Is print on demand actually worth it for a small seller

Somebody sold you the pitch already. No stock, no upfront cost, upload a design and the printer handles the rest. That part is all true, which is what makes the rest of it hard to see.

Print on demand is worth it in one situation and not really in any other. If you already have an audience that will buy because it's yours, it's a sensible way to sell them merch without $2,000 of inventory in your hallway. If you don't have an audience and you're planning to find buyers with paid ads, the margin doesn't survive the customer acquisition cost, and I'd rather you knew that before you spent $240 finding out. A $28 shirt leaves you around $7 after the printer takes their cut. There is no version of that number that supports a $12 cost per acquisition.

What follows is the arithmetic, the things you can't control, and the alternative that most people don't consider.

The margin, written out

Prices move, so check today's rates with your own printer and your own blank. This is roughly where a mid-weight cotton tee with a single front print lands for a small seller in the US in 2026.

Line	Amount
Retail price	$28.00
Shipping charged to customer	$5.95
Collected	$33.95
Blank garment	$9.50
Printing	$5.00
Fulfilment shipping paid to printer	$5.95
Card processing at 2.9% plus 30c	$1.28
Sailo card commission, 1–3% of the goods	$0.14
Costs	$21.87
Net	$12.08

That looks decent. Now apply reality.

Somewhere between 3% and 8% of orders need a reprint or a refund, depending on the garment and how well you've handled sizing. Call it 5%, and half of those are your cost rather than the printer's, because the printer reprints their errors and not the customer's wrong size. That's about 60 cents off every order.

Then take off your time. Designing, mocking up, listing, writing the post, answering "what size am I", and dealing with the one that arrives with the print off-centre. Even at a brisk pace that's 20 to 30 minutes per design plus a few minutes per order.

So the honest figure is around $11 an order, and roughly a third of it is eaten if a single customer needs handling.

Now add advertising, which is where the model actually breaks. Say a click costs 80 cents and one in sixty clicks buys. That's $48 to acquire a customer who nets you $11. There are people who make paid traffic work in this category, and they do it with high average order values, back-end products, or margins you don't have on a single shirt. At $28 with a $12 net, paid acquisition is a subsidy you're paying to a platform.

None of this is a reason not to do it. It's a reason to be clear about what it is: a way to convert existing attention into a bit of money, not a way to manufacture attention.

What you don't control, and what that costs

The pitch says the printer handles it. That's exactly the problem, because "it" includes everything the customer will judge you on.

Print quality varies by garment colour. The same file prints differently on white and on black. On dark garments, direct-to-garment printing needs a pretreatment, and if it's applied sloppily the customer sees a faint square around the design after the first wash. You will not see this in the mockup. You will see it if you order a sample in black.

Colour shifts. The teal on your screen is not the teal on the shirt. Order the sample.

Blanks go out of stock. In the size that sells most, usually, in the week you post about it. Some printers substitute a different brand silently, and a customer who bought a second shirt six weeks later gets a different fit and thinks you cheated them.

Production time is before shipping, not included in it. Two to seven business days in the facility, then transit. A customer who ordered on the 14th for a birthday on the 22nd is not going to make it, and they'll be messaging you, not the printer.

Durability is not your call. Print cracking after ten washes is a manufacturing outcome you had no part in and will be blamed for entirely.

The parcel comes from somewhere else. Different country sometimes, different sender name, no note from you, no tissue paper, nothing that says the thing arrived from a person. For merch built on a personal following, that gap between the post and the parcel is a real cost that nobody puts on a spreadsheet.

You own the complaint and not the process. That's the trade, stated plainly. Every message about a crooked print, a slow delivery or a shirt that feels thinner than expected arrives in your inbox, and your only tools are apologising and eating the cost.

Order the sample, in the worst colour

If you take one operational instruction from this article, take this.

Before you list a design, order it. In black and in white, because those two fail in opposite ways. Then wash it five times at the temperature normal people use, not the temperature on the care label, and look at it in daylight.

You'll learn:

• Whether the pretreatment square shows.
• How much the print cracks along a fold.
• Whether your dark navy prints as dark navy or as a slightly dusty blue.
• Whether the "unisex medium" is a medium that any actual person would call a medium.
• What the packaging looks like when it lands, which is what your customer's first impression is.

Cost of that exercise: about $30 and ten days. Cost of skipping it: your first twenty customers finding out for you.

The same logic applies to sizing. Publish the flat measurements from the garment brand's own chart, in centimetres, chest and length, and photograph the shirt flat next to a tape measure if you can. Sizing returns on POD come out of your pocket entirely, which makes measurement the cheapest insurance in the business. If you're selling in multiple sizes and colours, how to handle sizes and variants is worth reading before you build the listings rather than after.

Where print on demand genuinely wins

I've been sceptical for eight hundred words, so here's the honest other side. There are four situations where it's clearly the right call.

Merch for an audience that already exists. A podcast with 3,000 listeners, a running club, a small band, a creator whose followers want the in-joke on a shirt. These people are buying belonging, not apparel, and the margin is fine because the acquisition cost is zero.

Testing a design before you commit. Print five, see if anybody cares, then order a screen-printed run of the one that worked. This is the smartest use of POD and almost nobody frames it this way.

