White background product photos: how to shoot, edit and scale them
UPDATED 8 MIN READBY THE SCENOX COMMERCE TEAM
SHORT ANSWER
A pure white background means RGB 255, 255, 255, which a white wall or sheet will not give you straight out of the camera. Light the backdrop separately from the product, expose so the background just clips to white, then finish with a levels adjustment or background removal and check the result with a colour picker.

ON THIS PAGE
A white background looks like the easy option, and it is the one most often done badly. The problem is not the backdrop. It is that "white" to the eye and "white" to a marketplace checker are different things, and most home setups produce a pale grey that nobody notices until the listing is rejected or the collection page looks patchy. This guide covers how to shoot it properly, how to finish it in editing, what the main platforms ask for, and how to keep it consistent across hundreds of products.
What "pure white" actually means
Pure white is RGB 255, 255, 255, or #FFFFFF. Every pixel of background should hold exactly that value, right up to the product edge. A background that is 248 or 250 looks white on a phone screen but is not, and it shows as a faint box when the image sits on a white page.
You check it with a colour picker, not your eyes. Open the image in any editor, sample the four corners and a few points along the product edge, and read the values. If they are not 255 across the board, it needs a fix.
A shot that is right at capture needs far less rescue in editing. The aim in the studio is a background that is slightly overexposed, and an edit that finishes the job.
Equipment you need
You do not need much. The kit that matters:
- Backdrop: a white sweep (paper roll, acrylic sheet or foam board curved up the wall) or a small lightbox for tiny items. A white bedsheet works but creases show up as shadows.
- Two lights on the backdrop: soft sources aimed at the background, separate from the lights on the product. This is the single biggest improvement over a one-light setup.
- One or two lights on the product: softboxes, umbrellas or a window with a diffuser. Side or front-side, not a hard overhead.
- Reflectors: white foam board to bounce light into shadows, black card to add definition on glossy edges.
- Tripod: so every shot of every product sits in the same position.
- A grey card or colour target: for colour accuracy, see below.
Lighting setups that work
The small-product tabletop
For jewellery, cosmetics, accessories and anything that fits on a table:
- Place the product on a white sheet that curves up behind it, so there is no visible horizon line.
- Put one large soft light on each side of the product, angled slightly forward.
- Add a third light behind the product pointing at the backdrop, or feed the backdrop from two small lights behind the sweep.
- Hold white card on the side opposite the key light to lift shadows.
- Expose for the product first, then raise the backdrop light until the background reads close to full white on the histogram.
The larger-item floor setup
Furniture, bags and bulky items need a floor or wall backdrop and more distance. The rule is the same: the backdrop gets its own light. Keep the product at least a metre or so from the background so light from the product lights does not turn the backdrop grey, and so that spill from the backdrop does not wrap onto the edges of the product.
Why one light is not enough
With a single light, the background is lit at roughly the same level as the product, and the camera meter averages everything to a mid-tone. The background comes out grey, and brightening the whole image to fix it blows out the product. Lighting the background separately lets the background clip to white while the product keeps its tone.
Phone or camera
Both work. The honest comparison:
| Phone | Camera with a proper lens | |
|---|---|---|
| Resolution | More than enough for web | More than enough |
| Distortion | Wide-angle lenses stretch small products near the edges | Longer lenses (a macro or 50 to 100 mm equivalent) look natural |
| Control | Limited without a manual app | Full control of aperture, shutter and white balance |
| Consistency | Good if you lock exposure and use a tripod | Good, and easier across lighting changes |
| Cost | Already owned | Higher |
If you are using a phone, use the longest lens it offers rather than the main wide camera, lock exposure and focus, and shoot in a controlled room, not near a window whose light changes through the day. If you are using a camera, shoot RAW so you can correct exposure and white balance without damage, and keep aperture small enough that the whole product is in focus.
Editing to true white
The editing sequence matters. Do it in this order:
- Correct exposure and white balance first. Use the grey card to set white balance so the product colour is accurate. This is the step that prevents returns from "not as pictured".
- Crop to your standard ratio. A square is the safest and is accepted across channels. Keep the product centred with an even margin.
