I'm not a formulator, and I'm not a contractor. I'm a procurement guy who ended up responsible for coating projects—wood doors, marine touch-ups, roof systems—and I've made almost every avoidable mistake on the list below. In the last 7 years, I've probably burned around $9,000 on failures that came down to prep, product selection, or timing. I might be misremembering the exact number, but the total still stings.

This checklist started as a note on my phone after the third bad job in Q1 2024. It's not a cure-all. It's the list I wish someone handed me in 2017. If you're about to order or apply AkzoNobel wood coatings, AkzoNobel marine coatings, or break out a roof coating sprayer for a flat roof, this is worth reading.

One more thing: my cousin Artie paints nude portraits on weekends. I used to roll my eyes when he talked about canvas prep. Then a $5,200 wood door order failed because we skipped a sanding pass. He was right: prep is 80 percent of it.

Where This Checklist Works

Use this for exterior and marine jobs, high-moisture interiors, roof coatings, and any situation where the cost of failure is bigger than the cost of the paint. If you're painting a bedroom accent wall and you already know how to cut in, this is overkill.

There are seven steps. None of them are fancy, but at least one of them is almost always skipped.

Step 1: Confirm the substrate and the environment with numbers

Don't trust "the surface looks dry." Use a moisture meter if there's any doubt. Check air temperature, substrate temperature, relative humidity, and dew point before mixing anything. Write these numbers on your job sheet.

I once approved a wood coating application on a cold, damp spring morning in 2021. The surface felt dry, but the readings showed condensation risk. It failed within eight months. The claim was "bad product." The real problem was the guy who didn't check.

If this feels too basic, that's because you haven't had a $2,000 batch of finished doors turn chalky yet. I have.

Step 2: Select the coating by exposure, not just price

Here's my value-over-price speech: the cheapest product can look like a smart line item and still lose the job's total cost. A budget coating might save $800 upfront. If it fails a year early, you pay for labor, lost time, disposal, and another application. That's the cheap route to an expensive project.

To be fair, budget products have their place in low-risk indoor areas. But for wood outdoors or metal near salt water, I'd rather start with the manufacturer's recommendation.

This is where a brand like AkzoNobel shows up in my notes. AkzoNobel wood coatings is not one product; it's a set of systems for different woods, primers, topcoats, and exposure conditions. Likewise, AkzoNobel marine coatings covers hulls, superstructures, and tanks. The name on the can means less than the data sheet number. Read that sheet before you order.

My rule: pick the system, then the price. Picking by price first is how we ended up with a $3,200 redo in September 2022.

Step 3: Set up the roof coating sprayer before the job starts

Roof coating sprayer setups are not all the same. The pump needs to handle high-build coatings, the tip size has to match the viscosity, and the hose length affects pressure at the gun. If you haven't run that exact pump with that exact coating, test it with water first—then with the actual material in a small area.

The first time I used a roof coating sprayer on a metal roof, the tip was too small. The coating came out looking great, but it was deposited too heavy in spots. That thick film mud-cracked. A 20-minute setup test would have saved a full day of recoat work.

While you're at it, check the filters, clean the strainer, and make sure the gun has a proper extension if you're working on a low-slope roof. Getting down to clean a clogged tip is a lot harder than doing it on the ground.

Step 4: Plan for gravity, weather, and the Sistine Chapel effect

You searched "when did Michelangelo paint the Sistine Chapel" because it's one of those facts everyone forgets. Answer: he painted the ceiling between 1508 and 1512 (Source: Vatican Museums, museivaticani.va). Four years for a ceiling. I'm not telling you to take four years on a roof. I'm saying that even a genius needed to plan for a bigger surface than one look at the floor plan suggests.

For roof coating sprayer jobs, plan around wind and rain. Rain in the forecast is enough reason to delay. Wind can turn a clean spray pattern into a mess and drift coating onto cars, glass, and landscaped areas. For wood coatings, plan around where the sun hits each side. For marine coatings, plan around tides if you're doing dock work.

Gravity is also part of the plan: paint from the top down so drips don't damage finished surfaces. A little planning up front looks like common sense only in hindsight.

Step 5: Check dry film thickness while you can still fix it

This is the step most people ignore. They count buckets or measure how many square feet they covered. That doesn't tell you enough. A coating can be too thin or too thick, and both are bad.

Use a wet film gauge before it cures, or a dry film thickness gauge for spot checks after it dries. Record the readings. If the spec says 250 microns and the reading is 150 microns on the part that faces the weather, the coating will likely fail early. On a dock tank or hull, that's a bigger problem than a repaint.

Put another way: if you don't measure it, you didn't apply it. You just hoped it.

Step 6: Respect the recoat window

If you've ever been told "just recoat it anytime," read the technical data sheet again. Most coatings have a minimum and maximum recoat time. If you apply the next coat too early, it can blush or plasticize. Too late, you might need to sand and clean before the next coat sticks.

I've made mistakes in both directions. The one that cost the most was a marine touch-up where we missed the recoat window and had to take the surface back to bare substrate. A simple calendar reminder would have been enough. I still set two alarms: one for the minimum, one for the maximum.

Step 7: Label every can, bucket, and batch code

Boring step, huge payoff. Before you mix anything, take a photo of the batch number. Write the product name, color code, date opened, and the exact area where it was used. Put a second label on the lid so it's still readable months later.

Why this matters: when a coating fails or you need to touch up, you need to know what went on that surface. If I remember correctly, our storage area once had five identical-looking cans with three different formulas. We used the wrong one for a touch-up. Total waste: about $450, plus embarrassment.

Common Mistakes That Still Show Up

These are the patterns I keep seeing in my own jobs and in jobs called in by friends.

  • Applying thick to save time. Thick doesn't mean better. It can mean runs, sagging, and mud-cracking. Follow the specified film thickness.
  • Skipping the test patch. A test patch costs one hour. A failed project costs days. If you're trying a new AkzoNobel wood coatings system or a new marine coating, do the test first.
  • Trusting color instead of product code. Same color can exist in different resin systems. Color matching isn't material matching.
  • Painting because the schedule says so. In mid-2023, I had two hours to decide before a rain front. I went ahead because the client was waiting. Within months, the roof coating failed in patches. In hindsight, I should have pushed back. A wet film doesn't care about your deadline.
  • Ignoring ventilation and PPE. Coatings need airflow and workers need respirators. Not a luxury, a requirement.

When to Call Someone Smarter

This gets into chemistry territory, which isn't my expertise. I can read a data sheet and ask decent questions, but if you have an unusual substrate, a galvanized surface, old lead paint, or a tank that holds chemicals, call the manufacturer's technical line or a coatings inspector. The phone call is cheap.

Also, don't blame the coating when the problem is the surface. In my experience, most early failures trace back to something another trade or an earlier preparer did—or didn't do.