One-off and event runs. A memorial shirt, a reunion, a fundraiser. Twenty units, no minimum, nobody stuck with a box.

Products where the blank is cheap relative to retail. Posters and art prints carry a much better margin than apparel, because the substrate costs almost nothing and the perceived value is in the image. If you're an illustrator, printing is a far better fit for prints than for tees.

Two things to steer clear of, hard. Don't print anyone else's lyrics, characters, team names or logos, however much they'd sell. Your printer will terminate your account and every platform you list on will too, and it takes one complaint. And don't build a shop of forty generic "aesthetic" designs hoping one takes off, because that's a lottery with a listing fee and the winners in that game have hundreds of designs and ad budgets.

The alternative nobody prices out

A pre-order run. Take the orders first, print second.

Screen printing 50 shirts at a local printer runs roughly $8.50 a unit all in for a one-colour print on a decent blank, sometimes less. Sell them at $30.

	Print on demand	Pre-order run of 50
Upfront cash	$0	$425
Cost per unit	$14.50	$8.50
Retail	$28	$30
Net per unit before shipping	$13.50	$21.50
Who packs it	The printer	You, at your kitchen table
Delivery time	5 to 14 days	Whenever you get to the post office
Risk	Almost none	You could be left with 12 shirts
Print quality control	None	You see them before they go out

The pre-order run makes 60% more per shirt and puts a note in the box. It also means an evening of packing, a trip to the post office, and a real possibility of leftovers. On 50 shirts that's $1,075 versus $675, and the extra $400 is payment for your Saturday and your nerve.

The way to take most of the risk out is to open the pre-order with a deadline and a minimum, and only place the print order if you hit it. Say it plainly on the page: "Pre-order closes on the 20th. Shirts ship the first week of next month." People are used to this now and they're fine with it. Building a waiting list for a drop covers how to get to the minimum before you open.

If you'd rather someone else handled everything and you'd rather not think about margin, a marketplace with a built-in audience might beat both. That's a genuine option and when a marketplace beats your own shop argues it honestly.

Worked example: Ryan, Portland, 1,400 followers

Ryan draws small illustrations of local landmarks and had about 1,400 people following him when he decided to sell shirts.

He spent roughly 30 hours on the first launch: eight designs, mockups, product photography he had to fake because he had no product, and a launch week of posting.

Launch week. 38 shirts sold at $28, plus shipping collected. Revenue $1,290. Printer and processing costs $831. Two refunds, one wrong size and one that never arrived, cost him $58. Net $401 for 30 hours, so about $13 an hour, and he was pleased with it because it was his first go.

The three months after. Four to nine shirts a month from the same audience, netting maybe $80 a month. The audience that wanted a shirt bought one.

The ads month. He put $240 into paid social across three weeks. Six sales. $72 of net margin against $240 of spend, so a $168 loss, which he described afterwards as tuition.

What he did next, and it worked. He picked the single best-selling design, opened a pre-order for a screen-printed run at $32 with a minimum of 40 and a two-week window, and hit 61. Print cost $9.10 a unit at that quantity. He packed them over two evenings with a friend and a case of beer, put a small sticker in each one, and cleared $1,190 net. Same audience. Same design. Nearly three times the money.

His conclusion, which I'd endorse: print on demand was a good way to find out which design people wanted. It was a bad way to sell it once he knew.

The honest limitation, and the shop side

Whatever you print, a merch shop grows listings fast. Eight designs is eight listings before you decide whether a different colourway gets its own page, and the Sailo free plan stops at 10 products. Pro at $19 a month takes it to 100 and Business at $49 makes it unlimited, and you'll want the Business plan anyway if you're taking card, which for merch you almost certainly are.

The division of labour is worth stating plainly, because it's the whole risk in this model. Sailo holds the catalogue, takes the order and puts the money in your own account. Your printer prints the shirt and posts it. Nobody in that chain is the person the customer messages when it arrives crooked, except you. On bank transfer, cash, WhatsApp, Instagram DM, Telegram, email and phone orders Sailo takes nothing at all and never touches the money. On card it's 1–3% of the goods, after discount and excluding delivery and tax, on top of your Stripe fee, and card needs a Stripe account cleared for charges. On a $28 shirt that's 14 cents to Sailo and about $1.11 to Stripe, which tells you where to look if you're trying to save money on fees. What a commission actually costs you does the same comparison across marketplaces.

One more thing worth budgeting: the shipping you charge and the shipping you pay are two different numbers and they drift apart when someone orders two shirts. How to work out delivery charges covers how to set a flat rate that doesn't quietly lose money on multi-item orders.

What to do next

Order one sample of your best design, in black, and wash it five times before you list anything. That's $30 and it's the difference between finding out about the pretreatment square now and finding out from a customer.

While it's in the post, work out your own version of the margin table above with your printer's actual current prices. Not the ones in this article, yours, today. If the net is under about $8 a unit, the product is too cheap or the blank is too expensive, and no amount of marketing fixes either.

Then be honest with yourself about the audience question. If you have people who'd buy because it's yours, print on demand is a reasonable way to serve them and you should go ahead. If you don't, build the audience first and sell them something later, because the shop was never the hard part.