- Push the background to white. Use a levels adjustment and drag the white point in until the background reaches 255. This works when the capture was close. For stubborn shadows or uneven backdrops, move to step 4.
- Remove the background if needed. Use a selection or background-removal tool to cut the product out and place it on a layer of pure white. Refine the mask by hand where the edge is soft, such as hair, fur, glass or fine mesh.
- Fix the edges. Zoom to 200 percent and look for grey halos, jagged outlines and fringes of the old background. A one-pixel feather and a slight contract of the selection usually removes them.
- Add a contact shadow if the channel allows it. Paint or generate a soft shadow on a separate layer, so you can switch it off for marketplaces that ban it.
- Export. Save at the highest resolution as your master, then export channel versions. Use JPEG for photos and PNG where you need to keep edges crisp.
- Check again with the picker. Sample the corners and edges of the exported file.
Rules for main images on marketplaces
White is the default for a main image on most channels, but the detail differs. In summary:
- Amazon: pure white background (RGB 255, 255, 255), product alone, no text, watermarks or badges, and the product should fill most of the frame. Size and fill details vary by category, so read the style guide in Seller Central.
- Google Merchant Center: no rule requiring white, but no promotional text, logos or watermarks on the main image, and no placeholders. White mains are the easy way to stay clear of image disapprovals.
- Shopify: no platform rule on background colour. Your theme decides how images sit, and consistent white mains look tidy in collection grids.
- Meta, Flipkart and others: clean product-only main images are standard, but sizes and category exceptions differ.
The detail for each channel, with limits we could verify from official pages, is in our product image requirements guide. Treat that as the starting point and confirm in the channel's own help pages before a large upload, because these rules change.
Common mistakes
A grey background. The most frequent one. Caused by lighting the backdrop at the same level as the product. Fix with a dedicated backdrop light or a levels push.
Colour casts. A white room with coloured walls bounces green or yellow onto the product, and a mixed-light room gives you a blue-orange mess. Shoot in a neutral space, use a grey card, and set white balance before anything else.
Clipped edges. An aggressive white point or a poor cut-out slices off the soft edge of the product, leaving a flat, cut-paper outline. Light-coloured products, such as white shirts on white, are the worst case. Keep a faint shadow or a hairline edge so the product stays readable.
Halos and fringing. A pale outline left from a bad background removal. Visible at 200 percent zoom and on hover-zoom on the live store.
Uneven framing. Products at different sizes in the frame make a collection page look chaotic. Fix the camera distance and the canvas, and place the product in the same position each time.
Over-retouching. Plastic-looking skin or fabric, or colours pushed beyond what the item really looks like. It raises returns.
Inconsistent shadows. Some images with shadows, some without. Pick one style and apply it everywhere.
Doing it at catalog scale
A single product takes minutes. Five hundred products need a system, or the quality drifts after the first fifty.
- Fix the setup. Mark camera, product and light positions with tape and keep the same settings for every shot.
- Batch by category. Shoot all the same-sized items together so the framing is identical, then edit in one batch.
- Write a spec. Canvas size, margin percentage, shadow style, file naming and export settings, in a one-page document anyone can follow.
- Use presets and actions. Save your levels, crop and export steps so they run on a folder of files at once.
- Review in a grid. Put 20 finished images side by side. Inconsistencies you miss one at a time are obvious in a grid.
- Keep masters and exports separate. The master is the highest-quality version; channel exports derive from it.
For large volumes, the tedious part is cut-outs and edge cleanup. AI background removal handles most of it quickly, and the output still needs a human check for edges, glass, fine detail and colour. That mix is how our AI product photos service works: the engine does the volume and a person reviews every image.
When to do it yourself and when to hand it off
Do it yourself if you have a few dozen products, a spare corner for a light tent or sweep, and the patience to build a repeatable routine. The equipment cost is modest and the skill is learnable in a weekend.
Hand it off if you have hundreds or thousands of images, supplier photos with messy backgrounds, a launch deadline, or channels that reject uneven images. Our product photo editing service takes photos you already have and returns them on pure white to your spec, and you approve before anything goes live. You can see how it handles your own images with the free 10-product sample, if the project is a fit.


