The "buy versus outsource" question is one of the most consequential capital decisions a small product team, hobbyist designer, or maker-studio runs into. Get it right and you compound a competitive edge — fast iteration cycles, custom material choices, no waiting on shipping. Get it wrong and you've parked $2,000 to $30,000 of capital in a noisy machine that gathers dust between rare jobs. The break-even math below tells you which side of that line you're on.
How the math actually works
The in-house monthly cost of a 3D printer is the sum of three things: amortized capital (printer cost divided by useful life in months), recurring overhead (maintenance, software subscriptions, storage), and variable cost per print (material plus electricity, multiplied by volume). Outsourcing has just two: per-print fee from the service bureau and shipping. The break-even is the print volume at which both totals equal each other.
Failed prints. Beginners typically scrap 10% to 20% of their first 50 prints to nozzle clogs, bed adhesion failures, and slicer settings. Even experienced operators average 3% to 8% scrap on production work.
Consumables beyond filament. Build plates last 6 to 18 months. Nozzles wear out every 200 to 500 hours. Belt tensioners, fans, and PTFE tubing all need replacement on a schedule.
Post-processing time. Most parts need support removal, sanding, painting, or assembly. Budget 15 to 60 minutes per print depending on geometry and finish requirements.
Storage and material handling. Filament absorbs moisture; you need a dehumidifier or sealed dry boxes. Resin handling requires gloves, ventilation, IPA wash baths, and UV curing equipment.
The learning curve. 20 to 40 hours of testing, calibration, slicer profile development, and YouTube watching before you reliably produce good first prints.
FDM, SLA, SLS, MJF — pick the technology that matches your output
FDM (filament): $200 to $4,000 printers. Best for functional prototypes, jigs, fixtures, and large parts. PLA, PETG, ABS, and TPU filaments. Lowest cost per print but visible layer lines.
SLA / DLP (resin): $250 to $5,000 printers. Best for high-detail miniatures, dental, jewelry, and presentation models. Smooth surface, brittle parts, requires washing and UV curing.
SLS (powder, laser): $25,000+ printers. Best for end-use nylon parts, hinges, snap-fits, and small production runs. Almost always outsourced — capital is prohibitive for under-100-parts/month volume.
MJF (Multi Jet Fusion): $200,000+ printers. Best for repeat production of nylon parts at scale. Always outsourced unless your business is contract manufacturing.
Walkthrough: a product design team printing 20 prototypes/month
Printer choice: prosumer FDM (Bambu Lab P1S) at $700
Material cost per print: $3 (average part, PLA at $20/kg)
Electricity per print: $0.40 (4-hour print at $0.10/kWh)
Capital payback: $700 / ($397.56 − ongoing overhead) ≈ 2 months — extremely fast
At this volume and price point, buying is obviously correct. Flip to 3 prints per month and the same math turns negative; the printer sits idle most of the time and amortization plus maintenance exceeds outsourcing cost.
When non-cost factors override the math
Iteration speed. An in-house printer collapses prototype-to-test cycle time from 5 to 10 business days to overnight. For product-development workflows, that compression is worth significantly more than the financial savings.
Confidentiality. Sending CAD files to a service bureau exposes your designs. In-house printing protects IP for unreleased products.
Material exotic-ness. If you need carbon-fiber-reinforced nylon or polycarbonate, only certain bureaus can deliver and at premium pricing. In-house high-temp printers solve this.
Variable demand. If you sometimes need 5 prints, sometimes 100, owning a single machine is more flexible than batching to a bureau's lead times.
Tax treatment for small businesses
In the U.S., 3D printers purchased for business use fall under the IRS de minimis safe harbor if under $2,500 — you can fully expense them in the year of purchase against ordinary business income. Larger machines can be depreciated under MACRS (7-year property class) or fully expensed under Section 179, subject to the annual limit. Hobby printers used solely for personal projects do not qualify; you'd need a documented profit motive and the activity must rise to the level of a trade or business under IRC §183.
Frequently asked questions
What about renting a printer instead?
Equipment rental for 3D printers is rare in the consumer space and limited in the industrial market. A few maker-spaces and university labs offer membership-based access at $30 to $200 per month, which is often cheaper than ownership for occasional users.
Are 3D printing services taxable as sales tax in the U.S.?
Generally yes when delivered as a finished product, though some states treat custom fabrication as a service rather than a sale. Check your state's revenue department guidance; California, Texas, and New York have explicit positions.
How do I price 3D-printed parts for clients?
Standard markups range from 2x to 4x the variable cost (material + machine time at a notional rate of $5 to $15 per hour). Industrial-grade SLS and MJF service bureaus charge based on bounding-box volume; consumer-facing services bill by part complexity and material.
Can I deduct electricity for home-business 3D printing?
If the printer is used exclusively for business and located in a dedicated workspace, yes — typically as a percentage of total household utilities matching the workspace's square footage. Mixed personal/business use requires careful allocation.
What's the resale value of a used 3D printer?
Consumer FDM printers retain about 30% to 50% of original price after 2 years. Resin printers depreciate faster (often 20% to 40% retained). Industrial machines hold value better because of certification, support contracts, and known service histories.
How does this compare to CNC vs outsourcing decisions?
The break-even math is similar but capital is typically higher for CNC, lead times are longer for outsourced machining, and materials are more diverse. The qualitative factors (iteration speed, IP protection, custom alloys) often weigh even heavier on the CNC side.