I've Specified Silicone Sealants for 9 Years. These 4 Mistakes Taught Me What Quality Really Means

I've spent nine years specifying silicone sealants and elastomers for construction, industrial, and—unexpectedly—medical projects. In that time, I've made mistakes that totaled roughly $41,000 in wasted budget, not counting the damage to my own credibility. The worst part? Every single failure followed the same pattern: I prioritized price or convenience over material quality, and my company's reputation paid for it.

Here's my opinion, stated plainly: the material you spec isn't just a technical decision—it's a brand decision. The client's first experience of your work is how it looks and how it holds up. If a sealant fails, they don't blame the sealant. They blame you.

Lesson 1: "Structural" and "General Purpose" Are Not the Same Language

In 2019, I was project lead on a curtainwall installation. The spec called for Dow Corning 995 silicone structural sealant. I typed that into the purchase order, and my colleague—experienced, competent, but not a sealant specialist—saw "silicone sealant" and ordered Dow Corning 785 silicone sealant instead.

I'm not blaming him. On the surface, they're both grey tubes of silicone. Same manufacturer, similar texture, both with "sealant" in the name. If you're not living in this world daily, they look interchangeable.

They absolutely are not. Dow Corning 785 is a general-purpose acetoxy sealant: excellent for weathersealing, sanitary joints, and glazing that isn't load-bearing. Dow Corning 995 is a structural glazing sealant with high tensile strength, designed to transfer wind loads from glass panels to the building structure. It's tested to ASTM C920 and C1184, and it's the documented choice for structural silicone glazing.

We caught the error before installation—thankfully. But only because the glazing contractor's engineer noticed the product code during a pre-install inspection. If he hadn't? Structural glass panels held in place with a sealant never rated for structural loads. I get a little nauseous thinking about it, even now.

The mistake cost us $1,200 in reordering and a three-day schedule delay. But the liability we narrowly avoided? That would have been measured in millions, not thousands. Quality perception is one thing; structural safety is another. Both start with specifying the right product code.

Lesson 2: Cheap Thermoplastic Is the Most Expensive Mistake I've Made

Here's a misconception I see all the time: "thermoplastic is cheaper than silicone, so it's the budget-friendly option." That's a causation error. Lower upfront price doesn't equal lower total cost. In demanding applications, the opposite is generally true.

In 2021, I was managing an exterior joint-sealing project for a food-processing plant. The spec required a low-modulus silicone. But the subcontractor proposed a thermoplastic sealant instead, claiming it would save roughly $1.80 per linear meter. The price pressure was real, the jobsite was clean, and the joints looked simple. So I approved the substitution (not my finest judgment).

Within 18 months, the thermoplastic had hardened, cracked, and pulled away from the substrate on over 400 linear meters of joints. Silicone manufacturers—including Dow Corning—publish movement-capacity and weathering data precisely because the substrate and temperature cycling environment is unforgiving. Thermoplastics have their place: interior joints, low-movement applications, controlled environments. Outdoor exposure and thermal cycling? Not their place.

The rework: $6,200 for removal, surface prep, and reapplication, plus a weekend shutdown of a production line. We paid more than triple the original savings, and we did it in front of a client we'd have preferred to impress. The lesson is now embedded in our checklist: compare total service life cost, not unit price. Silicone costs more upfront for a reason, and a 20+ year service record is that reason.

When I talk to thermoplastic suppliers and silicone manufacturers side by side, they'll each tell you their material is right for the job. The difference is that silicone's performance claims are backed by decades of outdoor exposure tracking and project case histories. I've learned to ask both for the weathering data. It's a fast way to separate expectation from reality.

Lesson 3: Medical-Grade Silicone Is Not a Marketing Phrase

This is the one that embarrassed me (ugh). In 2022, we were approached about silicone components for body-contact medical and prosthetic devices—think protective cushioning, liner interfaces, and pads that sit against skin for extended periods. The purchasing contact asked if our standard industrial silicone elastomers would be suitable.

I said yes. Quickly. Without checking.

