Selling on Amazon means accepting strict visual discipline. The platform doesn't just recommend photo best practices: it imposes a precise spec, and any listing that strays from it risks suspension, the loss of the Best Seller badge, or worse, total invisibility in search results. For an FBA or Vendor Central seller, a non-compliant Amazon product photo is revenue left on the table.
This complete guide details all of Amazon's technical requirements in 2026 — main image, secondary images, pure white background, 85% framing — then shows how generative AI now lets you produce compliant visuals in about thirty seconds, with no studio or shoot. The payoff: a scalable catalog, ready for the Vendor approval workflow, and compliant with Seller Central's automated checks.
Why Amazon product photos are so critical
On Amazon, the photo isn't an aesthetic detail: it's the first purchase trigger. The platform's internal studies and independent reports (Jungle Scout, Helium 10) converge on the same figure: a listing with professional photos converts up to twice as well as one with mediocre photos. On a €30 product, going from an 8% to a 16% conversion rate changes the economics of an entire seller account.
But the stakes don't stop at conversion. Amazon's ranking algorithm, A9, factors visual quality into its signals: a listing with a non-compliant main image is downranked or even suspended, making it invisible in search. Vendor Central sellers, meanwhile, see their new ASINs blocked in the approval workflow if the visuals don't meet the guidelines. The Amazon photo is therefore both a marketing tool and a technical prerequisite for eligibility to sell.
Amazon Seller Central requirements for the MAIN image
The main image — the primary image that appears on the product page and in search results — is the most tightly regulated. Amazon applies zero tolerance here. Here are the exact rules in force.
Technical format: JPEG, TIFF, pixel dimensions
Amazon accepts JPEG (recommended), TIFF, GIF and PNG. JPEG remains the de facto standard: it offers the best weight/quality compromise for web display, and it's the format most ASIN integrations use by default. PNG is technically accepted but discouraged for product photos (excessive weight with no visible quality gain).
On dimensions, two thresholds to know:
- 1000 × 1000 pixels minimum — this is the threshold that enables the zoom function on the product page. Below it, the image displays but isn't zoomable, which massively degrades the desktop UX.
- 2000 × 2000 pixels recommended — this is the size that allows high-resolution zoom on Retina and 4K screens. All serious sellers target this resolution.
The ratio must stay square (1:1) for the main image, matching Amazon's standard catalog format. Landscape or portrait images may be accepted but will display with white bars that visually penalize the listing.
The pure white background (#FFFFFF — not approximate)
This is the strictest rule, and the one that gets the most listings rejected. The main image must have a pure white background, RGB value 255-255-255, hex code #FFFFFF. Not off-white, not light gray, not a slight gradient imperceptible to the eye. The check is automated: Amazon scans the background pixels and rejects anything that isn't exactly #FFFFFF.
This requirement is a real problem with classic photography. Even in a professional lightbox, the background captured by the camera is almost always around #F8F8F8 or #FAFAFA — close to white but not pure. You therefore systematically need a post-production cutout (Photoshop, GIMP, or a dedicated tool) to reach true #FFFFFF. Across 100 SKUs, this retouching step quickly becomes a bottleneck.
85% framing: the occupancy rule
The product must occupy at least 85% of the frame of the main image. This rule exists for a simple reason: on the SERP thumbnail (the results grid), Amazon displays the image at a very small size. If the product doesn't fill the frame, it becomes illegible and the CTR collapses.
85% is an operational threshold: in practice, aiming for 85-90% leaves a comfortable safety margin. Beyond 95%, you risk visually cropping the product (touching the edges), which is also penalized. Clean framing means a centered product with 5 to 10% white margin all around.
The absolute no-gos
Amazon explicitly lists what's forbidden on a main image. The slightest breach gets the listing rejected:
- Text of any kind — no slogans, no "100% organic" claims, no prices.
- Logos and watermarks — even a small brand logo in a corner is forbidden.
- Accessories not included in the sale — if you sell a mug alone, the spoon and saucer must not appear.
- Mannequins — except in clothing and lingerie categories where they're allowed or even required.
- Multiple products — only one product visible, unless the SKU is explicitly sold as a bundle (in which case the full bundle must be visible).
- Borders, frames, illustrations — the main image is a photo, not a graphic.
