Silicone Looks Expensive. My 6-Year Procurement Log Says Otherwise.

The Conversation I Keep Having

Every quarter we review material costs. And every quarter, someone at the table points at a line item for Dow Corning silicone and asks the same question: "Can't we get away with EPDM here? Or neoprene? They're a fraction of the price."

I understand the instinct. The price book comparison is brutal for silicone. On paper, the material cost difference is undeniable: EPDM compounds generally run $3–6 per kilogram; silicone runs $8–15 per kilogram, based on published commodity indices I checked in January 2025. That's roughly a 2–3x multiplier. In one 2024 quote comparison I ran for a custom seal, silicone was 180% more expensive than the incumbent EPDM on a per-unit basis.

But per-unit price is the wrong number to look at. I've been managing procurement at a 120-person manufacturing company for six years. Our annual materials budget is about $1.2M, and I've tracked every order in our cost tracking system. After six years of data, I can tell you that unit price predicts maybe 10–15% of the real cost of a material over a component's lifecycle. The other 85–90% lives in failures, replacement labor, downtime, warranty claims, and the quieter damage: what happens to your reputation when your product fails in front of your customer.

That last part took me the longest to learn.

The March 2023 Lesson

In March 2023, a customer who builds rubber wheel loaders called about a batch of failed hydraulic seals. Actually, let me back up. A year before that call, the customer's own procurement team had asked us to switch to a cheaper rubber seal for their hydraulic cylinders. It was a generic EPDM seal that looked comparable to the Dow Corning silicone we'd been supplying. The savings: $1.10 per unit. On a 1,200-unit annual order, that's $1,320. They were happy with the saving.

"We've got three units down with hydraulic leaks. We need to understand what changed, and we need it fixed fast."

Each repair took four hours of shop time. At $240/hour shop labor, that's $960 per unit just in labor—before parts, before logistics, before the machines sitting idle on jobsites. The total direct cost of that switch: roughly $4,800 in rework on a $1,320 savings.

That event changed how I think about material substitution. I didn't fully understand the value of silicone's reliability until a $1.10 saving turned into a $4,800 field failure—and then into a lost account. They quietly moved that product line to another supplier the following year.

The Chemistry Behind the Cost Difference

Here's what I learned when I finally dug into the actual science. Silicone has a silicon-oxygen backbone. Most organic rubbers—EPDM, NBR, neoprene—have carbon backbones. That silicon-oxygen bond gives silicone a much wider service temperature range: roughly -60°C to 230°C for some grades (i.e., far beyond where most organic rubbers can operate). It also resists UV, ozone, and many chemicals that degrade carbon-backbone rubbers.

Honestly, I'm not sure why some engineering teams don't factor this in when they approve substitutions. My best guess is they see "rubber" in both columns of the spec sheet and assume the difference is mostly cosmetic. But in six years of tracking field failures, the most common patterns I've seen with organic rubber seals are:

• Compression set—the seal flattens and stops sealing
• Hardening and cracking after UV or heat exposure
• Chemical attack from oils, solvents, or cleaning agents
• Ozone cracking, especially in powered equipment

Silicone handles these conditions measurably better. That's not a marketing claim—it's thermal stability and oxidation resistance that follow directly from the polymer structure. When I see a seal fail in an automotive or heavy-equipment application, the failed material is almost never silicone.

The Automotive Example That Killed the "Cheaper Is Fine" Argument

There was another one in Q2 2024 that turned a dependable product into a warranty headache. This involved a Dow Corning silicone automotive component—a wiper seal we supplied to a tier-2 automotive supplier.

They were paying $2.85 per unit. Their cost-reduction team found a crosslinked rubber version at $2.10, a $0.75 saving per unit. They switched. Eight months later, their quality department flagged a 6.8% field-return rate on that assembly: the seals were hardening after extended engine-heat exposure, and the vehicles were past the 50,000-mile mark.

When we ran the numbers together, the $0.75 saving per unit turned into a $17.50 cost per returned unit, once you included warranty replacement, freight, and administrative handling. On their annual volume of about 40,000 units, the original savings was $30,000. The warranty exposure was over $47,000. That doesn't include the reputational cost of a part that fails well before the customer expects it to.

