Capacity Utilization in Food Manufacturing: What Good Looks Like

Capacity utilization in food manufacturing is the number everyone quotes and almost nobody counts the same way. One plant owner measures against a 24/7 calendar and thinks he's dying at 40%. Another measures against last week's schedule and thinks she's crushing it at 92%. Both numbers lie. Here's what a good rate actually looks like for a co-packer, how to count line-hours so the number means something, and how to tell whether a low number says "go sell" or "go fix changeovers."

What Does Good Capacity Utilization in Food Manufacturing Look Like?

75 to 85% of staffed hours. That's the target for a $2-15M co-packer. Not 100.

For reference, the Fed has tracked capacity utilization across US food manufacturing for decades, and the industry-wide number typically sits in the mid-to-high 70s. That's big-plant, all-shifts data — but it should recalibrate anyone who thinks 95% is the goal.

Why not 100? Because a full plant is a fragile plant. At 95% you have no room for a sanitation window that runs long, no slack for a line trial with a prospect, no way to absorb a rush order from your best customer. Every breakdown becomes a missed ship date. And here's the part that bites growing co-packers: you can't sell capacity you don't have. A plant at 97% has nothing to offer the next brand that calls.

The bands look roughly like this:

• Under 60%: fixed overhead — rent, insurance, salaried QA, that $400K filler — is spread across too few hours. Margins get crushed even when jobs price well.
• 60-75%: viable but thin. One lost customer puts you back in the danger zone.
• 75-85%: the sweet spot. Efficient enough to make money, loose enough to say yes.
• Above 85%: you're quoting six-week lead times, losing bids, and running your people ragged. Raise prices or add a shift.

How to Calculate Line Utilization Without Lying to Yourself

The formula is simple: hours the line ran product, divided by available line-hours. The fight is over the denominator. There are three ways to count it, and they answer different questions:

1. Theoretical hours — 24 hours, 7 days: 168 per line per week. Use this only when you're thinking about expansion headroom. It's your ceiling, not your report card.
2. Staffed hours — the shifts you actually crew. This is the denominator for utilization. If you run one shift, five days, your line has 40 available hours whether or not the building is open at midnight.
3. Scheduled hours — what you planned this week. Dividing by this gives you schedule adherence, not utilization. Useful, but a different number.

The numerator needs honesty too. Count hours the line actually produced sellable product. Changeovers don't count. CIP doesn't count. Waiting on a late ingredient truck doesn't count. Idle-but-staffed doesn't count.

Worked example. A 38-person sauce plant in Ohio runs two filling lines, one shift, five days: 80 staffed line-hours a week. Last month the lines averaged 46 hours of actual production. That's 57.5% utilization — the real number. Measured against theoretical capacity (336 line-hours), it's 14% — which isn't a grade. It's a map of how far a second shift and weekend runs could take them without buying a single machine.

OEE vs Capacity Utilization: Two Different Questions

Owners mix these up constantly, usually because an ERP vendor put them on the same dashboard.

OEE — availability times performance times quality — measures how well a line runs during the hours you run it. Slow cycles, short stops, rework: that's OEE territory. World-class is around 85%; most food plants honestly sit in the 50s and 60s.

Capacity utilization measures how much of your staffed time you had product to run at all. It's a commercial number wearing an ops costume.

You can post 85% OEE at 50% utilization: the line hums when it runs, and it doesn't run because the pipeline's empty. You can also post 90% utilization at 55% OEE: booked solid and bleeding hours to jams and giveaway. Different diseases, different medicine.

Capacity Utilization Benchmarks for a Co-Packer

Here's where the bands land once you're counting against staffed hours honestly:

One caveat: annual averages hide seasonality. A BBQ sauce co-packer might run 90% from April through July and 45% in January, and the annual 68% describes neither plant. That swing is a seasonal capacity planning problem with its own playbook — averaging it away just hides it.

When 55% Means "Sell" and When It Means "Fix Changeovers"

Two plants at 55% can have opposite problems. The diagnosis lives in the shape of the missing hours.

Pull last month's line log and sort every lost hour into two buckets: no order to run, and had an order, line wasn't running it. Whichever bucket is bigger is your real problem.

If the empty time comes in big blocks — whole afternoons, a line dark every Friday — that's demand. Your changeovers could be pit-crew fast and you'd still be at 55%. No amount of ops work fills a calendar that has no orders on it. That's a sales problem, and it has a sales solution.

If the lost time is shredded, it's ops and scheduling. Twelve changeovers a week at three hours each is 36 hours; on 80 staffed hours, changeovers alone cap you at 55% before a single machine hiccups. Fix allergen sequencing, batch like SKUs together, attack the two worst changeovers, and you can claw back 15-20 points without adding one customer. And a lot of that shredding traces back to brands that drop POs on Thursday for Monday — there's a way to handle scheduling production when brands won't forecast that doesn't involve eating the chaos yourself.

Closing the Gap

If the diagnosis is demand, treat empty line-hours like the perishable inventory they are — an hour that passes unsold is gone forever. There's a full playbook on how to turn idle line time into contract revenue, but the short version: nobody's coming to find you. Referrals and trade shows fill capacity by accident; outbound to emerging brands that need a manufacturing home fills it on purpose. Building that outbound motion is exactly what Feed The Line does for co-packers — but whether you build it in-house or hire it out, build it. A plant at 55% doesn't have a utilization problem. It has a pipeline problem with a utilization symptom.

If the diagnosis is ops, spend nothing on sales until sequencing is fixed. Selling into a shredded schedule makes the shredding worse — more SKUs, more changeovers, angrier customers.

Either way: count against staffed hours, target 75-85%, and diagnose before you fix. The number only helps if you count it honestly.

Frequently asked questions