How Much to Charge for 3D Prints

There is no single right number, but there is a method that holds up in 2026: add up what the print actually costs you — material, power, machine wear and your time — adjust for the prints that fail, and then at least double it. Small, repeatable prints usually end up priced near 4–10× raw material cost; labour-heavy one-offs need the cost-plus floor instead. This guide walks through the full formula with a worked example, the rules of thumb experienced makers actually use, and the marketplace fees that quietly eat 10–35% of your revenue.

What a 3D print actually costs (five components)

Most underpricing comes from counting only filament. A print's honest cost has five parts, and three of them are almost always skipped:

You don't have to do this by hand: the 3D printing cost calculator models all five, and the filament cost calculator converts your actual spool price into cost per gram and per print. Each term's math — including 2026 filament price benchmarks and the failure adjustment — is expanded in the 3D printing cost formula guide.

Two pricing methods that (helpfully) agree

Method A — cost-plus with a markup

Compute the true cost per good part, then apply a markup. 2× cost is a bare-minimum resale; maker sellers of small consumer prints commonly quote 2.5–4×; the "10× material" rule that circulates everywhere is really a proxy for this on prints where material dominates — and it fails exactly when your time dominates.

Method B — price per print-hour

The other community shorthand: charge roughly $2–4 for every hour the machine runs, plus material, plus your hands-on time on top. It bundles power, wear and profit into one number so you stop itemising cents. It only works if labour is billed separately — the mistake is reading "$3/hour" as including your slicing time.

Worked example: an 8-hour articulated dragon

Method A says: at 1.5–2× cost, sell at $34–45. Method B says: 8 × $3 + material ≈ $25.60 — and that's before paying you for the hour of hands-on work, so anything under roughly $25 is subsidising the buyer. Notice what the example proves: labour, not filament, is the largest cost — which is exactly why "10× material" ($16 here) would have lost you money.

The failure rate you're not charging for

Spaghetti towers are inventory you threw away. If one print in ten fails, every successful print must carry part of the failure's cost: divide the material + machine portion by (1 − failure rate). At 10% that's a 11% surcharge on those lines — small on filament, significant on long prints and resin batches. Well-tuned machines drift below 10%; first prints of a new file or filament often run above it. Estimate honestly in the cost calculator rather than assuming zero, and re-check it quarterly.

Selling on Etsy or other marketplaces: fees come off the top

A price that works in your workshop can quietly lose money online. As of October 2026, on Etsy's published fee schedule, a US listing costs 6.5% of the item price (including shipping you charge) plus $0.20 per listing, plus payment processing of roughly 3% + $0.25 per order — so real-world fees on a small order land near 15–20% of the sale.

STL marketplaces take larger shares. Cults3D's standard cut is about 15% (much higher royalties if you join its exclusivity tiers), and platforms like MyMiniFactory and CGTrader commonly take around 30% of each sale — always re-check the current schedule before listing, because these rates move and many guides quote years-old numbers. The pricing math changes: use the 3D model pricing calculator to build a marketplace commission into your model price, and the Etsy fee calculator (or PayPal fee calculator for direct sales) to work backwards from what you want to keep.

The one-line rule: gross up. To keep $30 after a platform taking ~20%, your list price is 30 ÷ 0.80 = $37.50, not $36.

Should you charge per hour instead?

Per-hour pricing protects you on exactly the jobs cost-plus underprices: long prints, multi-colour prints, large-format machines and custom work. The community's per-print-hour band ($2–4) plus a labour rate (≥ $20/h hands-on, and design and file-repair work quoted around $75–150/h by freelance 3D artists) covers most cases. Get an honest print time first — slicer estimates are optimistic, so estimate the print time and add 10–20% before it goes in a quote. If you sell custom modelling rather than prints, the modeling hourly-rate calculator does that gross-up for you.

Five pricing mistakes that kill maker margins

1. Charging materials ×10 without clocking your time. Ten times filament pays you nothing when an hour of prep is in the job.
2. Treating slicer estimates as truth. They are usually 10–20% optimistic; electricity and depreciation follow the real hours.
3. Ignoring failed prints. A 10% failure rate is ~11% of your material and machine cost if it isn't priced in.
4. Listing at "what others charge". Competitors may be mispriced, dumping, or not counting labour — benchmark demand, price your own math.
5. Forgetting the fee stack. Marketplace + payment + shipping materials compound; gross-up from the amount you actually want to keep.

Sanity checks before you publish a price

Three questions do most of the work. Does it clear cost + my labour? If no, you are buying your customer a hobby. Is it inside the band buyers pay for comparable finished size and quality? If yes, you have room to test upward. Does the profit per printer-hour justify this job over the alternatives? On a shared machine, that ranking is your product mix. Track it with the 3D print profit calculator, and if you are planning a small print farm, the same inputs scaled up answer the break-even question — maker-business reports typically put it at roughly months 6–14 after a $750–2,800 start-up.

Frequently asked questions

Sources & verification. Marketplace and payment fees were checked against platforms' published schedules (Etsy, PayPal — see the linked calculators' source lines) on 1 October 2026. The pricing rules of thumb above are the community consensus repeated across r/3Dprinting pricing threads and maker-business write-ups such as grandpacad's "boring math". Individual results vary with machine, material and market — treat every figure as an input to your own math, not a substitute for it.