Google Shopping image disapprovals usually come from a small set of preventable mistakes: too-small source files, promotional overlays, placeholder images, variant mismatches, blocked image URLs, and landing pages that hide product images from crawlers. Fixing those issues is not just an ad approval task. The same image discipline helps products appear across Google Search, Google Images, Google Lens, and the Shopping tab.
The Fast Diagnostic Table
Start with the disapproval reason in Merchant Center, then use this table before editing creative.
| Symptom | Likely Cause | Fast Fix |
|---|---|---|
| Image too small | Source image below Google's minimum | Replace with 1500 x 1500 px or larger when possible |
| Promotional overlay | Sale text, shipping claim, watermark, logo, border | Submit a clean product image as image_link |
| Wrong variant image | Feed image does not match color, pattern, or material | Generate one image URL per variant |
| Placeholder or generic graphic | Main image is not the actual product | Use a real product photo for image_link |
| Product cropped | Main product not fully visible | Re-crop with full product visible |
| Crawl or fetch error | URL blocked, changed, redirected, or too slow | Use a stable HTTPS URL and allow Googlebot |
| AI metadata issue | Synthetic image metadata removed | Preserve required IPTC metadata for AI-generated images |
Do not fix every item manually in Merchant Center first. Fix the source image library and feed logic. Otherwise the next feed push can overwrite your repairs.
Mistake 1: Treating the Minimum Size as the Target
Google's Merchant Center image attribute requires product images to meet minimum dimensions and file limits. But the minimum is not the operating target.
| Rule | Practical Seller Target |
|---|---|
| Minimum size | At least 500 x 500 px |
| Recommended performance size | Around 1500 x 1500 px or above |
| Maximum resolution | No larger than 64 megapixels |
| Maximum file size | No larger than 16 MB |
If you sell apparel, home goods, electronics, beauty, or accessories, upload a square master image around 1500-2500 px. That gives Google enough detail for thumbnails, zoom-like surfaces, and visual matching without creating a huge file.
Fix
Create a product image master:
| Product Type | Recommended Master |
|---|---|
| Small accessories | 2000 x 2000 px |
| Apparel | 2000 x 2000 px or 2400 x 3000 px with square-safe crop |
| Furniture | 2000-3000 px on long side |
| Beauty / packaging | 2000 x 2000 px |
| Bundles | 2000 x 2000 px with all included items visible |
If your current image is 800 x 800 px and blurry, do not upscale it and call it fixed. Reshoot or find the original high-resolution source file.
Mistake 2: Using Marketplace Infographics as the Main Image
Images that work as secondary marketplace graphics often fail as Google Shopping main images. Google explicitly rejects promotional elements and overlays in the primary image.
Common rejected elements include:
- Buy-now wording
- Free shipping claims
- Sale prices
- Promotional adjectives
- Compatibility claims
- Watermarks
- Brand overlays that are not part of the actual product
- Borders
- Barcodes
The main image should show the product. Put marketing graphics in additional images, landing page content, or marketplace-specific secondary slots.
Fix
Split your image set by role.
| Role | Use This Image |
|---|---|
image_link |
Clean product image, no overlay, actual product visible |
additional_image_link |
Angles, lifestyle, scale, feature images |
| Landing page gallery | Full product story, including infographics where allowed |
| Marketplace secondary images | Platform-specific callouts and dimension graphics |
If your image library only has infographic versions, create clean base images first. Do not try to remove text from flattened JPEGs if you still have the original product photo.
Mistake 3: Reusing One Image Across Variants
Google recommends submitting a unique image that shows the distinguishing details of each variant. This matters for color, pattern, material, size pack, and bundle contents.
| Product Setup | Bad Feed Image | Better Feed Image |
|---|---|---|
| Couch in green and blue | Every SKU uses green couch | Each color variant uses matching image |
| T-shirt with print variants | One generic blank shirt | Correct print shown per variant |
| Jewelry finish variants | Gold image for silver SKU | Finish-specific photo |
| Multi-pack bundles | Single item photo | Bundle photo showing all included products |
Variant mismatch causes disapproval risk, buyer distrust, and wasted clicks. A shopper who clicks a blue product and lands on a green image is less likely to buy even if the variant selector is technically correct.
