- Migrate in stages: start with one to three products that sell steadily and have settled artwork.
- Artwork built for a mockup often needs vector files, fewer colors, and exact placement for a real press.
- Always see a physical or photographic proof, and wash test one sample before approving the run.
- Size the first order to about two months of sales and use a preorder window if demand is uncertain.
- Keep on-demand live as a fallback until the new setup has proven itself over 60 to 90 days.
Before you move anything
Leaving print-on-demand is a project, and projects fail when they start with a purchase order. Start with a week of looking at your own shop. You are trying to answer three questions: which products are worth moving, what customers have actually been receiving, and what you will need to run the new setup.
Export the last six to twelve months of orders. Group by product and design. For each, note units sold, revenue, refunds, and the blank and print method. You will usually find that a small share of designs produce most of the money. Those are the candidates.
Then read your own reviews and support threads. Customers describe the product more honestly than any dashboard does. If four people mentioned a print that cracked or a color that looked off, you have learned something about what to change in the next version, not just where to make it.
Choosing what to move first
Move a small number of products, and move the right ones. A good first candidate has four traits. It sells steadily, at least 25 to 30 units a month. The design is settled, with no pending tweaks. The artwork is well suited to a traditional method, which usually means up to four colors and clean edges. And the margin gap between on-demand and a run is large enough to matter.
A poor first candidate is a seasonal item, a design you changed last month, or anything with a photographic gradient that screen printing will struggle to hold. Leave those for later.
Resist the urge to move the entire catalog. A staged migration, one to three products at first, keeps your fallback intact and gives you real data from the new setup before you expand.
Picking a print partner
The questions you ask a production partner for a run are different from the ones you asked an on-demand app. Here is a checklist that applies whether you choose a trade printer, a local shop, or an overseas factory.
Ask about the proof process. Will you see a physical proof or a digital one, and who approves it? Ask about pricing structure: what is the setup, the per unit price at your quantity, and the next break up? Ask what happens on a reorder. Ask for production and shipping times in business days, and what rush costs. Ask whether they can ship under your name with no outside branding. Ask how they handle defects: reprint, credit, or refund, and within what window. Ask what artwork formats they need.
Our guide on how to choose a trade printer goes through these questions in more depth. And if you plan to sell to clients or an audience as a business, the comparison of printers, brokers, and resellers explains the differences between supply models so you know who you are buying from.
Preparing artwork for a real press
Artwork that looked fine on an on-demand mockup may not be ready for production. A few common fixes:
Vector files. Logos and line art should be supplied as vector files, not screenshots. A press cannot make a sharp edge from a low resolution image.
Spot colors. For screen printing, you will specify ink colors, usually by a standard color matching system, instead of relying on how a color looks on a monitor. Pick them from a physical swatch book if you can.
Color count. Each color usually costs a screen. If a design uses seven colors because that is what the digital print allowed, ask whether it can be simplified to three or four. Simplified artwork often looks stronger on a shirt and costs far less.
Placement and size. Define exact dimensions and position relative to the collar, seam, or edge. Provide those in writing and on a template. Vague placement is the leading cause of a run that looks different from what the creator imagined.
Halftones and gradients. Gradients can be printed with halftone techniques, but they behave differently from digital. If your design relies on them, ask the printer to show you a test on the actual blank.
Proofing: the step you cannot skip
The main thing you gain by leaving on-demand is the ability to inspect before you commit. Use it.
Ask for a physical sample or a first article before the full run starts. If that is not possible, ask for photos of the first pieces off the press under neutral light. Check four things: color against your swatch, placement and scale, print quality at the edges and in fine details, and the blank itself, including weight, fit, and fabric hand.
Test a wash. Wash and dry one sample according to the care label, then check for cracking, fading, or shrinkage. A single wash test can save you from a run of 250 that fails by the third wear.
For print products such as stickers and cards, a digital proof is the standard first step, and a physical proof is worth requesting when color is critical. Our proofer demo shows how a trade proofing workflow looks in practice.
Sizing your first order
Use your sales history for this. Take the average monthly units of the product and multiply by two for the quantity you want to hold. If you sell 40 a month, a first order of 80 to 100 is appropriate. If the number is uncertain, run a preorder window and order exactly what sold, plus a small buffer of 10 to 15 percent for exchanges and walk-ups.
For sizes, review your own distribution rather than splitting evenly. A typical unisex apparel distribution has medium and large taking about half the units, small and extra large taking about a third, and the extremes the remainder. Your audience may skew, so look at the data.
Resist the price break at the next tier unless you are confident. A price that drops 12 percent at double the quantity does not help if half the order sits unsold for a year. Ask whether the printer will let you split the quantity across two shipments, or hold the second half under the original price terms. We cover why these breaks exist in our guide to minimums and setup fees.
Receiving, storing, and shipping
Inventory is a physical problem. Decide before the boxes arrive where they will live. A dry, clean shelf in a spare room is enough for 250 shirts. Plan for labels, bags, and mailers. Count and check the delivery against the packing list the day it arrives, and report any shortages or defects while the window for claims is open.
Fulfillment can stay in-house at low volume. At 100 or more orders a month, pick-and-pack can eat several hours a week, and a third party fulfillment partner starts to make sense. Compare their per order fee, storage fee, and receiving fee against your own time. Most Shopify apps for fulfillment connect to a store with a few clicks, so the technical side is rarely the barrier.
