Sarah spent fourteen hours hand-stitching a silk coin purse. It was, objectively, a tiny masterpiece. She photographed it on her dining table, which happened to be covered in a cheerful chaos of mail, a half-eaten granola bar, and what appears to be a tax form from 2019. The listing got twelve views. Zero sales. One person messaged her to ask if the granola bar was also for sale.
This is not a story about photography skill. This is a story about backgrounds - the silent assassins of online product listings.
The Brutal Truth About Product Backgrounds
Marketplaces like Etsy, Amazon, and eBay have increasingly strict (or at least strongly encouraged) guidelines around product image backgrounds. Amazon outright requires pure white backgrounds for main product images. Etsy stops short of mandating it, but their own internal data and every conversion rate study in existence confirms the same thing: clean backgrounds sell products. Cluttered backgrounds sell confusion.
The psychology is not complicated. When a buyer looks at your product, you want exactly one thing competing for their attention: your product. Not your sofa. Not your cat, however photogenic. Not the pile of laundry that wandered into frame with the quiet confidence of someone who lives there.
The traditional solution to this problem costs money. A proper lightbox photography setup runs anywhere from $40 for a flimsy pop-up tent to several hundred dollars for a proper sweep setup with softbox lighting. Then there's the time investment of learning to use it, the storage problem of owning it, and the uncomfortable realization that your apartment does not actually have a room you can dedicate to photographing earrings.
What AI Background Removal Actually Does
The good news is that for most product photography, you no longer need any of that equipment. AI background removal has gotten genuinely impressive at isolating subjects from their surroundings, even in tricky situations like transparent objects, hair, fur, or products photographed against backgrounds that are almost the same color.
The process works by having a neural network analyze your image and identify what is foreground versus background. Good implementations handle edges with real nuance - they do not just draw a rough rectangle around your product and call it done. Fine details like the wispy fringe on a handwoven scarf, the individual prongs of a ring setting, or the translucent edge of a glass bottle all get preserved rather than brutally cropped away.
What you get is a clean cutout of your product, which you can then place on a white background, a lifestyle scene, a branded color, or literally nothing at all - which is useful for websites that let you layer images.
The Right Way to Shoot for Background Removal
Even with AI doing the heavy lifting, a few simple habits when taking your original photo will dramatically improve results. Think of it as giving the algorithm a fighting chance.
- Light your product, not your background. High contrast between subject and background makes edge detection much cleaner. Natural window light pointed at your product works beautifully.
- Avoid matching colors. If you are photographing a white ceramic mug, do not put it on a white tablecloth. The AI will get philosophical about where the mug ends. A dark cutting board or a contrasting fabric gives the algorithm something to work with.
- Shoot at the highest resolution your phone allows. More pixels mean more detail in the edges of your subject. If you later need to resize the image for your listing, you want to start big and go small, not the other direction.
- Keep the product in focus and centered. Background removal does not fix blur. If your product looks like it was photographed through a frosted window, no amount of edge detection will save you. Use the tap-to-focus feature on your phone and actually use it on the product itself, not whatever happens to be behind it.
Beyond White: What to Do With Your Cutout
White backgrounds are the safe, universally acceptable choice. They are also, if we are honest, a little boring. Once you have a clean cutout of your product, you have options that small sellers rarely use but absolutely should.
Lifestyle backgrounds - a wooden table surface, a linen fabric texture, a tasteful marble - can make handmade products feel more premium without the expense of a full photoshoot. The trick is to use the same background across all your product listings. Visual consistency makes your shop look like a real brand, not a collection of photos taken at random intervals across several different kitchens.
You can also use your cutout to create variations. The same candle photographed on a white background for your main listing image, then composited onto a cozy autumn scene for a seasonal banner, then dropped onto a branded color card for Instagram. One photo, three uses, zero additional photoshoots.
If your products need any additional touch-ups after background removal - maybe the lighting looked a bit flat once isolated - running the image through AI auto-enhancement can quickly correct brightness, contrast, and color balance to make the product pop.
The Privacy Angle Nobody Mentions
Here is something worth knowing: when you upload product photos to various editing apps and cloud services, you are often giving that service rights to your images, or at minimum leaving your work sitting on someone else's server. For handmade sellers, your product photos are essentially your intellectual property. The designs in them represent your livelihood.
COMBb2's background removal runs entirely in your browser. Your product photos never leave your device. No server upload, no cloud storage, no Terms of Service paragraph 47b quietly claiming a license to your handmade goods photography. The processing happens locally, which means your unreleased designs stay unreleased until you choose to post them.
A Realistic Assessment
AI background removal is not magic, and it is worth setting honest expectations. Very reflective surfaces like mirrors or polished metal can confuse edge detection. Transparent products like glass vases require extra care. And if your product is the same color as your background - see above re: white mug on white tablecloth - you will have a bad time regardless of how clever the algorithm is.
For the vast majority of handmade products, though - jewelry, ceramics, textiles, candles, bath products, stationery - results are genuinely excellent with minimal effort. The technology has crossed the threshold from impressive-but-finicky to actually-reliable.
Conclusion
The gap between a product that sells and a product that sits has less to do with quality than presentation. Buyers cannot touch your coin purse, smell your candle, or feel your knitted hat. All they have is the photograph. Giving that photograph a clean, distraction-free background is one of the highest-return improvements a handmade seller can make - and it no longer requires a studio, a photographer, or anything other than the phone already in your pocket and a free tool that runs entirely in your browser. Sarah relisted her coin purse. It sold within a week. The granola bar, for the record, was not included.
Try it yourself
Free, private, runs in your browser. No sign-up required.
