Cost Per Item vs Profit Per Printer-Hour — When to Switch

Cost per item is the right number when you're starting out. Here's how to tell when you've outgrown it — with real numbers from a real client quote.

A client sent me a model last month and asked for a price. Simple enough: a two-part display piece, multi-colour, and they wanted a few sizes to choose from.

I did what I always tell people to do — I sliced it before quoting. (I've been burned: I once quoted $60 for something that turned out to be a $175 print, because I priced it from memory instead of from the slicer. Never again.)

Chasing down what the slicer showed me changed how I price everything. But before we get there, I want to start somewhere less glamorous — because if you're early in your journey, the fancy metric at the end of this post isn't the one you need yet.

Start here: cost per item

When you're getting going, one number matters more than any other: what does it cost me to make this thing, and am I charging more than that?

Materials, electricity, a bit of machine wear, your time. Add it up, compare it to your price, and make sure the gap is positive. That's cost per item, and it's the right first metric for a simple reason:

When you're starting out, your printers have spare time. They're sitting idle more than they're running. So the machine hours are effectively free — there's nothing else they'd be doing anyway — and the only real question is whether each sale clears its own costs.

If that's where you are, don't overthink it. Our 3D print cost calculator does exactly this — filament, electricity, machine wear, labour, and a suggested price — for free, no signup. Start there. It's the right tool for this stage, and plenty of studios never need anything more.

But some of you will. And the day you do, the metric that got you here starts to lie.

The day your printers are full

Here's what changed for me. I sliced that client's model, and two settings turned out to matter enormously.

First — support in places you'll never see. The model has recessed areas, and the default profile was building support material down inside a cavity that's invisible once the part is finished. So I switched the slicer to "critical regions only" support. Result: ~28% off the cost. Not from the filament saved — that's pennies — but because support removal is hand labour, and pulling supports out of a tight cavity is slow, fiddly work. Less support, less time with an X-Acto knife.

Second — trade a little filament for a lot of time. The multi-colour version swaps filament on nearly every layer, and the interior sliced into thousands of tiny disconnected segments, each needing the nozzle to travel to it. So I closed the walls up: more perimeters, zero infill. The extra walls turn those fragments into continuous paths.

The trade:

+30% filament

−50% print time

I added material and the print got twice as fast.

Here's the before and after on the full-size piece:

ㅤ	Before	After
Print time	123.75 h	58.5 h
Cost to make	$211	$119
Profit at my price	$264	$356

Same model, same printer, same price to the client. Two checkboxes, and I make more per unit while the machine is tied up for less than half as long.

Now here's the thing. If my printers were idle, this would be a nice-to-have. Cheaper is cheaper. But my printers aren't idle anymore — and that changes which number I should be looking at.

Why cost per item stops working

When your machines are busy, they stop being free. Every hour a printer spends on one job is an hour it can't spend on another. Machine time becomes your bottleneck — and cost per item can't see bottlenecks.

Watch what it misses. Two products:

Product A: $12 profit, ties up a printer for 3 hours

Product B: $5 profit, takes 30 minutes

Cost per item says A wins — more than double the profit. But per hour of machine time, A earns $4/hour and B earns $10/hour. When your printers are the constraint, B is more than twice the business A is. Cost per item will happily steer you toward the worse product, because it can't see the resource you're actually short on.

That's why, on that client's job, the number I really cared about wasn't the cost. It was this:

Profit per printer-hour: $1.86 → $6.09

Two checkboxes moved it more than 3×. And once you start thinking this way, everything shifts:

A "big" order that ties up a printer for two weeks might be worse than a steady stream of small ones.

A faster slicer profile is worth more than a price increase you'd have to negotiate for.

The question stops being "what does this cost?" and becomes "is this the best thing my printer could be doing right now?"

Which stage are you in?

Not everyone needs to make this jump, and I want to be honest about that. A two-printer side hustle with time to spare can live on cost per item forever and do just fine. There's nothing wrong with that stage — I was there, and the calculator above is genuinely all you need for it.

The point isn't "cost per item is bad." It's:

Printers mostly idle? Cost per item is your metric. Charge above cost, keep the machines busy, don't overthink it.

Printers mostly busy? Profit per printer-hour is your metric. Your constraint is hours, not dollars, and you should be optimising the thing you're short on.

Know which one you're in. Use the metric that fits it. And notice when you've crossed over — because the crossover is quiet, and the old number keeps looking fine long after it's stopped being right.

What you can do today

If your printers are busy, try the profit-per-printer-hour test on your top five products. Take each one's profit, divide by its print time in hours, and rank them. The order will surprise you.

Attack your slicer settings before your prices. Support in hidden cavities and fragmented infill are quietly costing you machine hours on every print. A checkbox is cheaper than a price negotiation.

Never quote from memory. Slice it first. Ten minutes, and it's the whole job. I learned that one the expensive way.

We're building the tool that does the second half of this automatically — slice once, price every size, and see profit per printer-hour on every job. It's coming soon, and the people on our list will be the first through the door.

Join the waitlist to be first to know when it's out — and in the meantime, the cost calculator is live and free for wherever you are in the journey.

Come argue with us on Discord — we're building this in the open.