Think about the parcel. This is the part of the migration customers will actually notice. A branded mailer, a thank you card, a sticker, tissue, and a clear packing slip all cost pennies at volume and change how the brand feels. If you ship from your own stock, you can add a sticker pack for 40 cents that would cost $3 on-demand.
Updating the store
The storefront needs work too. Products moving from on-demand to inventory change in a few ways that customers will see. Delivery estimates drop, so update the product page and the shipping policy to say so. Stock counts need to be real, and an inventory tracking setting should be on, so you cannot oversell a limited run. Photography should be updated to show the actual product, since you now own one and can take real photos instead of mockups. Real photographs lift conversion on nearly every shop that makes the switch.
Keep both versions live during the transition. Create the run product as a new listing, or add a variant, rather than editing the existing one in place. Redirect or hide the old listing once inventory is available, and keep an on-demand fallback for sizes you ran out of.
Telling your customers
You do not need an announcement about operations. You need a message about the product. If the new run is better, say so: a heavier blank, a softer print, a faster ship time. Customers who have bought before will notice and appreciate the upgrade, particularly if you tell them in a newsletter or a note in the parcel.
If you are making a limited run, make that the headline. "Two hundred made, first access this Friday" is a marketing event that on-demand never gave you.
A realistic four week plan
Week one. Export data, pick one to three products, request quotes from two or three print partners at three quantities, and order samples of the blank.
Week two. Choose a partner, finalize artwork, approve a digital proof, and request a physical sample if possible. Open a preorder window if you want one.
Week three. Approve the first article, place the order, and set up the store listing, inventory tracking, packaging, and shipping settings.
Week four and beyond. Production and delivery typically take two to three weeks, so inventory arrives by the end of week five or six. Our guide to production timelines can help you plan around peak season. When stock arrives, check it, swap the listing live, and monitor orders closely for a month.
What to measure afterward
Compare the migrated products to their on-demand history over 60 to 90 days. Track margin per unit, units sold, refund rate, repeat purchase rate, and review score. Track the time you spend packing. If margin rose, refunds fell, and the time cost is reasonable, move the next product. If one metric disappointed, find out why before expanding.
Keep a simple sheet: product, units sold, cost, revenue, refunds, and reorder date. Reordering at the right time matters, because a stockout on your best seller costs more than the inventory ever saved you.
What the migration costs
Plan a budget before you start so nothing surprises you in week three. For a first move of two apparel products, a realistic outline looks like this. Inventory for two products at about 100 pieces each at $7 to $9 per piece is $1,400 to $1,800. Samples and a proof round add $50 to $150. Packaging supplies, including mailers, tissue, thank you cards, and stickers, run $80 to $200 at that volume. Photography for the new listings is a few hours of your time, or $100 to $300 if you hire it. A storage box and a label printer, if you do not already have them, add another $50 to $100. The total sits between roughly $1,700 and $2,600.
That money comes back through margin. At a saving of $5 to $7 per unit on 200 units, the recovery is $1,000 to $1,400 on the first pass alone, plus the value of better packaging and faster shipping. Most creators who migrate their top sellers reach break even within two to four months.
Returns and exchanges when you hold stock
On-demand shops often handle a wrong size by reprinting and shipping a new unit. With your own stock, an exchange can use inventory you already have, which is faster and cheaper, but it means you need a process. Decide what you will accept back, in what condition, and within how many days. Write the policy on the product page and the packing slip. Keep a shelf for returned items, check them, and restock the good ones.
Defects deserve a separate rule. When a piece arrives with a flaw, replace it and photograph the issue, then send the photo to your printer. A supplier who stands behind the work will credit or reprint, and the paper trail makes that conversation short. Ask about the defect policy before you place the order, not after.
When not to migrate
Some shops should wait. If the catalog changes every month, if no design sells more than 15 units a month, if you have less than a couple thousand dollars you can afford to tie up, or if you are about to rebrand and the artwork will change, hold off. The cost of being wrong is real, and on-demand does its job well when the business is still finding its footing. Revisit the numbers every quarter. The day a design crosses your threshold, move that one.
If you are building a brand around the merch and not only selling it, our brand launch playbook is a useful companion, because settled artwork and a settled identity make a migration much smoother.
Frequently asked
How long does it take to move a product off print-on-demand?
Plan on four to six weeks from the first quote to stock in hand: about a week to choose products and collect quotes, a week for artwork and proof, and two to three weeks for production and delivery.
What does a first migration cost?
For two apparel products at roughly 100 pieces each, expect about $1,700 to $2,600 including inventory, samples, packaging, and photography. That figure varies with the garment, print method, and how much you do yourself.
Do I need to change my artwork?
Often, yes. Logos should be vector files, colors should be specified for the press, and designs with many colors usually print better and cost less when simplified. Photographic gradients may need halftones or a different method.
Should I use a fulfillment partner or ship myself?
Ship yourself at low volume. Once you are packing roughly 100 orders a month, compare a fulfillment partner per order, storage, and receiving fees against the hours you spend. Many Shopify apps connect a fulfillment partner to a store in a few clicks.
What if the new product does not sell?
Limit the risk by sizing the run to two months of sales, choosing designs with a track record, and using preorders. Keep a plan for leftovers, such as bundles or giveaways, and revisit the numbers every quarter.
Ready to move your best sellers?
Apply for the Trade Print program to get published pricing, proofing, and blind shipping on your first run.
Apply to the Trade Print program