sealing the face. On any job where a void is suspected we set a hard volume budget per packer, typically one to two litres, and stop rather than pumping a drum into the ground. It also means some cracks that appeared after 2022 are settlement cracks, and settlement is a geotechnical problem, not a waterproofing one.

East-coast lows. A single thunderstorm loads a wall for twenty minutes. An east-coast low loads it for days. Sustained saturation is what raises the water table, fills the drainage layer behind a retaining wall and turns a damp patch into running water — and it is the reason we ask to observe a repair across a rain event, not just for 24 to 48 hours of fine weather, before we call it verified.

What a real investigation involves

A quote-only visit is a look, a measure and a price. That is appropriate and usually free where the path is obvious — one crack, one wall, water running out of it. Where it is not obvious, an investigation is a piece of work with a deliverable, and charging for it is what allows it to be done properly.

Here is what one actually consists of. Not all of it applies to every job; the first two apply to all of them.

1. Moisture readings at marked pointsReadings taken at physically marked positions so the same points can be re-read after the repair. This turns 'it seems drier' into a number, and it is the only before-and-after evidence that survives a dispute.
2. Exposure of the actual defectPaint, render and wall linings hide the true extent of a crack or joint. Nobody can price what they cannot see, so exposure is quoted as its own item. A crack that looked like 1.2 metres under paint is regularly 3 metres once the render is off.
3. Crack monitors read over 4 to 12 weeksA tell-tale gauge fixed across the crack with a scale on it, read across a temperature and moisture cycle. This is cheap insurance against the most expensive mistake in the trade — a rigid repair in an active crack — and it is the step most contractors skip.
4. Water or dye testingControlled wetting of a suspected entry zone while somebody watches the dry face, or dye introduced upstream. This is how lateral tracking gets proven rather than assumed, and it is the only way to establish that the crack you can see is or is not the entry point.
5. GPR scanning where anything gets drilledGround-penetrating radar maps the reinforcement and any post-tensioning tendons before a hole goes anywhere. On a post-tensioned slab this is not optional and neither is written engineer approval — cutting a tendon is not a recoverable mistake.
6. A written report with a recommendation you can tenderThe diagnosis, what was ruled out and why, the recommended method with packer spacing and resin family named, and the things that would change the answer. If it is written well enough that three contractors could quote the same scope from it, it is worth what it cost.

Take your symptom to the right page

Find the row that matches what you are actually looking at. Each destination page carries the method with real numbers, the honest limits, and an indicative cost band for that specific situation.

What it costs

Indicative only, and excluding GST. Build a line-by-line estimate for your own job with the price book, or read the working behind the rates in the 2026 cost guide.

Questions we actually get asked

How much does it cost just to find out what is wrong?

A quote-only site visit is commonly free, and for a single obvious crack that is all that is needed. A genuine diagnostic investigation — moisture readings at marked points, crack monitors installed and read, water or dye testing, GPR scanning where drilling is contemplated, and a written report with a tenderable recommendation — sits at roughly $450 to $1,500 plus GST. An engineer's structural assessment is a separate engagement again, typically $1,200 to $3,500 and up. Charging for real diagnosis is defensible; it is also why the diagnosis gets done rather than guessed.

Water is coming up through the floor rather than out of a crack. Can that be injected?

Sometimes. If it is arriving along a defined line — nearly always the kicker joint where the floor slab meets the wall, which is the single most common path in a basement — then yes, that joint is injectable, with packers at roughly 200 to 400 mm centres and a typical production rate of 8 to 15 linear metres per crew per day. If water is coming through the body of the slab across a broad area, there is no line for resin to grip and injection is the wrong tool entirely. That situation is a negative-side tanking system, a drained cavity floor, or getting the groundwater level down with drainage and a sump. The plastic sheet test on the floor will usually tell you which of the two you have.

My builder says it is just shrinkage cracking and nothing to worry about. Is that right?

Often it is. Concrete shrinks as it cures and fine cracking is expected — AS 3600, the Australian concrete structures standard, works to serviceability crack widths in the range of roughly 0.10 to 0.40 mm depending on the element and its exposure classification. So the existence of a crack in that band is not itself a defect. The questions that matter are different ones: is water coming through it, and is it getting wider? A dormant hairline crack in a dry internal wall genuinely needs nothing. The same crack passing water into a below-ground space needs sealing, because water sitting against reinforcement is a corrosion problem later, and corrosion is a far more expensive scope than injection.

There is white powder on the wall but it feels dry. Is that a leak?

That is efflorescence — salt left behind when water carrying dissolved minerals evaporates off the face. It is proof that water has been moving through, not proof that it is moving today. In South East Queensland it is very common to find a wall with heavy efflorescence in August that ran freely in February. Two practical consequences: first, it tells you there is a real path even when the wall is dry, so do not dismiss it; second, efflorescence and calcite deposits inside the crack physically block resin penetration and have to be water-flushed out through the packers before injecting, or the resin will simply not travel.

It only leaks in summer and dries out completely by winter. Does that change the repair?

It changes the material, which is the most important decision on the job. Hydrophilic polyurethane holds water permanently inside the cured polymer, which is what makes it bond so well to damp concrete — but if the crack dries out completely for months, that entrapped water evaporates and the foam shrinks. In theory it re-swells when the wet season returns, commonly quoted at two to four times; in practice a shrunk foam can lose its key in the crack and never fully recover, and the leak comes back. So a seasonal SEQ leak is normally a hydrophobic job, because hydrophobic foam uses water to cure but does not incorporate it, and stays dimensionally stable through wet and dry cycling. Tell whoever quotes you that the wall dries out — it should change what they propose.

The garage floods after every big storm. Can you just seal the wall?

Probably not, and sealing it might make things worse. Flooding that correlates with heavy rain is usually a water management problem rather than a concrete problem: a blocked or non-existent subsoil drain, a stormwater line that cannot cope with the volume, surface water running down a driveway toward the building, a downpipe discharging straight at the footing, or a failed sump pump. Sealing cracks in that scenario raises the water level against the outside of the wall, and on a retaining wall it increases the load on a structure that may not have been designed for saturated backfill. The first thing we look at on a flooding enquiry is where the water goes, not where it comes in.

How do we know whether we need an engineer before we call a waterproofer?

For ordinary leak sealing in a basement wall, lift pit, planter box or car park joint, you do not. You do need one when the cracking might be structural rather than a durability and water issue: diagonal cracks, cracks that widen toward the top, any lean, bulge or step cracking in a retaining wall, a change in floor level, or cracking in beams, columns or other primary members. You also need one to approve any drilling into a post-tensioned slab — which must be GPR-scanned first — and to nominate a maximum safe injection pressure in writing on thin or slender elements such as tilt panels, which are commonly only 150 to 200 mm thick. Any contractor worth engaging will tell you that before quoting, not after.