Cut-Out Product Images at the Touch of a Button: How DACH Retailers Automate Their Image Processing

What “Freigestellt” Means and Why It Matters

A “freigestelltes” product image — cut-out image or white background image — shows the product alone, without any context, on a clean white or transparent background. This is not an aesthetic choice: it is a hard requirement on Amazon and a strong quality signal on most other e-commerce platforms.

The reason is functional. A cut-out image focuses the customer’s attention entirely on the product. It is also the standard format that allows marketplaces to display consistent product grids — no distracting backgrounds, no color clashes with the platform design, no competing visual elements.

For Amazon specifically, the main product image must show the product on a pure white background (RGB 255, 255, 255). Listings that do not meet this standard are suppressed. There is no workaround and no exception. This makes automated, scalable background removal not a nice-to-have — it is infrastructure.

The Traditional Workflow and Its Problems

In-house with Photoshop

The classic approach is manual background removal in Photoshop using selection tools, masking, or the Magic Eraser. For a skilled designer, a simple product image takes 5–15 minutes. For complex products — shoes with laces, jewelry, glasses, products with fine textures — 30 minutes or more per image is realistic.

At 1,000 products with one main image each, manual processing requires 83–250 hours of designer time. At 10,000 products, this scales to 833–2,500 hours — and that is just for the initial batch, not ongoing updates.

Outsourcing to agencies or freelancers

Many retailers outsource image processing to specialized agencies or low-cost freelancers on platforms like Upwork. The typical cost ranges from €0.50 to €2.00 per image for basic background removal. At scale, this becomes a significant line item — and it introduces turnaround times of 24–72 hours, which creates bottlenecks when launching new products quickly.

The fundamental problem: neither scales

Both approaches have the same structural limitation: they do not scale. When your catalog grows from 2,000 to 10,000 SKUs, or when a supplier sends 500 new products on a Friday afternoon, manual processing and agency outsourcing cannot keep pace. The backlog grows, products launch without proper images, and channels either suppress listings or show incorrect images.

How AI Background Removal Changes Everything

Modern AI background removal models — trained on hundreds of millions of product images — can process a standard e-commerce product image in under one second with accuracy that equals or exceeds what a careful human editor produces.

The key capabilities of AI-powered image processing:

• Background removal: Isolate the product from any background — studio shots, warehouse photos, supplier images with complex backgrounds — in milliseconds per image
• White background fill: Place the product on RGB 255, 255, 255 white for Amazon compliance, or keep transparency for channels where that is appropriate
• Format conversion: Convert between JPEG, PNG, WebP automatically based on channel requirements
• Resizing: Scale images to channel-specific dimensions — 2000×2000 for Amazon, 2048×2048 for Shopify — without manual steps
• Batch processing: Run all of the above on your entire catalog simultaneously — 10,000 images in hours, not weeks

The Quality Question: Is AI Good Enough?

For most e-commerce product categories, AI background removal is indistinguishable from careful manual work. Electronics, packaged goods, footwear, apparel on mannequins, furniture, toys — all of these are handled at very high accuracy.

The edge cases where AI may need a manual review: fine jewelry with hairline details, clear glass or transparent packaging, products with fur or fabric with complex textures, and products that blend visually with the original background. For these categories, AI still removes 90%+ of the work — the final touch-up is minimal.

For a realistic assessment: at 10,000 product images processed by AI, perhaps 200–500 will require a quick manual check. That is a 95–98% reduction in manual effort compared to the traditional workflow.

Productbay: Full Image Automation in One Pipeline

Productbay’s DAM integrates AI image processing directly into the product data workflow. Here is how it works end-to-end:

1. Image import: Supplier images, studio shots, or photographer deliveries are imported into the DAM — in bulk, via URL, or via supplier portal
2. AI processing: Background removal, white background fill, and format variants are generated automatically — no manual trigger required
3. Product linking: Images are automatically linked to the correct products based on EAN or article number matching
4. Channel export: When a product is exported to Amazon, OTTO, or Shopify, the system delivers the channel-appropriate image variant — JPEG on white for Amazon, WebP for Shopify — without any manual step

The AI features in Productbay extend beyond images — AI also enriches product texts, fills missing attributes, and generates channel-specific descriptions. Image processing is one part of a fully automated product content pipeline.

Every hour spent manually removing image backgrounds is an hour not spent on strategy, new products, or customer experience. AI automation returns that time — at any catalog size, starting from day one.

Frequently Asked Questions