- Background colors other than white — no gradient, no vignette, no overly artistic shadow.
The 8 secondary images: seize the opportunity
Amazon allows up to 8 additional images beyond the main image. These secondary visuals are much less regulated: the white background isn't required, text is allowed, and this is where most of the commercial persuasion happens. Under-using these slots means letting competitors take the advantage.
Lifestyle: the product in use
A lifestyle image shows the product used in its natural environment: a mug on a breakfast table, a backpack worn on a hike, headphones on someone on the train. These visuals create a mental projection in the buyer — they already picture themselves using the product. It's the most powerful conversion lever after the main image.
Infographic: dimensions, materials, features
An infographic overlays text and visual indicators on a product photo to communicate key specs: precise dimensions (in cm and inches), materials, certifications, strengths. This type of image is highly valued because it answers technical questions before they arise — and reduces returns for "wrong size" or "not what I expected."
Detail zoom: textures, finishes, quality
A close-up on a stitch, a finish, a texture, an engraved logo, proves manufacturing quality. On premium products (leather, electronics, cosmetics), these zooms are decisive: they differentiate a premium product from a low-cost equivalent whose main photo would be identical. For beauty specifically — material rendering, label legibility — see our guide to AI cosmetics product photos.
Packaging comparison: show what the buyer receives
An image showing the box contents (product + accessories + manual + packaging) reassures and clarifies the offer. For a bundle product, it's essential. For a simple product, it reduces support questions.
Contextual use: the product in a real situation
Different from pure lifestyle, this type of image shows a specific function of the product in action: the pot simmering, the vacuum picking up a precise type of dirt, the bag revealing its open compartments. More didactic than lifestyle, more focused on functional proof.
Vendor Central specifics: volume and approval workflow
Vendor Central is for brands that sell wholesale to Amazon (Amazon becomes the customer, not the marketplace). The photo rules there are identical to Seller Central on the main image, but two additional constraints change the game.
Volume: a Vendor account typically manages several thousand SKUs, with seasonal catalog refreshes (spring/summer, fall/winter, capsule collections). Producing even the main image for each SKU with a traditional shoot becomes economically untenable: at €30 minimum per studio photo, 1000 SKUs = €30,000 just for main images. Hence the pressure on scalability.
The approval workflow: every new visual goes through an internal Amazon validation cycle, which can take several days. A rejected main image means a blocked ASIN, sometimes during a critical commercial week. Vendor teams therefore have every interest in iterating quickly and not having to wait for a re-shoot to fix a detail.
Traditional method vs AI for Amazon
The full production cycle of an Amazon-compliant photo, classic method vs generative AI, compares as follows:
| Criterion | Classic shoot + retouching | Caralens (AI) |
|---|---|---|
| Production time / SKU | 3 to 6 hours (logistics, shoot, background retouch, export) | ≈ 30 seconds (upload source photo → compliant photo ready) |
| Cost / SKU | €30 to €80 (studio + photographer + retoucher) | ≈ €0.44 with the Pro plan (50 photos / €21.99) |
| Pure #FFFFFF white background | Requires a systematic post-production cutout | Generated natively by the AI, no retouching |
| 1:1 format (1024×1024 and up) | Manual cropping after the shoot | 1:1 format selectable at generation time |
| 85% framing | Depends on the photographer + cropping | Default centered, compliant framing |
| Iteration (color variants, angle) | A new shoot required | Re-generation in 30 seconds |
| Catalog scalability (1000 SKUs) | Several weeks, five-figure budget | A few days with a Pro plan, budget < €500 |
Tutorial: generating an Amazon-compliant photo with Caralens
Here's the concrete workflow, usable from your very first upload on Caralens.
Step 1: the source photo (smartphone OK)
Take a photo of the product with your smartphone, in reasonable conditions: daylight or a neutral lamp, the product on a plain surface, an angle that clearly shows the shape. No studio needed, no white background at this stage — the AI will handle it. Just avoid blurry, underexposed photos, or ones with a cluttered set that hides the product.
Step 2: choose the 1:1 format
In the Caralens interface, select the 1:1 (1024×1024) format — it's the standard Amazon catalog format for the main image. If you want to go beyond the 1000 px minimum (recommended), Caralens natively provides 1024×1024, enough to enable zoom. To target the recommended 2000×2000, you can upscale the result with a third-party tool, but 1024 remains perfectly acceptable and compliant.
