Drainage data for home inspectors

What arrives is not an opinion about the house. It is three numbers per wall, computed from a bare-earth model of the ground, with the uncertainty attached to each one.

The three quantities, and why an inspector wants them

1. Ground slope, per wall, not per house

Grade is not a property-level fact. A house can shed water cleanly off three sides and collect it on the fourth, and the fourth is the only one that matters.

So each wall is measured along its own outward normal. The slope is measured from 0.75 m to 3.25 m out, repeated at several stations along the wall and combined — so one bad reading, or a parked wheelie bin, cannot carry a wall on its own.

Nothing is sampled nearer than 0.5 m to a wall. Inside a footprint a bare-earth surface is not ground at all; it is the interpolation across a building that was removed.

The reference is the familiar one: a 5% fall over the first three meters. That is a widely used target rather than a code citation, and local requirements vary.

2. Contributing area — the part you cannot see from the yard

Standing at a wall tells you the last three meters. It does not tell you that two thousand square meters of hillside behind the fence drains into that corner.

Contributing area is computed by routing water across the whole surface and counting what arrives at each wall. It is sampled at nine points beside the wall and published with a range, because a single sample point on a flow boundary is a coin toss and printing a decimal point would be false precision.

3. Standing water, as a depth and an extent

Every closed hollow in the surface is filled to the point where it spills, and the depth it was raised is kept rather than discarded. That number is how deep water gets before it runs off, counted within 20 m of the footprint and outside every building. Anything under 3 cm is measurement noise and is not reported.

That paragraph is not legal throat-clearing, it is the product boundary, and it is the reason this is safe to sit beside an inspection report. Ground falling toward a foundation raises the risk of water reaching it — ordinary building science, not in dispute. That a particular house has water is a diagnosis, and this instrument does not make diagnoses.

The sentence that ships with the number

"the ground falls toward this structure on 14 m of its walls" measure · drainage · lead signal

Quotable verbatim. Not "this basement leaks", not "this house needs work" — a length of wall and the direction the ground tips at it.

Why a wall direction is only printed sometimes

A published building outline is a trace, and it can sit a meter from the wall it claims to be. Shifting a real Pittsburgh footprint north in 0.2 m steps moved one wall from +5.4% to −23.5%. It crossed zero inside 1.4 m, which is less than the outline error we concede in our own limits.

So the whole footprint is displaced one meter in each of sixteen directions and every wall is measured again — seventeen positions in all. A direction is printed only when all sixteen agree on the sign.

Walls that fail come back marked undecided, with which way they lean and how many of the sixteen agreed, so a near miss is still worth walking and is never dressed up as a finding.

The test asks the directions to agree, not the magnitudes. A meter of movement on steep ground can legitimately swing a grade a long way, which tells you the hillside is not a plane and nothing whatever about which way the water goes.

Which building, which is the failure nobody audits

The best terrain model in the world is worthless pointed at the house next door. Identity is decided by graded evidence, and the weakest grade — the nearest outline is all there is — locates rather than identifies: it is labeled that way in the payload, and it is never sold as a report.

We publish no headline identification rate, because one figure would hide the thing that decides whether an answer is safe to use: what identified the building on this address. Where the underlying records are thin the evidence is thin too, and the engine says so rather than guessing harder. And not every disagreement is a different building — some are our source and somebody else's tracing the same roof differently.

What it does not replace

• Anything below grade. Footing drains, tile, waterproofing membrane and the condition of a sump are invisible to a remote measurement.
• Gutters, downspouts and where they discharge. That is a visual finding and it is yours.
• Soil type and infiltration. A hollow in the surface is a shape, not a prediction that water sits in it.
• Anything about the interior. We never see it.

Questions inspectors ask