I should add that this was a well-established tier-2 supplier with a mature quality system. This wasn't negligence. It was the same mistake I'd made earlier in my career: looking at the price book instead of the total cost of failure.

"Can Foam Board Insulation Get Wet?"

Let me take a detour into a completely different industry, because the same pattern shows up in construction.

Somebody in our extended network once asked, "Can foam board insulation get wet?" It's a real question. The short answer is no—standard expanded polystyrene (EPS) absorbs moisture, and even modest water uptake measurably reduces its R-value. Some tests show a 20–30% loss in thermal performance when the board is wet. Trapped moisture can also create conditions for mold growth.

But here's the connection to my job: I've seen builders spec foam board insulation in exterior walls and treat it as if it were a waterproofing layer. It isn't. The gaps, penetrations, and joints—where water actually gets in—need a weather-resistant sealant. In commercial construction, Dow Corning 3145 RTV silicone adhesive is a standard choice for those joints: it bonds to the foam without melting it (solvent-based sealants can eat EPS), and it stays flexible enough to handle thermal movement.

The lesson is identical to the wheel loader story. Foam board is excellent at what it's designed to do—insulate. But when you use it as a stand-in for something it isn't (a waterproof barrier), the cost shows up later. In this case, it shows up in energy bills, mold remediation, and callbacks.

Rubber Stamping: The Tolerance Problem Nobody Expects

One more example, because this one surprised me. In rubber stamping—die-cutting rubber sheet into gaskets and pads—we had a run where a customer requested a "generic equivalent" of our Dow Corning silicone sheet. The generic material was 40% cheaper per sheet.

The first batch stamped fine. Tolerances looked good. But by the second production run, the punched holes were off by half a millimeter and the surface finish was inconsistent. We traced it back to the material: the cheaper rubber had wider gauge tolerance across the sheet, so the stamping press produced out-of-spec parts when the thickness varied.

The visible rework cost was $1,200. The material savings on that batch was about $380. But the real cost was the two-week delay to our customer's production schedule. I don't have a spreadsheet column for "customer frustration," but I know what happened next: the customer's next O-ring kit order went to a competitor.

That's the quality-perception angle that doesn't show up in any TCO calculator. When a product fails or a delivery misses schedule, the customer remembers you. Not the material spec. Not the supplier who sold you the cheaper rubber. They remember your company name.

What Changed in Our Procurement Policy

After these examples, I restructured how we evaluate materials. We now require quotes from at least three vendors for any new material, but we compare total cost of ownership (meaning: unit price plus failure risk plus replacement labor plus warranty exposure), not just unit price. I built a cost calculator after getting burned on hidden field failures twice—you can't see the problem if you're only looking at the invoice.

The "silicone premium" actually looks small when you model it properly. For a $2.85 silicone seal vs. a $2.10 rubber alternative, our model shows silicone breaking even at about 800 units if the failure-rate difference is just 0.5%. The break-even point comes even earlier in applications with high downtime costs—like wheel loaders, automotive modules, or building envelopes.

I want to be clear about something: I'm not anti-rubber. EPDM, neoprene, and polyurethane are the right materials for many applications. We still use organic rubber for plenty of our own components. The point is to make the decision on the right number. If you're choosing materials based on per-unit price alone, you're not comparing the actual cost of what you're building. You're just hoping the failure rate doesn't find you.

Bottom Line

Six years of procurement data taught me this: the most expensive decision I see repeated isn't the premium material. It's the substitution made on price alone, without accounting for what happens when it fails in the field. The $1.10 seal. The $0.75 automotive part. The 40%-cheaper rubber sheet. These feel good in the quarterly cost review—and they hurt in the warranty report.

You don't have to take my word for it. Pull your own field failure data. Build your own TCO model. If you're honest about the consequences of a field failure for your customers, you'll probably find the same thing I did: quality silicone is rarely the most expensive option once you count everything.

Paying a few cents more per part is a reasonable trade. The alternative is a phone call from a customer whose machine is down—and that's a cost you can't negotiate down.