-
The comparison framework — and why five buckets
-
Dimension 1: Upfront quote — where unbundled almost always wins
-
Dimension 2: Design handoff — the quiet killer of unbundled projects
-
Dimension 3: Installation coordination — where the real comparison lives
-
Dimension 4: Quality and warranty gaps
-
Dimension 5: Timeline — and the counterintuitive finding
-
Where I've had to make the call under pressure
-
So which model should you choose?
I manage a $2.4M annual capital equipment budget for a 180-person specialty chemical plant. Over the past six years I've tracked invoices from 34 chemical process equipment purchases, built TCO models on 11 capital projects above $100k, and — not proudly — signed off on at least two sourcing decisions I'd reverse if I could.
When I started this role in 2019, I assumed the lowest quote was the best deal. It took a $47,000 surprise invoice (installation subcontractor, billed separately, buried on page 4 of a PDF) to teach me that a low-cost chemical process equipment supplier and an expensive one can land within 5% of each other on total cost of ownership — just not on the purchase order.
This article compares the two sourcing models I've evaluated most: unbundled sourcing (separate design firm, separate chemical process equipment manufacturer, separate installation contractor) versus integrated sourcing (one chemical process equipment supplier that holds or coordinates design, fabrication, and installation). I've organized the comparison around five cost buckets, because that's how our tracking system categorizes them.
The comparison framework — and why five buckets
Every capital project over $50k in our system gets tagged into five TCO buckets: (1) upfront quote, (2) design handoff cost, (3) installation coordination, (4) quality and warranty gaps, and (5) timeline impact. I'm using the same five here so you can replicate the math on your own numbers.
Quick scope note: when I say "chemical process equipment" I mean reactors, heat exchangers, pressure vessels, distillation columns, and similar unit operations. If you're sourcing pumps or instruments under $10k, the comparison overhead below is heavier than the savings — don't replicate this process for a $6,000 order. (I tried once. It took three weeks and didn't move the needle.)
Dimension 1: Upfront quote — where unbundled almost always wins
Least surprising comparison, so I'll keep it short: on sticker price, unbundled wins. In 2023 I compared a 3,000-gallon jacketed reactor across eight vendors. Two chemical process equipment manufacturers quoted the vessel only: $118,000 and $127,000. An integrated supplier quoted $172,000 for the same spec — their number bundled design review, skid assembly, and installation supervision.
That's a 35–46% spread on the purchase-order line. On sticker price alone, unbundled is a clear win, and I'd be lying if I said that didn't tempt me (it did, more than once).
The catch is that the $118,000 vessel doesn't arrive installed. The moment you add design, transport, rigging, piping, and commissioning, the gap narrows fast, and sometimes it inverts. Which brings us to dimension two.
Dimension 2: Design handoff — the quiet killer of unbundled projects
Here's where my initial assumption broke. When you hire a separate design firm and a separate chemical process equipment manufacturer, the interface between them is billable. Both ways.
On a 2022 distillation column project, we hired an engineering consultant for design ($38,000) and sent drawings to a manufacturer ($142,000). The manufacturer's engineer flagged 11 "clarifications" that required design revisions — billed at $180/hour by the consultant. Total design handoff cost: $61,000 against a $38,000 estimate. That's a 60% overrun on a line I'd mentally locked.
I said "clarifications." The manufacturer called them "errors." The design firm called them "constructive changes." Everyone was using polite language for the same thing (unfortunately, this is still how it works in 2026).
The most expensive mismatch that year: I said "standard connection." The manufacturer heard "ASME B16.5 Class 150" — the standard was current then and I'd assumed no gap, which was a mistake on my side. What we actually had on the header was a different facing. Result: two flanges on site that didn't fit, $4,200 in rework, and 9 days of schedule. Not catastrophic on its own. But it compounds across 15–20 interfaces on a mid-size project.
An integrated chemical process equipment supplier absorbs most of this cost internally — their engineers and their fabrication team talk to each other without invoices. You pay a premium on the quote, but you're arguably buying coordination.
Comparison result: Unbundled wins on quote, loses on handoff. On our 2022 and 2023 projects, the handoff overrun ate 55–70% of the apparent 30% quote gap.
Dimension 3: Installation coordination — where the real comparison lives
Most procurement teams stop measuring at the PO. That was my mistake for about two years. Installation is where the differentiator shows up, and where the honest limitation of unbundled sourcing becomes obvious.
Two 2024 projects, same facility, roughly comparable scope:
- The 2024 skid-mount project (unbundled): Equipment from a lower-cost chemical process equipment supplier, installation by a contractor we'd used for years. Installation came in at $88,000 against a $70,000 estimate. The contractor's crew spent 40 hours reading manufacturer drawings that didn't match our plant's tie-in points — hours we paid for.
