A year ago I started a print-on-demand store as an experiment. The pitch — that you can build a passive product business by uploading designs to platforms that handle printing, fulfillment, and shipping — has been all over the side-hustle internet for half a decade. I wanted to know if the numbers held up when an actual person ran the actual thing for an actual year. They mostly did not, and the way they did not is the interesting part.
The setup, briefly
Three platforms, picked because they did not require me to hold inventory: Redbubble, Society6, and Merch by Amazon (after a four-month wait for account approval). I licensed three sets of stock illustrations and modified them — within the licenses — into 2,840 design variants over the year. The mix was the usual: typography quotes, line-art animals, niche-hobby visuals (hiking, knitting, board games), and seasonal designs.
The store had no separate website. I posted designs to a small Instagram account that grew to 3,400 followers over the year — most of them through one viral post about Hokkaido in February — and did $0 in paid advertising for the first eight months.
Revenue, by month
| Month | Gross sales | Platform royalty | Notes |
|---|---|---|---|
| May 2024 | $220 | $68 | First month, mostly friends |
| June 2024 | $380 | $118 | |
| July 2024 | $510 | $158 | |
| Aug 2024 | $640 | $198 | Merch approval came through |
| Sep 2024 | $1,210 | $372 | Back-to-school traffic |
| Oct 2024 | $1,440 | $436 | Halloween designs |
| Nov 2024 | $2,130 | $644 | Black Friday week, biggest single week |
| Dec 2024 | $2,840 | $854 | Holiday gift buying |
| Jan 2025 | $680 | $208 | Steep post-holiday drop |
| Feb 2025 | $540 | $166 | |
| Mar 2025 | $340 | $104 | Tried $150 in paid ads, broke even |
| Apr 2025 | $330 | $101 | |
| Total | $11,260 | $3,427 |
Two things are visible in this table that the side-hustle blogs do not mention. First, holiday Q4 was 65% of the year's revenue. The "passive" income, in months when there is no gift-buying surge, was meaningfully below what gets sold as "achievable." Second, the platform royalty rate was 28-32% — what is paid out to the designer after the platform takes its share. The remaining 70% is the platform's cut, not just printing and shipping. POD is a thin-margin business for the seller, by design.
The costs I underestimated
Royalties paid out are not the same as profit. Real costs against the $3,427 royalty figure:
| Category | Annual cost |
|---|---|
| Stock illustration licenses (3 packs) | $420 |
| Design software (Affinity Designer one-time + Canva Pro) | $220 |
| Mockup/template software (Placeit) | $168 |
| Stock font licenses (two professional purchases) | $95 |
| Paid ads experiments (Pinterest, Meta) | $340 |
| Niche keyword research tool (3 months) | $87 |
| Bookkeeping software (Wave was free; QuickBooks not) | $0 |
| Schedule C tax-prep portion (estimated) | $95 |
| Total expenses | $1,425 |
Then there was the time. I logged hours weekly in a simple sheet. Total time, including design work, uploads, platform admin, customer service emails (rare but real), and Instagram posting: 412 hours over the year. That averages 7.9 hours a week, with very uneven distribution — five-hour weeks early on, fifteen-hour weeks in October and November.
Net profit
$3,427 in royalties minus $1,425 in expenses = $2,002 in pre-tax profit. After self-employment tax (15.3% on net SE income), that becomes about $1,696. After federal income tax at my marginal rate, about $1,375. State tax took another $90.
Net after-tax profit: $1,285.
Divided by 412 hours, that is $3.12 per hour. The federal minimum wage is $7.25. I made less than half of minimum wage running a side business that the genre of motivational YouTube videos describes as "set it and forget it" passive income.
A side hustle is not a side hustle if it pays less per hour than a job you already have. It is an expensive hobby that uses entrepreneurship language.
What I would do differently
Three things, in case anyone is reading this and considering trying it themselves.
One: do not chase volume. Of the 2,840 designs, about 60 produced 80% of the revenue. The other 2,780 were noise that took time to make and never sold. A focused store with 200 carefully iterated designs in two niches would have done the same revenue with a quarter of the work.
Two: niche down, not across. My niches were too broad. "Hiking" sold less than "Pacific Crest Trail thru-hiking." "Knitting" sold less than "knitting with cats." Specificity, even at the cost of audience size, won every time. The deepest niche I had — board-game-specific quote designs aimed at one game — did $480 in revenue from 18 designs, the highest revenue-per-design in the store.
Three: do not bother with paid ads on a POD store. I lost money on every dollar I spent. The unit economics — 30% margin on a $25 average sale, which is roughly $7.50 of royalty per unit — cannot absorb a $4 customer acquisition cost without obliterating profit. POD as a business is not built for paid acquisition; it is built for organic discovery on the platform itself.
Would I do it again? Probably not at this scale. The interesting version, if I were starting now, would be a very narrow store — one niche, one platform, 50–80 designs, three hours a week. The economics on $3/hour scale; the economics on $30/hour might. That is a separate experiment I have not yet run.
