How to Make T-Shirt Mockups Without Photoshop (The Real POD Workflow)

Most guides tell you to drag your design onto a stock template and call it a mockup. That is not how print-on-demand sellers who actually move product do it, and none of it needs Photoshop. Here is the real workflow: build your own mockups from the blank you actually sell, test them on Meta to find the one that gets cheap clicks, lock that winner in across your store, and only then start testing designs.

Why the usual "mockup" advice is wrong for real sellers

Stock lifestyle mockups, the model-in-a-cafe shots from a generator, are great for a one-off ad or a landing page. They are the wrong tool for your product images. Serious t-shirt sellers do not scatter different lifestyle scenes across their store, because buyers should be comparing your designs, not the models or the backgrounds. So the goal is not one pretty mockup. It is one proven mockup style you can put on every product, chosen because it earns cheap clicks. That takes a real, repeatable process, and the tools are free.

Step 1: Start with the real blank you sell

Skip the generic stock shirt. Start from the actual blank you stock, a Comfort Colors 1717, a Gildan 64000, or whatever your niche runs. You want a clean photo of that exact garment to build on.

The fastest source is a marketplace like Etsy or Creative Fabrica, where sellers offer photographed blanks and mockup images of these exact styles, usually cheap and ready to drop a design onto. Buying the real blank means your mockup shows the true garment a customer receives, which no generic template can promise.

Step 2: Build the mockup (you do not need Photoshop)

Photoshop is just one tool here, and not a required one. Plenty of sellers prefer Affinity; Photopea is a free, near-identical stand-in; Canva covers the basics. Use whatever you like, because the tool is not the point. What you are actually chasing is first-inspection realism: a shirt that reads as genuinely real in the half-second before someone scrolls past. That is honestly hard to nail with traditional editors.

This is where AI image tools have leapt ahead. ChatGPT's image tools, Google's Gemini (Nano Banana), Ideogram, and MysticPOD (an AI image tool built specifically for POD sellers) can now turn out mockups with a realism that used to take serious retouching. They are tools too, not magic, but they have closed the realism gap fast, and many POD sellers lean on them now.

That opens up a few routes sellers actually use: start from a real blank you bought and composite your scene onto it; use one blank and recolor it to match every variant color you sell; or have an AI tool produce a realistic mockup, then change only the shirt color for each variant. All three are legitimate, and the choice matters less than one rule you will see in a moment.

Whichever tool you use for the base image, the classic composited approach still works and needs no Photoshop. In a browser tool like Canva you build a mockup from three layers:

1. A niche-matching background. The scene should signal your niche in a glance: weathered wood for a rustic brand, brick for streetwear, snow for a winter drop. The background does a surprising amount of the selling.
2. The blank shirt, placed on that background. Worn on a model, laid flat, or floating; each reads differently to a buyer.
3. One or two niche props around the shirt to complete the scene, without cluttering it.

Now make several variations, and change only one thing at a time: swap the background but keep the shirt and props, or change the props but keep the background. If you change five things at once and one performs, you will not know which change did it. One variable per version is the whole discipline.

What separates a good test mockup from a merely pretty one

A few principles make the difference, whichever tool you build in:

• The shirt is the product; the scene is the frame. Keep the shirt the obvious star, roughly three-quarters of the visual weight. Props and backgrounds support it; the moment a viewer wonders whether they are buying the backpack, you have lost.
• Niche-coherent, not random. Every element should say the niche at a glance. For a hiking store that means trail, wood, boots, a compass, so the shopper thinks "that's my world." Test different concepts that are all clearly the same niche, not twenty unrelated photos.
• Start with one style. A clean flatlay is a reliable starting point. Add model and lifestyle shots later, once you have a winner, so you are not testing ten things at once.
• Square, consistent, and big enough. Build every image square at 2048 x 2048 px, the size Shopify recommends for product images and the size PrntFlow outputs by default. Meta pulls your product images from Shopify for catalog ads and downsizes as needed, so a larger 2048 source always looks right, while Meta's 1024 x 1024 catalog minimum is a floor, not a target. Keep every test image the same square size; do not mix aspect ratios.
• Quality-check every image, especially AI ones. Reject warped folds, malformed props, distorted graphics, or unreadable text. A single obvious artifact becomes an unwanted variable that muddies your data. "It is only a test" is not an excuse.

Step 3: Test your mockups on Meta to find the cheap-click winner

Here is what separates sellers who guess from sellers who know. Before you ever test designs, you test mockups, because the mockup is your ad creative and your product image at the same time. If the mockup itself pulls expensive clicks, you can never tell whether a design is good or just badly framed.

