Amazon Main Image Requirements, and What Actually Gets a Listing Suppressed

The rule, exactly as Amazon states it

Amazon's own guidance on the main image is specific rather than suggestive: a pure white background at RGB 255, 255, 255 — not off-white, not "light," the exact value — showing the real product, photographed rather than rendered, filling 85% or more of the frame. No props, no packaging in shot unless the packaging is the product, no text, no watermark, no badge graphic. The file needs to sit between 500 and 10,000 pixels on its longest side, in JPEG, TIFF, PNG or non-animated GIF, and Amazon's own guidance prefers anything above 1,000 pixels because that is the threshold the on-page zoom feature needs to work at all.

None of that is unusual as a rule. What is unusual is how it gets enforced: an automated scan reads the actual pixel values in the background, not what a human eye judges as "basically white" on a monitor. That gap between what looks compliant and what measures compliant is where most suppressions start.

What actually trips the check

Amazon runs new and edited images through an automated quality and compliance scan before they go live, checking the background colour, the frame-fill ratio and basic image quality (blur, jagged edges from a bad crop or a low-resolution source) against the published thresholds. This is not a human reviewer eyeballing each upload — it is a scan looking at specific measurable properties of the file, which is exactly why it catches deviations a person would call close enough. Understanding that the check is mechanical, not subjective, is the first step to debugging a rejection rather than guessing at it.

A background that reads white to the eye but measures 251, 251, 248 in the file will fail the same way a visibly grey background fails — the check does not grade on a curve. The most common causes we see are not deliberate rule-breaking: a JPEG compression pass that shifts white pixels toward a warm or cool cast, a background removed with a tool that leaves faint anti-aliasing at the product's edge, or a photo shot under warm studio lighting and never colour-corrected back to true white before export. All three produce an image a person would call compliant and a scanner would not.

Frame fill is the second common cause, and it fails in both directions — a product photographed too small leaves too much white margin below the 85% threshold, and a product cropped too tight can trip the same check from the other side if part of it is cut off. Text, a size chart baked into the main image, or a small "New!" badge in the corner are rule violations a human catches immediately, but in our experience they are less common than the pixel-level background failures, because sellers who know the badge rule exists do not add badges on purpose.

A worked example: the same photo, two outcomes

Take a bottle photographed on a genuine white backdrop, product filling roughly 80% of the frame, 2,200 pixels on the long side. On the file's own numbers this already fails one rule before anyone looks at the background colour — 80% is under the 85% frame-fill floor, so the fix is cropping in, not re-shooting. Now take the same bottle re-shot to fill 90% of the frame correctly, exported at 95% JPEG quality with a slight warm colour profile carried over from the studio lights. An eyedropper tool on the corner pixels reads 254, 251, 247 — visually indistinguishable from white, numerically not 255, 255, 255. That second image can fail a stricter automated pass that the first, worse-cropped image happened to pass, because compliance checks do not evaluate the two rules with equal strictness in practice. The lesson is not to relax on frame fill because background gets flagged more — it is to check both independently, because passing one tells you nothing about the other.

What to do when a listing gets flagged

Manage All Inventory's Suppressed filter names the specific attribute Amazon flagged rather than leaving it to guesswork — check that before re-shooting anything, because a wholesale reshoot when only the background needs a colour correction wastes days a targeted fix would not have cost. If the flag is genuinely the background, run the file through an eyedropper or histogram check before re-uploading; a fix that still is not exactly 255, 255, 255 will fail again and reset the review clock.

On a live listing, change one thing at a time. A wholesale swap of every image at once forces Amazon to re-read the whole record rather than the one attribute that changed, and can cost rankings that were built on the version being replaced — that is standing practice across the accounts we manage, and it applies as much to a compliance fix as to a deliberate creative refresh. One brand we worked with had four listings go into search suppression at once; getting all four reinstated took 40 to 45 days of escalations and appeals, and the recovery afterward still had to be staged one change at a time rather than restored in a single batch.

The mistake worth naming directly

The recurring failure is not ignorance of the rule — most sellers who have listed more than a handful of ASINs know the white-background requirement cold. It is treating "looks white on my screen" as the same test Amazon runs. A monitor's colour profile, ambient lighting in the room, and even the browser rendering the preview can all make an off-white background look correct to the person checking it. The only test that matches Amazon's is reading the actual pixel values in the file, which takes under a minute with any image editor's eyedropper tool and catches the failure before upload rather than after.

Where this sits in the bigger picture

On the latest listing-quality snapshot across 47 brands we manage, 214 ASINs were sitting in search suppression and 648 more in attribute suppression — part of a snapshot averaging 93 open listing issues per brand across all issue types — and image compliance is a recurring share of both categories. These come from Full Circle, a full-service Amazon management company with $500M+ in managed revenue across 100+ brands. Dr. CRO builds every Hero Shot to this exact spec from the start — pure white, correct frame fill, no added text or props — so the compliance check is passed before the image is ever uploaded rather than discovered after a suppression notice.