Why Do Foundations Fail in North Texas? 2026 Soil Guide

Why do foundations fail in North Texas? Ask around, and nearly everyone gives you the same two words: expansive clay.

That's true, but on its own it doesn't help you much. It doesn't tell you which clay is under your house, or how much it really moves. It doesn't explain why your neighbor two suburbs over has never had a problem, while your doors stopped closing in August.

So we went and found the numbers. The USDA keeps a national soil survey with the measured shrink and swell behavior of every mapped soil in the country. We sampled it on a 25-point grid across eleven cities in and around Garland. What came back is a lot more useful than "expansive clay," and we think it'll help you make sense of your own house.

Soil scientists measure swelling with linear extensibility. It's the percent a soil sample changes size between wet and dry. The bands are simple:

• Under 3 is low
• 3 to 6 is moderate
• 6 to 9 is high
• Over 9 is very high

Houston Black, the soil under much of Garland, reads 17. It runs 55% clay.

That's a lot more than "a bit expansive." It's close to the top of the national range. A soil at 17 can lift a corner of a slab after a wet spring and drop it again through a dry August, year after year, for as long as the house stands.

Two things follow from that, and they'll help you more than anything else on this page.

First, the movement is cyclical, not one-off. Houses here don't settle once and stop. They ride up and down with the seasons.

Second, the damage comes from differential movement. If the whole slab rose evenly, almost nothing would crack. Damage happens when one part of the soil is wet and swollen while another part is dry and shrunken, so the slab has to bend.

Here's where the numbers start to get useful. This is the main soil in each city we sampled, with its reactivity:

• Plano: Houston Black, 64% of the city, reads 17
• Garland: Houston Black, 32%, reads 17, with Ferris another 16% also at 17
• Mesquite: Houston Black, 32%, reads 17
• Rowlett: Houston Black, 36%, reads 17
• Sunnyvale: Houston Black and Trinity, 16% each, reading 12 and 13
• Rockwall: Houston Black, 36%, reads 12
• Murphy: Houston Black, 60%, reads 12
• Wylie: Heiden, 32%, reads 12.2
• Sachse: Houston Black, 28%, reads 12
• Richardson: Austin, 44%, reads 7.5
• Balch Springs: Rader, 24%, reads 1.5 at the surface and 7.5 below it

Look at the two ends of that list. Plano is 64% covered by soil reading 17. Richardson, its neighbor to the south, is 44% covered by soil reading 7.5.

That's more than double the reactivity, across a city limit line.

It explains something that puzzles a lot of homeowners. Two houses of the same age, built by the same builder ten miles apart, can have completely different histories. They simply aren't on the same ground.

Four soil names account for most of the trouble in this area.

A fair picture includes the places where the problem is smaller, and there are some.

Austin is a chalk-derived soil that reads 7.5 at 45% clay. It covers 44% of Richardson, more than any other soil there. That's still high, but it's less than half the movement of Houston Black for the same swing in moisture.

Eddy reads 1.5, which is genuinely low. It runs about 30% clay and drains well. It covers 16% of Plano, 12% of Sachse and 12% of Murphy.

That last one is worth a moment. Plano is the most clay-heavy city we sampled, and it still has a sixth of its area on soil that barely moves at all.

If your house sits on Eddy and your floors are going out of level, shrink and swell probably isn't the cause. It's worth looking for a plumbing leak, poor grading, or roof water landing in one spot instead.

So if someone tells you every house in a city needs piers, the soil survey doesn't back them up.

We'll be upfront about this one, because the lesson is useful to you.

Our first run at this data read only the top layer, the soil at zero centimeters. It reported Balch Springs at 1.5, and we nearly published that the city had low shrink-swell risk.

That was wrong. Balch Springs is sandy or loamy on top. Underneath, Axtell reaches 45% clay at 7.5 from as shallow as four inches. Rader hits 42.5% clay at 7.5 by about 33 inches. Wilson runs 42.5% clay at 7.5 from under a foot down.

A slab doesn't bear on the topsoil. It bears on or near the subsoil, so reading the surface and stopping gives you exactly the wrong answer.

Our tool now records the surface figure and the peak figure separately, with the depth and clay percentage for each, and we quote the peak. When we re-ran all eleven cities, Balch Springs was the only one affected. Everywhere else, the main soil already reads at or near its peak right at the surface.

There's a second lesson in there too. Balch Springs also has Houston Black and Ferris, both at 17, on another 20% of the city. So "Balch Springs has sandy soil" is just as wrong as "Balch Springs has black clay." It has both, and the answer depends on your address.

It's the same soil with three different ways to fail, depending on when your house went up.

Census figures put the median build year in Garland at 1978, with 11.4% of homes predating 1960 and 63.3% built between 1960 and 1989. Murphy's median is 2005, with 90.8% of its housing built in 1990 or later. Mesquite has the biggest share of older homes we looked at, with 12.7% predating 1960.

Those are build years, not foundation types. The Census doesn't record what your house sits on. What follows is the industry's general pattern, not a statement about your home:

• Pre-1960 tends to be pier and beam, with a crawl space, wood beams and interior piers. These move with the soil and can often be adjusted.
• 1960s through 1980s tends to be a conventional slab reinforced with rebar. That's most of Garland. A rebar slab is stiff, so it tends to crack rather than flex.
• 1990s onward is more and more often post-tension, where steel cables are tensioned inside the concrete. The one firm date here comes from the Post-Tensioning Institute, which has published approved methods since 1976.

The practical difference is what happens next. On a post-tension slab, nobody should core or drill without scanning first, because a live cable is dangerous. On pier and beam, the fix is often an adjustment rather than underpinning.

Reactive soil sets the stage. Something still has to change the moisture to set it off, and in practice it's nearly always one of three things.

Notice what these three have in common. None of them is about how much rain falls. All three are about rain, or water, arriving unevenly around one building.

Here's one more sign that local ground really does vary: the cities themselves don't agree on how to regulate the repair.

• Garland requires a permit once the repair is worth $2,000 or more, and charges $45
• Richardson triggers on a pier count instead: four or more piers
• Rowlett and Mesquite require a permit for any foundation repair, with no threshold at all
• Plano runs a full plan review, with two copies of sealed construction documents
• Rockwall writes it into the ordinance: every foundation or footing, or any size addition to an existing foundation, must be sealed by a Texas registered engineer

That's five different approaches inside a ten-mile radius. Building officials who deal with this every day have reached five different conclusions about how closely it needs watching.

The short answer to why do foundations fail in North Texas is that the soil changes size with the weather, and here it changes a lot. The useful answer needs two more facts: which soil is under your address, and when your house was built.

The good news is that both are easy to find. The soil survey is public, and we've sampled it for eleven cities on this page. Your build year is on the county appraisal record.

A good assessment will look at both. "Expansive clay soil" is where a real answer starts, not where it ends. If someone can't tell you whether you're on Houston Black at 17 or Austin at 7.5, it's fair to ask them to take a closer look.