Set up a small, controlled test in Meta Ads Manager. Run a handful of your mockup variations (around six to eight) head to head: same shirt color, same graphic, same ad copy, only the mockup differs. Keep the budget deliberately small, roughly $25 for the cycle (about $12.50 a day for two days). You are not buying sales here; you are buying information.

Optimize the test for purchases or conversions, not just engagement, so you are measuring people likely to buy rather than people who merely like to scroll. The target you are hunting: a mockup that produces a cost per click under $1, with enough spend behind it to trust the number.

Step 4: Read the results in the right order

When the cycle finishes, go to the Ads level and sort by Amount Spent, highest first, before you look at anything else. Meta votes with your money: if its algorithm pushed most of the budget to one mockup, its system saw something promising in that creative. Then look at CPC.

The order matters because a great-looking CPC on almost no spend means nothing:

A mockup showing a $0.71 CPC after spending only $0.71 has not proven anything. A mockup that took real spend but sits at $4-plus CPC caught Meta's interest without converting cheaply. The one you want took at least $5 of spend and still held a sub-$1 CPC (a sale during the test counts as a win too).

Step 5: No winner on the first try? That is normal

Most sellers do not get a sub-$1 CPC on their first cycle. That is not failure; it is the process working. Do not panic and spin up a random new batch. Instead, study your best performer and ask why it did better:

• What was the background doing? Wood, brick, snow, solid color?
• Was the shirt worn, laid flat, or floating?
• Which props were in frame, and how many?

Then build your next batch around what worked. If the wood-surface mockup pulled the most spend, keep the wood and change the background behind it, or keep the background and swap the props. Run another $25 cycle. Across a few rounds you will have tested roughly fifteen to twenty mockups and closed in on a real winner, moving toward the target on purpose instead of by luck.

Step 6: Lock in the winner and roll it out across your store

Once you have a validated sub-$1 CPC mockup, stop changing your product imagery. This winning mockup style becomes the consistent foundation for your whole store: the same proven frame on every product, so buyers focus on your designs and your Meta catalog ads (which pull your Shopify product images) stay predictable.

Rolling that one style across a real catalog is the hidden mountain of work. A typical POD product carries 5 to 8 colors, and a store built for paid traffic runs 100-plus products, which is 500 to 800 variant images to composite, upload, and assign by hand, at roughly 20 minutes per product. This is exactly the grind our guide on assigning images to Shopify product variants walks through, and where a tool earns its keep.

Step 7: Now, finally, test your designs

With a proven mockup locked in everywhere, you have a clean testing environment at last. The mockup variable is fixed, so when you now test designs, any difference in performance is the design, not the frame around it. That is the whole reason you tested mockups first: to earn a foundation you can trust before the testing that actually grows the store.

Where PrntFlow fits (and where it does not)

Full disclosure: this is our app, so judge accordingly. PrntFlow does not make your mockups and it does not run your tests. You build the mockups and you find your winner on Meta, exactly as above. What PrntFlow removes is Step 6, the rollout: once you have your winning mockup style, it composites your design onto every color, uploads the images to Shopify, and assigns each one to the right color variant, across your whole catalog, in a fraction of the manual time. Set the placement once, reuse it on every product, and rerun it whenever you refresh. It works whether you print with Printify, Printful, or Gelato.

FAQ

Do I need Photoshop to make t-shirt mockups?

No. A real blank plus a free browser tool like Canva is enough to build a background, place the shirt, and add props. Photoshop is optional, not required.

Where do POD sellers get their t-shirt blanks and mockups?

Marketplaces like Etsy and Creative Fabrica sell photographed blanks and mockup images of the exact styles sellers stock, such as the Comfort Colors 1717 and Gildan 64000. Start from the blank you actually sell.

Why test mockups before testing designs?

Because the mockup is both your ad creative and your product image. If the mockup pulls expensive clicks, you cannot tell whether a design is good or just poorly framed. A cheap-click mockup gives you a clean foundation to test designs on.

What is a good CPC for a t-shirt mockup?

Under $1, with enough spend behind it to trust the number (at least around $5 of spend on that mockup). A sale during the test counts as a win too. Ignore a great CPC that only spent a few cents; there is not enough data.

Does PrntFlow make or test mockups for me?

No. You make and test your mockups. PrntFlow automates the rollout after you have a winner: compositing it onto every color, uploading to Shopify, and assigning each variant across your catalog.