Short answer: a “good” OEE is 85%, the world-class benchmark. But most manufacturers run around 60%, and shops measuring accurately for the first time often land between 40% and 60%. Fewer than 1 in 10 plants ever clear 85%. So if you’re in the 50s or 60s you’re normal, and the gap to 85% is capacity you already own.
| Benchmark | What it is |
|---|---|
| 85% | World-class OEE, the benchmark to measure against |
| ~60% | What a typical manufacturer actually runs |
| ~6% | Share of plants that ever reach world-class |
Every owner who starts measuring machine performance asks the same two questions within a week: is this number good, and what does everyone else run. Fair questions, and the real answers are more useful than the scary ones. Here’s the real benchmark data, what it means for a metal fab or machine shop, and the one thing that decides whether your OEE number is worth trusting at all.
The figures below are third-party industry benchmarks (OEE.com, Evocon), not measurements from Spall’s own customer base. They’re here as a reference point, not a claim about what Spall specifically delivers.
What is a good OEE score?
The widely accepted world-class OEE benchmark is 85%. That figure isn’t arbitrary. It’s what you get when a machine runs about 90% availability, 95% performance, and 99% quality at the same time (per OEE.com’s world-class benchmark). Multiply those three and you land near 85%. Hitting it means very little downtime, running at rated speed, and almost no scrap, all at once, all the time. That combination is rare.
What is the average OEE for manufacturers?
Across the industry, the typical OEE is around 60%, and real-world datasets put the center of the pack a touch lower. Evocon, monitoring more than 3,500 machines across 50-plus countries, found scores cluster in a 36% to 60% band, with only about 6% of operations reaching the 85% mark. The people who publish these numbers add a caveat: once you correct for inflated cycle times and misclassified downtime, the share actually hitting world-class is closer to 3%.
| OEE range | What it usually means |
|---|---|
| 85%+ | World-class. Rare. Fewer than 1 in 10 plants, and closer to 1 in 30 once the numbers are corrected. |
| 60 to 85% | Good and improving. You’re measuring, you’re acting on it, there’s still upside. |
| 40 to 60% | Typical. Where most shops actually live, especially early on. Big recoverable gap. |
| Below 40% | Common on a first accurate measurement. Not a failure, a starting line with a lot of found money in it. |
Why is my OEE lower than the benchmark?
If you just started measuring and the number stings, that’s the normal experience, not a sign your shop is broken. A first accurate OEE reading is almost always lower than people expect, because it finally counts the losses that used to be invisible: the machine that’s “running” but sitting idle between parts, the slow cycle no one flagged, the setup that ran 20 minutes long, the rework that never made it onto a report. None of that showed up on a spreadsheet. All of it was always there.
The gap between your number and 85% isn’t a scolding. It’s a map. Every point of it is capacity you’re already paying for: machines you own, floor space you rent, people you employ, and it’s the cheapest capacity you’ll ever add because you don’t have to buy anything to get it back.
The one thing that decides if your OEE is even real
Here’s the part most benchmark articles skip. An OEE number is only as good as the standard behind it. If your “ideal cycle time” is padded, or downtime gets reclassified as planned, you can post a beautiful 85% that means nothing. A soft 85% is worse than an ugly 55%, because the ugly one is telling you the truth.
So before you compare yourself to anyone, make sure the number is coming straight off the machine, not off a clipboard or a hopeful standard. That’s the difference between a benchmark you can act on and a vanity metric. It’s also why the accurate first number matters more than the impressive one.
How world-class OEE is built
OEE is three numbers multiplied, so a weak spot anywhere drags the whole score down. This is what the 85% benchmark is actually made of:
What’s a realistic OEE target for your shop?
Chasing 85% out of the gate is the wrong goal, and it’s the wrong goal for most shops ever. High-mix, low-volume work with constant setups will never look like a plant running one part around the clock, and that’s fine. The target that matters is your own number, moving up.
A useful way to think about it: find your real baseline, then aim to add 10 points. Going from 50% to 60% on a single machine is often a full extra part every few hours, real capacity, real margin, no capital. Once that’s steady, aim for the next 10. The plants that win aren’t the ones fixated on 85%. They’re the ones who know their real number and nudge it every quarter.
The takeaway for the person signing the checks If your OEE is in the 40s or 50s, you’re in good company, and you’re sitting on capacity you’re already paying for. The number to distrust is a suspiciously high one built on soft standards. Get a real baseline first. The gap to world-class is the cheapest expansion you’ll ever run, and our ROI calculator puts a dollar figure on it.
OEE benchmark FAQ
What is a good OEE percentage? 85% is considered world-class. 60% to 85% is good and improving. Most manufacturers run in the 40% to 60% range, so anything in the 50s or 60s is normal and leaves real room to grow.
What is the average OEE in manufacturing? Around 60%, with large real-world datasets showing most plants clustered between 36% and 60%. Only about 6% of operations ever reach the 85% world-class mark.
Is 60% OEE good? 60% is right around the typical benchmark, so it’s normal, not bad. It also means roughly 40% of your capacity is being lost somewhere, which is a large and usually recoverable opportunity.
What is world-class OEE? World-class OEE is 85%, built from about 90% availability, 95% performance, and 99% quality running at the same time. Fewer than 1 in 10 plants achieve it.
Why is my OEE so low? A first accurate measurement usually looks low because it finally captures losses that were invisible before: idle time between parts, slow cycles, long setups, and rework. The number didn’t get worse, you’re just seeing it accurately for the first time.
What’s a good OEE for a job shop or machine shop? High-mix, low-volume shops naturally run lower than single-part production because of frequent setups. The right target isn’t a universal number, it’s your own real baseline plus about 10 points, repeated.