Most people who start selling prints price by feel, discover after a few months that they are busy and no better off, and conclude that printing is not worth doing commercially. Usually the arithmetic was wrong in one of two specific ways: they left out the machine and their own time, or they confused margin with markup. Both are fixable in an afternoon.
Margin and markup are different numbers
This is the error that costs the most and is the easiest to make.
Markup is a percentage added to your cost. A part costing ten dollars, marked up two-fifths, sells for fourteen. Margin is a percentage of the selling price that is not cost. The same part at a two-fifths margin sells for just under seventeen, because the margin is taken out of the larger number.
The calculator on this site works in margin, because margin is what tells you whether a business survives — it is the fraction of every pound taken that is available to cover everything that is not this specific job. It states the equivalent markup alongside so you can convert between the two rather than guess.
If you have been "adding 50%" to your costs, you have been running about a third margin, not a half. That gap is the difference between a hobby that pays for itself and one that quietly does not.
A commission costed line by line
Four identical parts on one plate. Together they use 150 g at $28/kg and take 8 hours on a machine drawing 110 W. Machine time at 75 cents an hour, US average electricity at 17.5 cents, 25 minutes of your own labour at $25 an hour, and a 10% failure allowance.
| Line | Amount |
|---|---|
| Filament | $4.20 |
| Electricity | $0.15 |
| Machine time | $6.00 |
| Labour | $10.42 |
| Base cost before failures | $20.77 |
| Failure allowance | $1.15 |
| Total cost for the plate | $21.92 |
| Cost per part | $5.48 |
| Price per part at 40% margin | $9.13 |
That last row is a 67% markup on cost, which is the same decision expressed the other way. Labour is the largest line in that table and the one most hobby quotes leave out entirely, which is the whole of the gap between a hobby price and a commercial one. Set the labour rate to nothing in the cost calculator and the price per part roughly halves.
Two things in that table deserve attention. Electricity is 15 cents — a rounding error, and the term people spend the most time worrying about. Labour is the largest single line, larger than the filament and the machine combined, and it is the term most beginners set to zero.
Setting the inputs you cannot look up
Machine rate. Take what the printer cost, decide how many hours of useful life you expect from it before it needs replacing or major work, and divide. Add consumables — nozzles, plates, belts, lubricant. Most desktop machines land somewhere between 25 cents and a dollar an hour. Pick a number, use it consistently, and revisit it yearly.
Labour rate. What your time is worth, not what you think a customer will accept. If the honest number makes the job unprofitable, the job is unprofitable and you have learned that before doing it rather than after.
Labour minutes. Time the whole loop once: opening the file, slicing, loading filament, starting, removing, cleaning up supports, checking, packing. It is nearly always longer than the guess, and for small parts it dwarfs everything else.
Failure rate. Use a real one, sized to the job — counting the cost of a failed print sets out how to measure yours.
What to do about design time
Design is not printing and should not be priced as though it were. Three workable approaches:
- Charge it separately, hourly. Cleanest for one-off commissions, and it makes revisions a conversation about money rather than a favour.
- Amortise it across an expected run. Reasonable when you own the design and will sell it repeatedly, provided you are honest about the expected quantity.
- Give it away and keep the design. Legitimate if you retain the rights and the customer is paying for parts. Do not give away both the design work and the rights; that is working for nothing twice.
The failure mode is quoting a per-part price that silently includes eight hours of modelling and then being asked for two parts.
Batching changes the answer more than any setting
Notice that the plate above held four parts. Labour, setup and the failure allowance are per-plate costs, not per-part costs, so filling a plate spreads them.
The same commission run as four separate prints would carry four lots of setup labour and four independent chances to fail. If a customer wants ten of something, quote the ten-up price, not ten times the one-off price — and if they want one, say plainly that a single part carries the setup cost of a plate, because it does.
The prices customers actually accept
Three practical observations that are not arithmetic.
Nobody buys a price per gram. Quote a part price. A customer comparing your per-gram figure to a spool's retail price will conclude you are charging four times what the plastic costs, which is true and beside the point.
Turnaround is worth more than material choice. A part on Thursday for a premium beats a part next week, more often than not.
A minimum charge is not rude. Anything under about twenty dollars costs more in correspondence, setup and packing than it returns. State it up front and it will never be an argument.
The check to run before you quote
Compute the cost, apply the margin, and then ask one question: if this job fails twice and I still deliver on time, am I still ahead? If the answer is no, the margin is too thin or the failure allowance is fictional. That single test catches almost every underpriced quote before it is sent.