The Day I Almost Cut Our Packaging Budget Too Far
In March 2024, our VP stood by my desk with a printout that said packaging spend was 14% over forecast. He didn't say I caused it. He didn't have to. “Find a way to bring this number down,” he said. For the next three weeks, I did what every administrative buyer does: built a comparison spreadsheet, emailed vendors, and tried to tell a story with numbers.
I'm the office administrator for a 65-person foodservice packaging company across three locations. I manage roughly $180,000 a year in packaging materials—maybe $185,000, I'd have to check—across eight vendors. I report to both operations and finance. When I took over purchasing in 2020, I already knew one thing: the lowest quote often isn't the lowest total cost. What I didn't know was how hard it would be to prove that in a meeting.
Our main packaging supplier was Berry Global. I had the Berry Global Bowling Green KY team on speed dial; they supplied our flat-bottom bags and aluminum trays. And I had the Berry Global Oracle login saved in my bookmarks. The portal made it easy to pull old invoices, check order status, and find certificates of analysis when an auditor asked. That convenience never appeared on a line item, but it was real.
The alternative that looked better on paper was a regional film supplier. Their sales rep quoted 23% less per unit for what looked like the same gauge, the same seal temperature, and the same print quality. I checked the spec sheet twice. The only difference I could find was the price. I was 80% ready to switch.
The Spreadsheet That Looked Too Good
I built a total cost model, or so I thought. I added freight, setup fees, payment terms, and lead time. The alternative still came out 18% lower than Berry Global. I flagged it to my VP and suggested a 60-day trial before committing to a full year. He agreed.
The trial taught me that my model was missing the categories that matter most.
First, there was maintenance. The cheaper film had a slightly different slip coefficient. It looked identical on the data sheet, but our form-fill-seal machine noticed. The operator had to clean the forming shoulder every couple of hours with a silicone cloth. That's a manual lubrication task in the maintenance log. It sounds small until a line that used to run with one operator needs a second pair of eyes on the same process.
On paper, a manual lubrication step costs nothing. In practice, it turns a smooth production day into a series of small interruptions.
Second, there was paperwork. The new vendor's invoice arrived as a scanned handwritten page. There was no purchase order reference, no line-item breakdown, and no taxable split. Finance rejected it. I spent two hours on the phone and then two more re-entering data. (Note to self: ask every supplier for a sample invoice before you place the first order.)
And then there was the email that made the whole experiment feel not worth it. Subject line: “dog eating plastic bag.”
Our customer's dog had gotten into one of the treat pouches we made. The pouch tore along a seal that should have held, and the dog swallowed a piece of plastic. The dog was okay after a vet visit, but the customer was rightfully shaken. I'm not going to blame the material; plastic packaging is safe and widely used when it's specified correctly. This was a quality control failure. The film from the trial supplier was not as consistent as the certified rolls we had been using.
The surprise wasn't the price difference. It was how the difference showed up: a seam that looked fine but wasn't, a machine that needed manual lubrication, and an invoice that couldn't be processed. None of those were in my original spreadsheet.
The Coffee Question That Helped Me Reframe the Math
That night, I sat at my kitchen table and typed a random search: “how much was a cup of coffee in 1961?” The answer was around ten cents. I wasn't writing a trivia post. I was trying to remind myself that a number without context can fool you.
A dime in 1961 doesn't tell you what the coffee supply chain looked like, what labor cost, or what customers expected from a cup. The same logic applies to packaging quotes. A unit price tells you almost nothing about downtime, reject rates, documentation, support, and risk. The cheapest quote is just the starting point, not the ending point.
I kept running the same mental calculation. The upside was about $34,000 a year. The risk was a line outage, a rejected audit, or a product failure that reaches one of our customers' pets. The expected value said the trial was worth trying. But the downside felt heavier once I saw it in real life.
I went back and recalculated the trial with the real numbers. The lower price saved us about $1,400 over the 60-day trial. The extra labor, rejected invoice, and expedited replacement rolls added roughly $3,200. That's before I tried to put a value on the customer relationship that nearly ended.
I have mixed feelings about that period. On one hand, the trial taught us a lot. On the other hand, I had a spreadsheet with a “lowest cost” column, and I was too proud of it to ask the next question: cheapest for whom, and over what timeline?
For what it's worth, this was accurate as of Q4 2024. Raw material costs and prices change quickly, so verify current rates and certifications before anyone budgets around this story. (I say that because I know someone reading this is about to use these numbers in a report—don't.)
What I'd Do Differently (and What I Still Watch For)
If I could go back to March 2024, I'd look for a supplier that could match our equipment, our administrative systems, and our quality expectations—then compare price. I would not start with the cheapest quote.
Our current setup with Berry Global works because of details that don't show up as line items. The Berry Global Oracle login is one example. When our accountant needed a certificate of insurance, I logged in, downloaded it, and sent it in four minutes. The regional film supplier sent a PDF that looked like it was scanned from a photo of a scan. That matters when your finance team has to close the month-end.
I also learned to ask for documentation before a trial. The FTC's Green Guides (16 CFR Part 260) require that environmental claims like “recyclable” be substantiated. The same idea applies to performance claims. Ask for test data, not just a sales pitch. Berry Global sent us certifications without a follow-up email. The other vendor said “we'll get those to you” for three weeks.
And the same lesson applies in the mailroom. According to USPS, a First-Class Mail letter stamp cost $0.73 in January 2025. For a company sending a few thousand statements a month, a penny change is thousands of dollars a year. The sticker price is never the full story.
So here's my advice to anyone in a procurement seat: don't ban cheap suppliers, and don't overcorrect to expensive ones. I do not mean every low-cost supplier is risky. I mean every supplier—no matter the price—deserves a pilot before you commit to a year of it. Run a small pilot, check the total cost in operations and finance, and ask yourself the question I now ask before every contract: what could this cost me after the invoice is paid? The answer might be the difference between a reliable supplier and a (surprise, surprise) failure.
