The Post-Construction Damage Problem: Why We Replace Construction Specialties Panels We Could Actually Repair
-
The surface problem: we default to 'replace' without ever asking whether we should
- The real reason: base-material damage and surface damage look identical during a walkthrough
- The actual cost: it's not just the swap, it's everything the swap drags with it
-
What actually works: three steps, not a new system
Last August I was walking a just-completed medical office corridor. Supposed to be the easy part of my day—post-construction walkthrough, ticking boxes against the punch list. By the third patient room, my map was dotted with 47 marks. Most were tiny. Surface scuffs on the wall panels. Paint chips the size of a thumbnail on two kick plates. A scratch on a frosted glass door leaf. A chip in a milk glass partition.
The contractor's PM looked at the first mark and shrugged. "Swap it out."
That's the default answer. It's also probably the wrong answer more often than not. But it's the default because when you're not sure, replacing is the version that definitely works—and repairing is the version that might fail in three months and become your problem under warranty. So I get it. I also think acting on that default costs projects more money than anyone has ever totaled up.
The surface problem: we default to 'replace' without ever asking whether we should
Here's the process most of us actually run. The building gets finished, the site walk happens, and someone spots a scuff, a chip, a paint defect. There's a quick triage—how many visual defects, is it at eye level, will the owner notice this immediately? If yes, replace. If no, log it and move on.
The part nobody has an incentive to do is figure out what the defect actually is. Because once you start asking "is this surface damage or base-material damage?" you have to answer it—and answering it wrong carries liability. Replacing is never wrong. Just expensive.
I'll put it this way: I don't have hard data on how many panel replacements per project are unnecessary, but based on four years of reviewing them, my sense is the ratio is somewhere between one in three and one in two. That's not a rounding error. That's line-item money.
The real reason: base-material damage and surface damage look identical during a walkthrough
This is the part that gets skipped. Most post-construction panel damage—especially on vinyl and acrylic-surfaced wall panels, coated metal, and painted gypsum—is entirely in the finish layer. A light scuff can be color-matched. A small paint chip can be touched up. A single damaged panel can often be repaired in place without pulling the whole run.
But there are also cases where repair is the wrong call and replacement is correct. Frosted glass with a scratch isn't a coating defect—you can't touch it up, you replace it. Milk glass is a fired finish, not a coating, so a repair kit isn't going to help you. Some cities now have glass-repair franchises that polish out light scratches on annealed or tempered glass, and they do genuinely useful work in the right application. That's not milk glass, though, and it's not every scratch. And if the damage reaches into a structural gap—small hole, corner knocked out—you're dealing with the substrate, not the finish, and a standard touch-up usually won't hold.
Here's the piece that actually matters: you cannot reliably tell surface damage from base-material damage by looking at the defect alone. You have to know what the panel is. And once you get that wrong, the number you quote is wrong by an order of magnitude in one direction or the other. At our shop, we used to default every scuff to replacement, and it was burning budget on every project.
One thing nobody wants to hear about color matching
If you're going to repair, you have to think about color. Industry standard color tolerance for brand-critical colors is Delta E < 2. Delta E of 2–4 is noticeable to trained observers; above 4 is visible to most people. Reference: Pantone Color Matching System guidelines.
Most field touch-up paint drifts in the 2–4 range. Meaning: most repairs are visibly imperfect to somebody who knows what to look for, and often to the building owner six months later. That's not because the product is bad. It's because a repair kit can't replicate a factory-applied finish under field conditions. So if a contract promises "color-matched repair," it needs to define what acceptable means. Ask what color deck, what paint system, and what Delta E is being held. Otherwise you're not buying a repair. You're buying a panel that no longer matches its neighbors.
A data gap I'll admit to
I wish I had tracked aging data more carefully from the start. I don't have solid numbers on milk glass replacement rates in this industry—it varies enormously by region, by sun exposure, by project type. What I can say anecdotally is that on our Construction Specialties jobs, most milk glass and frosted glass calls end up as replacement, not repair. That's dictated by the material, not by our preferences. And RSH-5700 panels, specifically, I can't give you a replacement rate on—we haven't run them long enough under enough conditions to be honest about it.
The actual cost: it's not just the swap, it's everything the swap drags with it
The direct delta between a panel replacement and a touch-up looks manageable on paper. The costs that hurt are the ones that don't sit on that line:
- Schedule drag. A repair happens alongside final punch work and is done in one visit. A replacement has to be scheduled, ordered, delivered, installed, inspected, and closed out—and post-construction schedules are the most expensive hours on the whole project.
- Adjacent work. Replacing a panel often means pulling base, moving a handrail, or re-doing a corner guard. Those become line items nobody planned for.
- Warranty exposure. This is the quiet one. If a repair fails and it shows up during the warranty period, the cost shifts from the contractor to the owner or facility manager. That's why "replace" feels cheap—it moves the risk. But it doesn't eliminate it. It prices it in.
- Hidden add-ons. Repair quotes are rarely transparent. The base number often doesn't include repair kits, doesn't include the repaint labor, doesn't include the color-match afternoon, doesn't include the return visit. I've learned to ask "what's not included?" before I ask "what's the price?"—because the vendor who lists every fee upfront, even when the total looks higher, usually costs less by closeout.
The mistake I made in year one
In my first year on this, I made the classic specification error: I assumed "matching repair" meant the same thing to every vendor. It didn't. We got a batch of 40 kick plates where the touch-up color drifted almost 4 Delta E from the original—visibly different to anyone walking by. The vendor claimed it was "within industry standard." It wasn't, per Pantone, but our contract didn't define the tolerance, so we redid it at our own cost on a project that was already 60% through. Now every single contract includes a written repair standard, and it names the color deck and the maximum Delta E.
What actually works: three steps, not a new system
I've been running this version for about two years. It doesn't require new software, doesn't require a new vendor, doesn't require much of anything except stopping the default long enough to make one decision per defect.
- Identify the substrate before you spec the fix. Is it coated metal, PVC-surfaced panel, painted gypsum, annealed or tempered glass, or milk glass? That single question separates repair-possible from replace-only. If you don't know, you don't have a plan yet—you have a guess.
- Sort repairs by material, not by defect. The same scrape on coated metal and on milk glass is two different line items at two different prices. Applying one default rule—say, "scuff equals replace"—guarantees you overspend on one and underspend on the other.
- Put the repair standard in the contract. Who pays, what tolerance is acceptable, what qualifies as replace-not-repair, and who signs off on the final call. These decided before the walkthrough are free. Decided after punch list submission, they're a change order.
Honestly, the thing that pushed me over wasn't the replacement cost itself—it was the two weeks I spent second-guessing a rush repair fee I'd already approved. Would the client have balked? Should I have negotiated? Didn't relax until a project manager on a totally different job said the number made sense to him. Now the decision runs on a checklist, and I sleep through the punch list.
So here's the number nobody wants to quote: not how many panels got replaced on the job. How many of them needed to be.
Leave a Reply
Your email address will not be published. Required fields are marked *