- The 2024 stripper column project (integrated): Equipment plus installation from one supplier at a bundled $210,000. Installation wasn't broken out, which annoyed my accounting team, but the total stayed within 2% of the quote. No interface charges. No 40-hour drawing-reading sessions.
On a per-unit basis the unbundled project was about 12% cheaper on the sheet. But it ran 23 days late against a 90-day install window, and the delay pushed our production restart by three weeks. The contribution margin on three lost weeks was roughly $26,000 — more than the 12% premium.
Here's the counterintuitive part: on the sticker, integrated cost more. On delivered TCO, integrated won. That isn't always true — but it was on both 2024 projects.
Dimension 4: Quality and warranty gaps
Unbundled sourcing splits the warranty into pieces. On the 2022 column project, design was warranted by the consultant, fabrication by the manufacturer, installation by the contractor. When a leak appeared at a nozzle weld in month 14, we spent six weeks in three-way finger-pointing before anyone dispatched a crew. The welder who failed the X-ray worked for contractor A. The weld spec came from consultant B. The nozzle landed on manufacturer C's drawings.
Nobody was technically wrong. But the coordination cost of resolving an unbundled warranty claim is real and it's never in the original quote. A 2023 trade-press estimate put installed pressure-vessel cost at roughly 2.5–3.5× equipment cost — treat that as a planning band, not a quote. Most of that delta is coordination, not steel.
Integrated sourcing doesn't eliminate quality problems — it consolidates who you call when they happen. For a plant running a 340-day annual operating schedule, "who I call" matters almost as much as "what broke."
Dimension 5: Timeline — and the counterintuitive finding
Conventional wisdom says unbundled is faster because you aren't waiting on one supplier to do three things in sequence. In our data, the opposite was more often true.
When I audited our 2023 and 2024 schedules (8 projects, $1.9M total), unbundled projects averaged 118 days from PO to production. Integrated projects averaged 104 days. Not dramatic — but consistent.
The reason isn't speed in any one phase. It's parallelization. An integrated supplier starts fabrication while installation planning is underway, because the same PM owns both. With unbundled sourcing, we were the integration layer, and our team wasn't always good at it. That's not a knock on procurement people — we're buyers, not project engineers, and the gap shows up on the schedule.
There is a limit. If the integrated supplier is overloaded, their internal queue can stretch past what unbundled would deliver. I've seen that happen once in six years. It's a real risk, just a rarer one than the interface problem.
Where I've had to make the call under pressure
One time worth mentioning because it's the kind of decision that gets romanticized afterward but wasn't clean at the time. In Q1 2023 a heat exchanger in our second train failed inspection four days before a scheduled shutdown. I had about 30 hours to commit to a replacement before we'd lose the shutdown window.
Normally I'd have run quotes through three vendors over two weeks. No time. I went with our existing integrated supplier because they already held our drawings and I trusted the PM personally. In hindsight I should have pushed back on the shutdown date — but the operations director was waiting on my call and I made it with incomplete information. The unit worked. It also cost about 8% more than an unbundled equivalent would have. I'd probably make the same call again under the same pressure, which is an uncomfortable thing to admit about a data-driven process.
So which model should you choose?
"It depends" is the only honest answer, and I've watched both models win on real projects. Here's where each one has a defensible edge:
Unbundled sourcing tends to make sense when:
- Your scope is simple and standard — replacing a pump skid, or a single pressure vessel with well-understood specs
- You have in-house engineering capacity that can actually own the design/fabrication interface (not outsource it to a consultant you'll then have to coordinate)
- The project is under roughly $75k, where coordination overhead eats the savings
- You have a preferred installation contractor already trusted on your site
Integrated sourcing tends to make sense when:
- Your process is specialty or non-standard, where design and fabrication can't be cleanly separated
- Your team lacks dedicated project-engineering capacity (ours does, most years)
- The warranty-failure scenario carries production impact you can't absorb
- Timeline matters more than the last 10–15% of purchase price
And there's a third case that's worth pricing out: hybrid sourcing, where an integrated chemical process equipment supplier handles design and fabrication, but installation splits to your existing site contractor. That captures most of the design-handoff savings without paying for installation markup you don't need. It's also harder to quote cleanly, which is probably why fewer suppliers pitch it. If you ask for it directly, some will say yes — they just won't lead with it.
If this is your first time running this comparison on a project above $100k, do the TCO math before you sign. The lowest quote is a data point, not a decision. Price the design interface, the installation coordination, and the warranty coordination, even as rough estimates. When the total spread is under 15%, the integrated option has usually been the safer bet in our tracking. When it's over 25%, unbundled has been worth the project-management overhead.
That isn't a universal rule — it's what 11 projects over six years have looked like in our plant. If your scale or process is materially different, run the numbers on your own data. The crossover point is probably different, and I'd be genuinely curious where it lands for you.