Most procurement teams evaluate plastic recycling line quotes in the wrong order. They compare price, then lead time, then — if anyone still has the energy — engineering specs. That order has cost me more money than any single vendor decision I've made in seven years.
If you're auditing specs after you've issued the PO, you're not sourcing. You're defending a number you've already committed to.
I'm the procurement manager at a 140-person plastics recycling company. I've owned our capital equipment budget since 2019 — call it $2.4M a year on average — and I've negotiated with 30+ equipment vendors. Every quote since 2021 is logged in our TCO sheet, mostly because I got tired of explaining why the actual figure didn't match the approved figure.
The $61,000 order that changed how I read a quote
In 2022 we tendered a waste plastic recycling line. Five vendors responded. We shortlisted two on capacity and cost-per-tonne-of-throughput, and picked the one that came in roughly 4% cheaper. The engineering review took about forty minutes.
The line installed fine. Commissioning went fine. Then actual feedstock started running through it.
The problem was in the feed assumption. Our vendor had priced the machine against a reasonably clean rigid stream. Our real input was around 40% rigid, 60% film, with a moisture range that swung hard depending on the season. That wasn't a secret — it was in our own feed survey. It just never got mapped against the quote, because the quote didn't state any assumptions at all. It said "throughput: 1,200 kg/h" and we nodded at the number.
Eleven months later we'd spent roughly $61,000 on pre-sort modifications, replacement screw elements, and eleven weeks of lost throughput. The 4% we saved was, if I remember correctly, about $14,000. Most expensive discount we ever took.
Five minutes of verification beats five weeks of commissioning.
I want to be honest about something here. My first instinct after that project wasn't "build a spec checklist." It was "negotiate harder next time." It took me two more years — and probably another four or five quotes — to understand that the money was never in the negotiation. It was in the review nobody had time for.
Three things a spec sheet won't tell you
A plastic recycling line quote without a stated feed assumption isn't a spec. It's a price.
1. Throughput is a range, not a number
"1,200 kg/h" only means something once you know what's going in. Film, rigid, washed, unwashed — the same machine behaves differently across all four. The same trap shows up on a mixed plastic recycling line, where the polymer split changes what the machine is actually being asked to do. If a vendor doesn't publish the range, ask for it in writing before you compare anything else.
2. Sorting purity is a function of input, not equipment
On a plastic recycling sorting line, the purity figure that gets quoted is usually measured against one specific polymer mix. Change that mix and the same machine, run with the same settings, produces a different result. Almost nobody puts that caveat in the proposal. I've since learned to ask which feedstock the number was derived from — and to ask for the test report, not the brochure spec.
3. A washing line hides its own operating cost
A plastic recycling washing line is not just equipment. It's water recirculation, chemistry, and sludge handling, and every one of those is a per-tonne cost that never appears on the quote. I once compared two washing setups that looked nearly identical on capex and diverged by roughly 30% on total operating cost per tonne. That was several years ago and I wouldn't swear to the exact figure, but the direction was unmistakable.
4. Money spent on material trials outperforms money spent on price pressure
This one took me the longest to accept. Most suppliers will run your material through a test cell or pilot rig. We paid about $4,800 across two candidates — freight and travel included — and that trial single-handedly eliminated the option that had looked stronger on paper, because its yield fell apart at our film ratio. The cost to retrofit it later, given that vendor's rejection terms, would have landed somewhere around $90,000.
So roughly $4,800 to avoid roughly $90,000. I now treat a material trial as non-negotiable on anything above $150,000 in capex. Not a nice-to-have. A gate.
Two objections I hear every time I say this
"This slows procurement down."
It does — by maybe two weeks. For context, we process three or four equipment purchases a quarter. Two extra weeks per project has never once been the constraint. The rework from one bad order, meanwhile, runs into six figures. The time you saved gets spent later, at a worse exchange rate.
"The vendor says their standard machine already handles our material."
They probably believe that. Salespeople believe what they sell. But "standard machine" is the phrase that hides everything. Ask for three things instead of a reassurance: reference sites running comparable feedstock, a sizing report against your actual feed survey, and a written statement of the conditions under which that throughput number was measured. If a vendor won't put it in writing, that's a signal — not necessarily a dealbreaker, but a signal.
What I'd do first if I started over
Two things, both before any quote gets opened:
- Write down your material. Polymer breakdown, film-to-rigid ratio, moisture range, contaminant types, how much the feed varies week to week. One page. Every serious vendor should receive the same document.
- Make each vendor respond to those assumptions line by line. Not just the throughput figure — the assumed feed behind it, and what happens when the feed drifts outside it.
Then compare technical proposals. Then discuss price. Flipping that order has saved us somewhere in the $40,000–$50,000 range on a single handling project, though I'd be careful with that number — there's indirect cost in the tracker that I can't cleanly attribute.
One more thing worth mentioning: compliance documentation. CE marking, ISO 9001 — these are useful reference points, and a supplier that genuinely meets them can usually produce the certificates without a lot of back-and-forth. Incomplete templates are a hint that the technical annex will be the same quality. Strictly speaking that's a commercial check, not an engineering one. It still tells you something.
Where I've landed
Nobody argues against checking things. Everyone agrees prevention is cheaper than rework. Then the actual process runs on price-first, and the review gets compressed into whatever's left of the calendar.
On capital equipment that trade is a bad one, because the numbers are large and the alternatives are expensive. The twenty minutes you spend before signing — reading what the spec actually commits the vendor to — is, in my experience, the single highest-return step in plastic recycling line sourcing. Higher than the negotiation. Higher than the reference calls. Higher than shopping one more vendor.
Cheap is real. It just isn't cheap until it survives contact with your actual material.