2026-11-13

Why Your LVT Floor Failed (And It Wasn't the Plank's Fault)

A flooring contractor who's wasted $14,000 on misdiagnosed LVT failures explains the real reason most floating and glue-down vinyl jobs die within 18 months — and it almost always traces back to the adhesive, not the plank.

By Ji-Won Han

The call I got in March 2023

A property manager called me about a 4,200 sq ft LVT install we'd finished eleven months earlier. Curling edges. Peaking at the seams. A couple of planks in the break room had popped loose entirely. She assumed the flooring was defective. Her words: "These planks just aren't holding up."

I thought the same thing. I'd seen bad batches before. I requested the manufacturer's lot numbers, pulled the install records, drove over with a moisture meter and a pry bar. Two hours later I was standing in a half-demolished break room feeling like an idiot, because the plank wasn't the problem. It was never the plank.

That job cost me $4,800 in replacement material, another $2,100 in labor, and — worse than the money — the referral relationship with that property manager, who has not called me since. I've since documented four more LVT failures across three different projects, and the pattern is embarrassingly consistent.

What everyone thinks the problem is

When an LVT floor fails, the default assumptions go something like this:

  • "The plank quality is bad" — especially with imported product
  • "The subfloor was wet" — always a good guess
  • "Someone didn't let the adhesive tack up long enough" — a popular scapegoat
  • "The building's HVAC isn't controlled properly"

Some of these are real. Subfloor moisture will absolutely wreck a glue-down install. But if you actually tear apart failed jobs and test the residue, you find something more uncomfortable: the adhesive was the wrong type for that specific LVT construction, or it was the right type applied outside its window. Not defective. Not expired. Just wrong.

I've now pulled up enough failed floors to have an opinion about this, and the opinion is that the industry has been slow to update how we spec adhesives for modern LVT. What worked beautifully in 2018 does not reliably work in 2025, and most of the people specifying jobs haven't caught up.

The deeper reason — and it's not what installers think

Here's the thing that took me three failed jobs and roughly 14 grand in write-downs to actually internalize: LVT construction has changed dramatically over the past decade, and adhesive technology has changed to keep up. But the specifications people write — and the habits installers carry from job to job — haven't moved at the same speed.

The old logic was simple: LVT is LVT. Pick a pressure-sensitive adhesive, roll it, lay the plank, done. That logic breaks when the plank has a rigid SPC core, when it's a loose-lay product being glued on perimeter only, when the wear layer chemistry is different, or when the back of the tile has a fiberglass mat that behaves nothing like the bare PVC backing everyone learned on.

I'm not 100% sure of the exact chemistry involved in every failing case — I'm a contractor, not a polymer chemist. But I can tell you what I've observed across multiple jobs: the adhesive that failed was always selected by habit, not by matching the specific product's backing and the specific job site's conditions. Every single time.

Take Mapei LVT adhesives, since that's what we standardize on now. The line isn't one product — it's a family, and each one has a different tack profile, open time, and moisture tolerance. If you're still reaching for whatever bucket was on the truck because "it's the Mapei LVT one," you're making a bet that the last job and this job have the same conditions. They don't. I've watched a crew pull the wrong bucket off the shelf three times in one week because the labels were similar and nobody checked the spec sheet against the plank's backing.

That's not a product problem. That's a habit problem.

What the failures actually cost

Here's a rough accounting from my own documented jobs, from 2022 through 2024:

  • The March 2023 property manager job: ~$6,900 in direct redo costs, plus a lost referral relationship
  • A 2022 school corridor job (1,800 sq ft): $3,400 redo, plus a three-day schedule slip that pushed the next job
  • A 2024 retail build-out: we caught it before handoff — $600 in material, no labor hit, because we finally had a check in place

That last one is the only one that didn't hurt, and it didn't hurt because we finally built a pre-installation check that most crews skip. The check is not complicated. It's six lines long. But it has caught 47 potential mismatches in the past eighteen months, and I should have built it after the first failure instead of the fourth.

The other cost people don't talk about enough is credibility. When an LVT floor fails at month eleven, the client doesn't think "the installer used the wrong adhesive." The client thinks "this contractor's floors fail." You don't get to explain the difference. The floor is the entire argument.

What actually fixes it

The fix is boring, and that's sort of the point. Before any LVT job, match the adhesive to the specific plank construction — not to the brand, not to the price point, not to what worked last time. Mapei publishes compatibility guidance for every LVT line they make, and the guidance is different across lines for a reason. Read it against the actual plank spec sheet, not against your memory of a similar-looking job.

Then verify the site conditions against the adhesive's published window. Concrete moisture, ambient temperature, and substrate pH all matter, and they matter differently for different adhesive chemistries. A pH reading that's fine for one product is out of range for another. This is one area where I've become a pain in the neck with my own crews, and I've stopped apologizing for it.

One more thing — and I say this as someone who learned it the hard way: don't confuse LVT adhesives with other adhesive categories on the same job. Repair Care epoxy resin, for example, is the right product for wood substrate repairs and structural filling, and it's excellent at what it does. It is not an LVT adhesive, and the two should never get pulled off the shelf in the same run. We now physically separate adhesive storage by application because I've seen what happens when a tired installer grabs the wrong bucket at 6:40 in the morning.

If you only take one thing from this: the next time an LVT floor fails, before you blame the plank, pull the adhesive spec sheet and the plank backing spec, and hold them next to each other. More often than not, you'll find the failure right there — in the gap between the two documents, where nobody bothered to check.

I know because I didn't check, four times, and I'm still paying for it.

Need the supporting report or SDS package?

Ask the Ppg team for documentation tied to your region, application and selected coating family.

Download our 2025 report