It was a gray Monday in March 2024, and the smell hit me before I opened the first box. Musty. Not "a little old" musty, but "this has been stored somewhere damp for a very long time" musty.
I'm a quality/compliance manager at an indoor entertainment company that supplies commercial venues. For the past five years, I've reviewed roughly 200 unique items a year—slot machines, trampoline safety pads, board games, fitness equipment, spare parts, you name it. In Q1 2024 alone, I rejected about 2.4% of first deliveries. When I implemented our verification protocol in 2022, our field failure rate dropped by 31%. That's the number I think about when a supplier tells me I'm being too picky.
The shipment on the dock was a batch of igt slot machine parts 1984, sourced for a customer who keeps a row of vintage machines in their venue. We'd put together an old IGT slot machines list a few years back to help operators identify which machine used which board. That list is one of the reasons customers come to us. It is also the reason I get to spend my Tuesday with a flashlight and a magnifier.
That old IGT slot machines list started as a service nightmare in 2022. We kept finding machines where someone had swapped a board with what looked like a compatible replacement—and it wasn't. So we spent a month documenting every revision we could find. The list isn't glamorous. It's just useful.
The supplier's photos made the parts look clean. The first open box had a reel motor with a cracked mounting tab. I put it aside. Second box: a power board with a burnt resistor near the rectifier. Third box: a wiring harness that didn't match the pinout listed in our old IGT slot machines list. Not a subtle difference—a 9-pin connector instead of a 12-pin. That's the kind of part that gets installed, fails instantly, and makes everyone involved look bad.
At that point, I did what quality people call "opening the sampling plan." Which, in plain language, means I stopped trusting the vendor and started checking every single box. Fourteen boxes. Three failures. The failures were not random: they were all in parts that had been sitting in secondary inventory too long, or pulled from a different product line. The vendor said it was "within industry standards." We're not in the industry-standard business. We're in the "it has to work in a commercial venue" business.
Then the real decision hit. The customer's technician was already scheduled, and the supplier said replacement parts wouldn't arrive for three weeks. The upside of accepting the lot was keeping the service appointment. The risk was shipping boards that could die on a machine floor and turning a one-time repair into a pattern of failure. I kept asking myself the same question I always ask in this situation: is the short-term save worth potentially making the customer doubt everything we send?
I had two hours to decide before the rush rework window closed. Normally I'd want to bench-test every board, but there was no time. So I made the call: accepted the boxes that passed the visual and continuity checks, rejected the three problem boxes, and negotiated a credit. Not ideal. But workable.
The Same Checklist, Applied to a Squat Rack Home Gym
About a month later, a client asked me to spec a squat rack home gym for a fitness corner they were building in their entertainment complex. No slot machine components in sight. No compliance laws, no reels, no wiring. But the inspection mindset was exactly the same.
We had two options. The first was a branded unit from one of our regular fitness suppliers. The second was a much cheaper rack that looked nearly identical in the product photo. I asked for dimension drawings. The drawings looked okay. Then I asked for actual samples.
Here's the thing: sample testing is not just for electronics. The cheaper rack's J-hooks were 4mm shallower than the approved drawing. On paper, 4mm seems like nothing. But when I loaded a bar with representative weight and racked it, the bar had almost no margin to sit on the hook. If a user racks the bar slightly off-center, it can slip. I rejected that rack and wrote a note: "J-hook depth must match approved drawing exactly." The client thanked me later because another venue in their franchise group had almost bought the same rack.
I also checked the rack's bolt holes. The cheaper rack had slots instead of holes on the uprights. Slots make assembly easier, but they also move under load. For a home gym that might see daily use, I didn't want to explain that tradeoff after an incident. We wrote the rack's spec requirement into the quote, and the client bought the branded unit.
Even a Strategy Board Game Follows the Rules
The same quarter, we introduced a new strategy board game to our catalog. It's a lightweight one, but it has to represent our brand on a shelf. The game board's red was wrong. The printer said it was "within tolerance." I measured it against our approved Pantone reference and got a Delta E of 5.2. For brand-critical colors, industry standard is Delta E < 2 per Pantone Color Matching System guidelines. That's not a place for opinions.
We rejected the first print run and had the covers redone. It cost us about $1,800 and a week of schedule. That delayed the catalog launch, which annoyed exactly one person: me. But the alternative—shipping a product with a color that weakens the brand—would have annoyed me longer. I've seen how fast a small spec miss turns into a shelf full of unsellable products.
The manual had a similar issue. The text was okay, but the map artwork looked soft because the file was only 180 DPI. Standard commercial print is 300 DPI at final size. The printer said it would be fine at that size. It wasn't. We sent new files, too.
The Question I Can't Answer
Around the same time, a venue owner asked me, "What are the best earbuds for iPhone?" I laughed, because I don't inspect earbuds. But the question kept coming back to me, because it's the wrong question in any purchasing process. "Best" needs a context. For a floor manager who spends shifts in a noisy game room, the best earbuds are the ones with good microphone noise rejection and a battery that lasts a whole shift. For someone who just wants to watch setup videos with the phone on the other side of the room, the priority is low latency and range. Same logic applies to a squat rack home gym, a strategy board game, or a box of 1984 slot machine parts: you can't evaluate the option until you define the job.
That's not a cop-out. It's the difference between a recommendation and a decision. If I just gave you the "best" list, I'd be doing you a disservice. The best earbuds for a sound engineer are not the best earbuds for a forklift operator. Same with racks and games and slot boards.
I have mixed feelings about NOS parts. On one hand, they can keep a machine from 1984 alive for another decade. On the other, "new old stock" can mean "old plastic that no one has powered up since the Reagan administration." That's why the list is only the start. An old IGT slot machines list tells you what fits. It doesn't tell you what will last.
That's what the 1984 parts order taught me. A list is only a starting point. A spec sheet is a promise, not a fact. And the best purchase decision isn't the one that's cheapest or fastest—it's the one where someone checked the actual item against the actual need.
There's something satisfying about all of this. After the stress, the rejected shipments, and the uncomfortable phone calls, seeing a corrected order roll out—knowing that a machine won't fail because someone took the time to measure a J-hook or check a Delta E—that's the payoff. The inspection is never perfect. But when it works, it makes everyone's life boring in the best possible way. And boring is exactly what I want.