Most of the failed floors I'm called out to look at didn't fail because of the floor. They failed because of what was underneath it, and in this region that means moisture coming up through the slab.
It's worth understanding, because a moisture test costs very little and it is the single best predictor of whether your floor is still flat in five years — and it is routinely the first thing dropped from a bid trying to be the cheapest one you get.
What's actually happening under a Houston slab
Almost every home here sits on slab-on-grade: a concrete pad poured directly on the ground. Concrete is porous. Ground moisture moves up through it continuously as water vapor, and it doesn't stop — it isn't weather, it's a steady state.
Since roughly the mid-1990s, standard practice around here has been to lay a polyethylene vapor retarder under the slab before the pour. That sheet dramatically slows vapor transmission and it's why newer houses test well.
Slabs poured before that frequently have nothing underneath them at all. Which means a 1985 house in Conroe and a 2016 house in Tavola are, from a flooring standpoint, two different substrates that happen to look identical once the carpet is up.
Why it matters more than people expect
Every flooring adhesive published by every manufacturer has a stated moisture limit. Install above it and the warranty is void — but more to the point, the bond degrades.
The degradation is what makes this insidious. A floor bonded to a slab that's too wet doesn't fail on install day. It looks perfect. It looks perfect for a year. Then it starts sounding hollow underfoot in patches, and the patches spread, and eventually the floor moves when you walk on it. By that point the crew is long gone and the warranty conversation is difficult.
Wood is the least tolerant material — its limits are the narrowest and the failure includes the wood itself cupping or crowning. Glue-down LVP is next. A floating floor is the most forgiving, because it isn't bonded to anything, though trapped moisture underneath can still cause odour and, in bad cases, microbial growth.
The two tests, and why the difference matters
Calcium chloride (ASTM F1869) measures moisture vapor emission rate — how much moisture comes off the slab surface over a set period. A dish of anhydrous calcium chloride is sealed under a dome for 60–72 hours and weighed. It's an established test and it has one significant limitation: it only reads the top half inch or so of the slab.
In-situ relative humidity (ASTM F2170) measures humidity at depth inside the concrete. Probes go into holes drilled to 40% of the slab thickness, sit sealed for 24 hours, and then read.
That depth difference is the whole story. The top of a slab dries first. A slab can present a perfectly acceptable surface reading while the interior is still carrying a great deal of water — and once you seal the surface with flooring, that interior moisture redistributes upward and reaches the adhesive.
For anything bonded, and for any house with water history, in-situ is the test worth having. On a house near the East Fork in Porter that's taken water more than once, a surface reading is the number most likely to mislead you.
What the numbers mean
Manufacturers set their own thresholds and you should follow the one on your adhesive's data sheet, not a rule of thumb. As a general orientation:
- In-situ RH at or below 75% — most adhesives are comfortable. Many wood systems want 75–80% or lower.
- 75% to 85% — you're into product-specific territory. Some systems are rated for it; most wood is not.
- Above 85% — mitigation before anything goes down. This is common on older slabs here and it is not a reason to panic; it's a reason to budget.
Where to test, not just whether
One reading in the middle of a room isn't a moisture test. It's a data point.
Readings genuinely differ across a single slab, and where they differ is predictable. Perimeter areas read higher where the ground outside stays wet — under heavy tree cover in Pinehurst, on properties where guttering is absent or the grade falls toward the house, and on septic properties near the drain field. A central reading can come back fine while an exterior wall is well out of range.
We test at multiple points around the footprint for that reason. On an acreage house it routinely changes the answer, and it can turn what looks like a whole-slab problem into a localized one — which is a meaningful difference in what mitigation costs.
What mitigation involves
Where readings are high, a moisture mitigation coating goes down first: a two-part epoxy or similar system rolled onto the prepared slab that forms a vapor barrier on top of the concrete.
It's a real cost and it adds a day. It is also dramatically cheaper than replacing a floor, and unlike most of what goes into a floor, it is entirely invisible once finished — which is exactly why it's the line item most likely to be quietly missing from a competing bid.
What to ask a bidder
Three questions, and the answers tell you a lot:
- "Did you test the slab, and can I see the reading?" A number, written down. Not "it looked fine."
- "Which test, and where did you take it?" Multiple points beats one. In-situ beats surface for anything bonded.
- "If it comes back high, what happens to my price?" You want that answer before work starts, not as a change order with the carpet already in a skip.
A contractor who tests is telling you something about how they work generally. One who doesn't is quoting your house on the assumption that it's like the last one — and around here, it frequently isn't.
If you'd like the actual numbers for your slab, we test as part of the free in-home measure rather than as an add-on, and the reading goes on the written quote. More on how each material handles a Gulf Coast slab is in our flooring guide.