Step 3: describe the expected result
In the description field, be precise about Amazon's constraints. Example of a prompt that works well:
"Professional product photo on a pure white background (#FFFFFF), compliant with Amazon requirements. Product centered, occupying 85% of the frame, soft natural shadow under the product, neutral studio lighting. No accessories, no text, no logo. E-commerce packshot style."
Caralens uses Claude to turn this description into a prompt optimized for the image generation model. The result is on average compliant from the first iteration.
Step 4: quick check before upload
Before sending the visual to Seller Central, run three quick checks:
- Eyedropper on the background — confirm the background is indeed #FFFFFF (ColorZilla, Photoshop, or any color picker).
- Framing — does the product visually occupy ≥ 85% of the frame without touching the edges?
- No-gos — no stray text, no unsold accessory, no exaggerated shadow.
If these three points pass, the visual will clear Amazon's automated checks without trouble.
Common mistakes that get a listing rejected
From experience, it's the same mistakes that keep coming up and push a listing into "Image issue" status:
- Off-white background — #F8F8F8 or #FAFAFA instead of pure #FFFFFF. The #1 cause of automatic rejection.
- Logo or watermark — even a small "© Brand" in the bottom-right is enough to fail the check.
- Ghost accessories — the spoon next to the yogurt, the flower in the vase sold empty, the mouse next to the laptop. Anything not in the box must disappear.
- Framing too tight — the product touches the edges, feels cramped and lowers the CTR.
- Framing too wide — the product occupies 40% of the frame, becomes illegible as a SERP thumbnail.
- Image below 1000 px — no immediate rejection but zoom disabled, which degrades conversion.
- Multiple SKUs in one photo — unless sold as an explicit bundle, showing several variants on the main image is forbidden.
- Competitor catalog photo — using an image that isn't yours = immediate suspension for IP infringement.
FAQ — Amazon product photos
Can I use an AI photo for my Amazon main image?
Yes, Amazon doesn't prohibit the use of AI-generated photos, as long as the visual respects the technical rules (pure white background, 85% framing, no text) and faithfully represents the product. Many FBA and Vendor sellers already use AI to produce main images and infographics at scale.
What pixel size should I target to be completely safe?
1000 × 1000 is the strict minimum to enable zoom. 2000 × 2000 is Amazon's recommended size and the one used by serious sellers. Caralens natively provides 1024 × 1024, enough for compliance and zoom — you can upscale to 2000 if you want maximum resolution.
Is a pure white background required on all my 9 images?
No. Only the main image (the first one) must have a pure white #FFFFFF background. The 8 secondary images can (and should!) be lifestyle, infographics, detail zooms, etc. — with colored backgrounds, overlaid text, and staging.
Vendor Central: how many images should I provide per SKU?
The strict minimum is 1 (the main image), but in practice all Vendor buyers require 5 to 9 visuals per SKU to validate a listing. The more expensive or complex the SKU, the denser the expected visual pack.
Can I change the photos of a listing that's already live?
Yes, at any time, from Seller Central → Inventory → Edit → Images. New images go back through an automatic check, but if they're compliant, the replacement takes effect within a few hours. It's an excellent lever to boost an underperforming listing without touching the title or bullet points.
Conclusion
The Amazon product photo is one of the rare areas where the platform imposes precise technical rules, automatically checked, and penalized in case of non-compliance. Pure #FFFFFF white background, 85% framing, 2000 × 2000 recommended, no stray text or accessory on the main image — these criteria are non-negotiable.
The good news is that in 2026 these requirements no longer need a studio or a dedicated retoucher. A smartphone photo, a well-configured AI workflow, and 30 seconds of generation are enough to produce a compliant visual. For a growing FBA seller or a Vendor account with a large catalog, it's the difference between a creative bottleneck and an industrialized pipeline.
To go further, two complementary reads on the Caralens blog: our complete AI packshot guide which details the source-shot technique and the best prompts, and our Shopify product photo guide if you manage a multi-platform catalog. And if you want to test directly, create a Caralens account and generate a compliant main image and its variants in minutes. Pricing details are on the pricing page.