I've been handling case packer sourcing orders for seven years. In that time I've personally made — and written down — four significant mistakes, totaling roughly $41,000 in wasted budget. I'll get to the number in a second, and I might be off by a couple thousand because I'm working from a spreadsheet I stopped updating in 2024 (note to self: update it).

The reason I keep the list is that none of those four mistakes were about picking a bad manufacturer. Every machine we bought was built to the spec we gave. That's the part that took me three years to understand.

The Surface Problem: Three Quotes That Look Identical

Picture the normal case packer sourcing cycle. You need a machine. You send out an RFQ with your case dimensions, your target rate, your corrugate grade. Three or four manufacturers come back. All of them quote a high speed case packer at 30 cases per minute. All of them list washdown, stainless frame, Allen-Bradley controls. Prices land within 15% of each other. Lead times are the tiebreaker.

You pick. You feel good about it. Fourteen months later the line is running at 22 cases a minute on a good shift and your operations manager is asking you why you bought a machine that doesn't do what it says on the tin.

That was my 2020. And for a while I genuinely believed the problem was vendor selection — that I'd picked the wrong one, that the next one would be better. It wasn't. The second machine had the same problem, just wearing a different logo.

What's Actually Going Wrong

You're comparing machines, not production realities

A datasheet describes a case packer machine running in isolation, fed by a perfect infeed, with ideal corrugate and an operator who never walks away. Your plant doesn't have any of that.

In September 2022 I approved a robotic case packer cell rated at 32 cases a minute. Great number. What I hadn't modelled was the case erector sitting upstream — it ran at 24 a minute with about three short stoppages an hour. The robot was never the constraint. It spent a meaningful chunk of every shift waiting, and when the erector hiccuped, cases backed up into the accumulation table and the cell faulted out. We spent $14,000 on integration rework and never once hit 32.

Here's the thing I'd tell anyone starting a case packer sourcing project: the machine speed matters less than the gap between your machine and whatever feeds it. That gap is where the money disappears.

The assumption is that slow machine speed causes downtime. Usually it's mismatch

People assume an underperforming case packer is a machine quality problem. Often it's a buffer problem. The machine is doing exactly what it was built to do — it's just being starved, or blocked, or forced to run at a rate its infeed can't sustain.

Once I understood that, my RFQs changed completely. Instead of asking "what rate can this machine hit," I started asking "what rate can this machine sustain given a 60-second upstream stoppage every 20 minutes." Only two of the six manufacturers I've worked with answered that question without hedging. Those two are the ones I still call.

Changeover time is where the money hides

My worst mistake was in Q1 2024, and it was embarrassingly small. The quote said "12-minute changeover." Nice. What I didn't read was the footnote: with a second set of change parts. The change parts weren't in the price. They were $9,800 on a separate order.

Even with them, actual changeover ran 40 minutes, not 12. Twelve minutes assumed two operators, both trained, working in parallel with everything staged beforehand. On a Tuesday afternoon with one operator and the parts still in the crate? Forty minutes, maybe forty-five.

That machine did two changeovers a day, 250 days a year. Twenty-eight extra minutes each, times 500: about 233 hours of lost production annually. I'll leave the dollar math to you, but on our line the fully loaded cost of unplanned downtime ran around $4,300 an hour — that figure's from our own 2023 accounting, it includes idle labor and late-shipment penalties, and it may not match yours.

Speed you pay for isn't always speed you use

I went back and forth between a 25-case-per-minute wrap around case packer and a 40-case-per-minute high speed model for about three weeks. The high speed option was $38,000 more. On paper it made sense — future growth, headroom, all the standard arguments.

What killed that idea was a two-hour conversation with our filler operator. Our filler was the constraint. It topped out at 24 cases a minute on a good day. Buying a 40 cpm case packer to sit behind a 24 cpm filler is buying an expensive machine that waits. We bought the wrap around unit and put the difference into a second accumulation table, which did more for throughput than the extra speed ever would have.

If your OEE is sitting in the 60s and your bottleneck isn't the case packer, extra speed buys you nothing but a bigger depreciation line. World-class OEE of 85% has been the reference point since Nakajima's TPM work in the late 1980s, and most packaging lines aren't close to it — ours wasn't.

What This Actually Costs

My first mistake, back in March 2020, was buying a wrap around case packer sized to a corrugate spec we never re-verified. Our board supplier had quietly shifted grade during a shortage, and nobody told us. We ran about 1,800 cases before a quality tech caught the flap scoring. Around $6,400 in scrap and rework, plus four days of line time.

The 2022 robotic cell was the expensive one — integration rework, a delayed start, one customer chargeback on a late shipment. Call it $19,000 when I add it all up, though I'd have to pull the file to be precise.

But the money isn't the worst of it. The worst is credibility. When you're the person who signs off on capital equipment and the equipment doesn't perform, every future PO you raise gets a second look from someone above you. That lasts longer than the invoice.

The Check I Now Run Before Any PO

I keep an 11-point list. The full version lives in a shared doc that three colleagues have edited. The five points that would have caught all four of my mistakes:

  • Model the line, not the machine. Ask every manufacturer to quote sustained throughput given your actual upstream stoppage pattern. Written answer, not conversation.
  • Get the changeover claim in writing, with staffing and parts included. If the number assumes two trained operators and a spare change part set, both of those need to be priced in the quote.
  • Confirm your corrugate spec in the same month you sign. Board grades drift. Get a letter from your supplier, date it, attach it to the PO.
  • Ask what the machine does when it faults. Recovery time matters more than cycle rate on a real line.
  • Ask for the last three installations running a similar case size — and call one of them without the manufacturer on the line.

For budget planning: wrap around case packers in the 15–25 cases/min range quoted us between roughly $45,000 and $95,000 through our 2024 sourcing rounds, and single-cell robotic case packer configurations ran $110,000 to $260,000 with tooling, guarding, and integration included. Those are ranges from quotes we collected between March and November 2024 — verify current pricing, because servo and steel costs move and I'm not going to pretend a 2024 number is a 2026 number.

What I Can't Help You With

I'm not a controls or safety engineer, so I can't speak to the risk assessment and validation work. If your machine falls under ANSI/PMMI B155.1 for packaging machinery safety, or your controls need performance-level validation under ISO 13849-1, that's a qualified integrator's job — budget for it rather than assuming the OEM covers it. Same with machine guarding obligations under OSHA 29 CFR 1910.212 in the US.

And I'm definitely not your compliance route into Europe. The EU Machinery Regulation (EU) 2023/1230 replaces the old Machinery Directive and applies from 20 January 2027 — verify the current requirements at eur-lex.europa.eu before you sign anything with a CE mark attached. That's outside what I do.

What I can tell you is what seven years of getting it wrong taught me: the specification conversation happens before the RFQ goes out, not after the quotes come back. Five minutes of verification beats five days of correction — and in capital equipment, sometimes it beats five months.