Product Imagery for E-commerce: Standards That Scale

Product imagery for e-commerce works at scale only when every image follows a written standard that systems can check. Without one, each photographer, agency and supplier delivers a different crop, background and filename, and product content teams spend their time fixing files instead of publishing products. This guide sets out the elements of a practical image standard, how it aligns with GS1 guidelines and marketplace rules, and how a DAM automates the work.

A written image standard is the foundation

An image standard is a document that defines, per product type, which images are required and how each must be produced. It turns subjective quality into checkable rules. A complete standard covers six elements:

• Shot list per product type, with mandatory and optional views.
• Angles and framing, including camera height, product orientation and margin.
• Background and lighting, usually pure white for the main image.
• Resolution and file format for the master file.
• Naming convention linking each file to a product.
• Metadata that must accompany each file, such as GTIN, view, colour and rights.

Store the standard in the DAM as a governed brand asset, and share it with every photographer, studio and supplier.

Shot lists and angles should follow the product type

A shot list differs per category. A typical starting point:

Fix angles once and reuse them. Consistent angles make category pages look orderly and allow automated cropping to work reliably across thousands of SKUs.

GTIN-based naming makes automation possible

A naming convention is the cheapest automation you can buy. The GS1 Product Image Specification recommends building file names from the product's 14-digit GTIN followed by codes for the view and other characteristics. A DAM can parse that name at upload, write the GTIN to a metadata field and create the link to the product record, as described in PIM and DAM integration architecture.

Keep the convention short and predictable: GTIN, an underscore, a view code and a sequence number. Avoid free text, spaces and campaign names in filenames. Campaign context belongs in metadata governed by your metadata strategy.

Background, resolution and format rules protect quality

Main images on marketplaces almost always require a pure white background. Amazon, for example, specifies pure white (RGB 255, 255, 255) and expects the product to fill most of the frame. Secondary images allow more freedom.

For resolution, shoot and store one high-resolution master and never upscale. In our experience a master of 2,000 to 3,000 pixels on the longest side covers zoom on most webshops and marketplaces, while packaging and print need more. Store masters in a lossless format such as TIFF or high-quality PNG, with a defined colour profile, usually sRGB for web use.

Marketplace requirements should set your minimum bar

Each marketplace publishes its own image rules. Amazon, bol.com and Zalando differ in minimum size, background, aspect ratio, the use of text or watermarks and the number of images per product. Fashion platforms such as Zalando typically prescribe detailed on-model and packshot formats for brand partners.

Build your standard to satisfy the strictest channel you sell on, then derive the others. Document each channel's rules as a channel profile in the DAM or syndication tool, and review them twice a year because marketplaces update guidelines without much notice. See product content syndication for the channel side.

Automated renditions remove manual export work

A rendition is a derived version of a master file with a specific size, crop, format or background. Modern DAMs and image CDNs generate renditions on the fly: WebP or AVIF for webshops, square crops for marketplaces, JPEG with fixed dimensions for retailer portals. One master feeds every channel.

Add automated quality checks at upload: minimum resolution, background colour tolerance, aspect ratio and filename pattern. Files that fail go to a review queue before they reach the PIM. AI-based tools can also remove backgrounds and generate alt text, though they still need human review for accuracy.

Recommendation

Write the image standard before you choose or configure tools. Base naming on GTIN, align the shot list with the GS1 Product Image Specification and your strictest marketplace, and store one master per view. Then configure the DAM to validate files at upload and generate renditions per channel. This turns imagery from a recurring clean-up task into a repeatable process that scales with your catalogue.