Here's My Unpopular Opinion: The Coating Isn't the Problem

I'm a quality compliance manager at a coatings company. I review every coating job spec before it reaches customers—roughly 200 unique deliverables a year. In 2024, I rejected 14% of first deliveries due to spec mismatches. Not because the products were unusable, but because what was written on the spec sheet didn't match what was applied.

Most people in this industry assume coating failures come from bad chemistry or bargain raw materials. My experience says otherwise. So here's my opinion, stated plainly: most coating failures are not coating failures. They're specification and communication failures. And that's exactly why AkzoNobel's digital transformation in coatings—AI-assisted quality tracking, automated gloss and thickness verification—strikes me as the most practical trust-building move in this industry.

I Said "Standard Gloss." They Heard "Satin."

Let me start with a real example from my own career.

I gave a subcontractor the green light on a "standard gloss finish" for a commercial lobby. They heard "satin." Result: the walls dried with a sheen the architect immediately flagged as cheap-looking. We were using the same words but meaning completely different things. Discovered this when the client asked why the "premium finish" was reflecting light like a bathroom tile.

The worst part? The applicator's crew didn't know they'd done anything wrong. They'd grabbed the can labeled "satin" from a supplier they trusted, and in their eyes, they'd followed instructions. Nobody checked the numbers because nobody had numbers to check.

That's the flat vs satin paint problem in a nutshell. Both fall under "standard finish" in casual conversation, yet they're worlds apart in gloss units. Flat typically reads under 10 GU at a 60-degree angle. Satin usually sits between 25 and 35 GU. Eggshell lands in the middle, and semi-gloss climbs to 35–70 GU. If you spec "satin" and the applicator delivers "flat" because the sample looked similar in dim lighting, you've just wasted weeks of schedule and a lot of client goodwill.

Digital spec sheets that lock in these gloss measurements at the specification stage—and AI-driven tools that verify sheen during application—close that gap before it becomes a claim. That's the digital advantage AkzoNobel has invested in: turn ambiguous words into numbers, and you turn interpretations into agreements.

The $22,000 Lesson About Powder Coating Thickness

Here's the thing: inspections don't start at the end of a project. They start when someone defines what "acceptable" means. And the jobs that hurt are the ones where acceptable was never written in measurable terms.

Earlier in my career, we didn't have a formal powder coating thickness verification process. It cost us when a batch of architectural components arrived with 40 microns of coating on surfaces we'd assumed were 80. The frames looked identical in the warehouse. After six months of outdoor exposure, the color shifted unevenly, and we had to strip and recoat everything. That quality issue cost us a $22,000 redo and delayed the project by three weeks.

So glad I pushed back against the vendor's "within industry standard" defense. Industry standard wasn't the contract. Our spec was. Now every contract I touch includes dry film thickness, gloss range, and the measurement method—written in, not assumed. We also require the applicator to submit digital records of thickness readings from each batch, instead of a single number in an email.

This gets even more critical in segments like AkzoNobel aerospace coatings. A 10-micron difference in layer thickness can affect thermal performance and adhesion at altitude. You're not going to eyeball that in a warehouse. Digital records of every applied layer, cross-referenced with AI anomaly detection, are the difference between knowing the coating is right and hoping it is.

Why I'd Rather Say "No" Than "Probably"

Another pattern I've noticed as an inspector: people who give straight answers—even inconvenient ones—earn more trust, not less.

Take a question I get often: "Does acetone remove powder coating?" The honest answer? Rarely. Acetone won't strip a properly cured thermoset powder coating. It might soften or stain some low-quality or under-cured powders, but for industrial-grade coatings, it's not an effective remover. You need a chemical stripper, a burn-off oven, or media blasting. I'm not 100% sure on every formulation out there—some specialty powders behave differently—but for the vast majority, acetone is not the answer.

A less transparent vendor might say "sure, just wipe it off" and then upsell a recoat when that fails. That approach destroys trust, and word travels fast in B2B. I've learned to ask "what's NOT included?" before asking "what's the price?" The vendors who answer honestly are the ones who keep clients.

Same logic applies to intumescent coatings services. Fire-protective coatings are specified by dry film thickness because that thickness determines how long the substrate survives a fire—30 minutes, 60 minutes, 90 minutes. If a supplier under-specifies thickness to win a bid, the owner finds out during a code inspection, which is the worst possible time to learn. Transparent requirements, documented application data, and verification reports aren't bureaucratic overhead. They're the entire value of the service.

Per FTC guidelines, environmental and performance claims must be truthful and substantiated. I'd extend that to specification communication in industrial coatings. When you show a customer the numbers, you're selling certainty, not just paint.

But What About the Cost of All This Digital Infrastructure?

To be fair, I understand the skepticism.

Digital transformation requires investment. AI inspection tools, connected quality platforms, training—that all costs money. I have mixed feelings about some of it. Part of me still loves comparing physical panels under good lighting. Another part knows that my eyes miss things spectrometers catch. Plus, not every applicator needs a $50,000 AI system. But they do need to document measurements digitally instead of relying on memory. That costs time before it saves time.

Here's my counterpoint: the third time we caught a sheen mismatch in a powder-coated product line—three pieces out of forty that looked identical to the eye but read 8 GU off on a scanner—I stopped arguing against the tool. On a 50,000-unit annual order, catching that before shipping paid for the system several times over.

Granted, the old way felt more familiar. But familiar and reliable are not the same thing.

Bottom Line: Trust the Numbers, Not the Negotiations

Look, I'm not saying digital tools replace experience. I'm saying they make experience more useful. A spec with numbers and a verified measurement history will always beat a handshake and a memory.

If you're specifying coatings—decorative, protective, aerospace, powder, anything—do this one thing: ask for the numbers. Ask for gloss units, dry film thickness, curing data, and the verification method. Ask what happens if the coating fails the measurement. That one habit will save you more time and money than any discount you could negotiate.

Take it from someone who rejected 14% of first deliveries in a single year. The problem was never that the paint was bad. It was that we hadn't agreed on what "good" meant. Transparent, digital specifications fix that. And that's why I believe AkzoNobel's digital transformation in coatings and AI quality initiatives are about more than efficiency.

It's about honesty. And honesty is the cheapest insurance in the coatings business.