The call I wasn't expecting on a Tuesday
In late April 2023, our pool assembly supervisor called me before 8 a.m. A pallet of ARC resin had come in for our plastic pools line, and the color on the trial-molded side panels looked, in his words, "a little off." He's not a color guy—he runs a line—so when he flags something, I pay attention.
I'm the quality and brand compliance manager at jsp. I review every incoming material batch before it touches production—roughly 240 batches a year across our product lines—and I rejected about 11% of first deliveries in 2023. That number isn't a boast. It's a cost. But it's a smaller cost than a bad logo on 4,000 pool panels.
The batch in question was supposed to match our approved jsp logo color, a specific blue we've held for years. The supplier had shipped it with a spec sheet, a certificate of analysis, and a scan-in through their jsp app that was supposed to give us batch-level traceability. On paper, everything was in order.
What the spec sheet didn't say
Here's the part that gets buried in supplier onboarding meetings: "thermoplastic resin" isn't one thing. If you've ever wondered what is thermoplastic resin, the short answer is that it's a polymer that softens when heated and hardens when cooled, which is why it's moldable and recyclable—and also why its final color depends on far more than the pigment you specify. Dispersal, pellet uniformity, and processing temperature all shift the result. Two suppliers can quote you the same ASTM D638 tensile numbers and still produce panels that look like they came from different decades.
We measured the trial panels against our master color standard. Delta E came back at 3.4. For anyone who doesn't live in color space: industry guidance from the Pantone Color Matching System puts the tolerance for brand-critical colors at Delta E < 2. Between 2 and 4 is noticeable to trained observers. Above 4 is visible to most people. Our jsp logo sits on a consumer-facing pool panel, so 3.4 wasn't a rounding error. It was a reject.
The supplier's response was polite and, frankly, predictable: "That's within industry standard for this grade." I've heard a version of that line maybe a dozen times in four years. It's technically true in the sense that some commodity applications tolerate Delta E of 5 or more. It's also irrelevant, because the standard that matters is the one in the contract—and ours said < 2.
The assumption I got wrong
Here's where I want to be honest about my own process. My first assumption was that the color drift came from our molding line—maybe a temperature fluctuation during the trial run. I assumed the resin was fine and we'd introduced the variable. Didn't verify before I said it out loud in a meeting, which was a mistake. I spent two days chasing our own process data before a technician asked the obvious question: had we run a blind comparison?
We hadn't. So we did. We molded identical panels from the disputed ARC resin batch and from a retained sample of our previous approved batch, same machine, same operator, same day. Then we showed them to six people from packaging and customer service—none of whom knew which was which—and asked which looked more like the jsp logo blue. Five of six picked the old batch. One asked if the other panel was a different product entirely.
That was the turn. The resin wasn't "a little off." The pigment dispersion in the new batch was uneven in a way that showed up under the specific molding conditions for our pool panels. The supplier's process was optimized for a different application—probably extrusion, based on the pellet geometry—and nobody had told us that when we placed the order.
People assume a broad material portfolio means a supplier can handle your application. What they don't see is how many of those materials are run through the same generic process window.
Where we drew the line
We rejected the batch. The supplier redid it at their cost, and the redo came back at Delta E 1.7—fine, but it took five weeks and pushed our pool panel run into a compressed window that cost us overtime on the assembly side. I won't quote the exact figure because I don't have the final invoice in front of me, but the overtime alone was somewhere in the low five figures.
The more uncomfortable conversation came after. This supplier had pitched themselves as a full-service partner: resin supply, custom molding, even finished plastic pools if we wanted to expand. That breadth was the reason we'd shortlisted them. And that breadth was exactly the problem. They were competent at a lot of things and excellent at none of them, at least not for our application.
I'd rather work with a specialist who knows their limits than a generalist who overpromises. The vendor who told me, in a later evaluation, "we're not the right fit for your pool panel color work—here's who is" earned my trust for everything else they quoted. That sounds counterintuitive when you're trying to consolidate suppliers. It isn't. A supplier who can articulate their boundary is a supplier who has actually thought about where their process breaks down.
What I tell every supplier now
Three things changed after that batch.
First, our incoming material contracts now specify Delta E tolerance explicitly, by application, with the Pantone reference cited and a defined measurement method. No more "industry standard" hand-waving.
Second, the jsp app traceability scan is now a gating step, not a nice-to-have. If a batch's scan data doesn't include the specific production line and the resin lot, it doesn't get accepted at the dock. We found out later that the disputed batch's app record had been entered manually, after the fact—something we now audit quarterly.
Third, and this is the one I care about most: we ask suppliers directly what they don't do well. The answers are revealing. The ones who pause and give you a specific, technical limitation are usually the ones whose quality holds up. The ones who say "we can do anything" are telling you they haven't measured where their process stops being reliable.
I'd argue the same applies to how we talk about ourselves. We're good at industrial thermoplastics and custom molding for our core lines. We're not a color-matching house, and we don't pretend to be. Saying that out loud has cost us exactly zero deals that I know of, and it's saved us at least one pool panel run since.
The rejected ARC resin batch is still on a shelf in our quality lab, labeled and dated. I keep it there for new hires. Not as a cautionary tale about a bad supplier—the supplier wasn't bad. It's a reminder that "within industry standard" and "within our standard" are two different sentences, and only one of them belongs in a contract.