I was wrong. Industrial silicone elastomers are formulated with catalysts, additives, and processing aids that are not necessarily compatible with prolonged skin contact. Medical-grade silicone is a separate category, manufactured under tighter controls and tested for biocompatibility according to ISO 10993. You really don't want to discover this distinction after a product has reached clinical evaluation.

We caught it before production, but only because the client's QA engineer asked for the ISO 10993 documentation. My credibility took a hit that I still feel.

And yes, I'm going to address a search term that probably led some of you here: silicone buttocks. If you're looking for silicone products intended for body-contact use—prosthetic seating, orthotic supports, protective pads—the same rule applies. Use medical-grade silicone formulations, not industrial ones. Check the certificate, ask for biocompatibility data, and don't rely on the word "silicone" alone. The difference between "medical-grade" and "industrial-grade" is not semantics; it's patient safety and regulatory compliance.

Lesson 4: Do You Actually Know How to Clean PVC Pipe?

I almost didn't include this one because it's so simple. But overlooking simple things is exactly how failures happen. So here it is.

In 2021, we installed new PVC plumbing penetrations on a residential retrofit and sealed around them with Dow Corning 785 silicone sealant. Within four months, the sealant was peeling away cleanly. It looked like we'd applied it to the pipe while the pipe was still wet. The reality: PVC pipes often have a manufacturing release film on the surface. If you don't clean it off, your sealant isn't bonding to the pipe—it's bonding to a thin layer of film that will eventually release.

So how to clean PVC pipe before sealing, in four steps:

1. Wipe the pipe with isopropyl alcohol (or the solvent specified in the sealant's technical data sheet) to remove the release film and any grease.
2. Let it dry completely—don't rush this.
3. Apply primer if the sealant spec requires it. Yes, primer is an extra step. That's why it exists.
4. Apply the sealant within the manufacturer's open time window.

That mistake generated two return trips to the same site and replacement sealant on eight penetrations. Total: $450 in rework, on a job where the margin was under $2,000. I remember that math every time someone on the crew says "it's just a wipe."

I Can Hear the Procurement Objection Already

"This is all well and good, but we have a target price."

I've been on the receiving end of that spreadsheet more times than I can count. Budgets are real, and nobody is asking you to ignore them. But the lens is wrong. The right question isn't "what's the unit price?" It's "what's the price per year of reliable service?"

A $3.50 tube of general-purpose sealant in a structural application doesn't save $30—it risks $30,000 in failure costs. A $0.80-per-meter thermoplastic substitution doesn't save money when you pay for removal and reapplication 18 months later. Even the medical-grade premium—and it is a premium—is trivial compared to the cost of a biocompatibility failure.

I'm not suggesting you buy the most expensive option every time. I'm suggesting you verify that the material is actually suitable for its application, then look at what failure would cost, then make a decision. If you run that sequence honestly, quality usually wins. Not because it's the right thing to do (it is) but because the numbers point there anyway.

My Current Pre-Purchase Checklist

After nine years and $41,000 in documented mistakes, I've condensed everything to this checklist:

• Define the application before the product. Structural glazing, weathersealing, sanitary joints, and body-contact devices all have different requirements.
• Cross-check the product code. It sounds absurd, but read the full name: Dow Corning 995 structural sealant vs. Dow Corning 785 general-purpose sealant. The difference is exactly what the name says.
• Ask for the data. ASTM standards, ISO 10993 biocompatibility documentation, weathering test results. If the supplier can't provide them, question it.
• Include surface preparation in the plan. Any sealant, ever, regardless of price: the surface matters more than the material.
• Calculate cost per year of service, not cost per unit. The numbers will surprise you.

Quality perception is not a vague brand concept. It's the direct result of what your work looks like on day one, day hundred, and day thousand. The material you choose determines that outcome long before the client notices—and if it fails, they'll notice precisely because it wasn't the premium material you promised.

I still use thermoplastics, where they belong, and I still check prices on everything. But I don't underspecify anymore. My invoices have the scars to prove it's a cheaper lesson to learn from someone else's.