Fix
Add an image QA column to your feed spreadsheet:
| Column | Example |
|---|---|
item_group_id |
linen-shirt-001 |
color |
sage green |
size |
M |
image_link |
URL of sage green shirt |
image QA |
matches color + material |
For large catalogs, automate checks by comparing filename patterns or DAM tags to variant attributes. Manual review should focus on top-selling SKUs and products with frequent disapprovals.
Mistake 4: Cropping Out the Product or Hiding Bundle Contents
Google says the product image should accurately display the entire product with minimal or no staging. For bundles, the image should show the products included in the bundle.
Bad crops happen when sellers reuse social ads, lifestyle shots, or marketplace heroes as feed images.
| Product | Risky Crop | Better Crop |
|---|---|---|
| Shoes | Toe and heel cut off | Full pair visible |
| Lamp | Shade visible, base cropped | Entire lamp visible |
| Bedding set | Styled bed but missing included pillowcases | All included pieces shown or clearly represented |
| Tool kit | Hero tool only | Tool plus included accessories |
Fix
Use a "full product first" crop:
- Place the full product inside the frame.
- Leave enough margin that automated thumbnails do not trim edges.
- Keep props minimal.
- Use additional images for detail close-ups.
- For bundles, include every material item in the main image.
This is especially important for products where the buyer cares about included parts: electronics kits, tool sets, cookware bundles, bed sets, and craft supplies.
Mistake 5: Changing Image Content Without Changing the URL
Google can take time to detect changed image content if the URL stays the same. Many ecommerce platforms overwrite product.jpg with a new image but keep the URL identical. That slows down recovery after a disapproval because Google may keep seeing the old image.
Fix
Use versioned image URLs:
| Bad | Better |
|---|---|
/images/product-main.jpg |
/images/product-main-v2.jpg |
/sku123.jpg overwritten repeatedly |
/sku123-green-front-2026-05.jpg |
| CDN URL with same path after edit | New path or cache-busting URL handled by platform |
If your platform generates URLs automatically, check whether replacing an image creates a new CDN URL. If not, use the platform's media replace workflow carefully and request reprocessing in Merchant Center after the URL changes.
Mistake 6: Letting Crawlers Miss the Image
Google Search Central says standard HTML image elements help crawlers find images, while CSS background images are not indexed the same way. Ecommerce pages often hide important product photos inside scripts, sliders, CSS backgrounds, or blocked CDN paths.
Fix
Ask your developer or theme owner to check:
| Check | Pass Standard |
|---|---|
| Product image markup | Uses an HTML img element with a valid src |
| Responsive images | Has fallback src, not only srcset |
| CDN robots access | Googlebot can fetch the image URL |
| Alt text | Describes the product without keyword stuffing |
| Structured data | Product schema references a valid image URL |
| Sitemap | Important image URLs are discoverable where needed |
This is not only about Shopping ads. Google can show ecommerce content across Search, Images, Lens, the Shopping tab, Business Profile, and Maps. Image crawlability affects more than one surface.
Mistake 7: Removing Required AI Metadata
Google's Merchant Center image guidance says images created using generative AI must contain metadata indicating that the image was AI-generated, and sellers should not remove embedded metadata tags such as IPTC DigitalSourceType.
This does not mean every AI-assisted image is rejected. It means your workflow needs to preserve the disclosure metadata when the image is created using generative AI tools.
Fix
For AI-assisted images:
- Keep the original export with metadata intact.
- Avoid stripping metadata during compression.
- Do not use AI-generated images to misrepresent the actual product.
- Use real product photography for the main image when accuracy is critical.
- Keep a clean audit trail of source image, edit date, and tool used.
When in doubt, use AI for background cleanup and secondary creative, not for inventing product details.
A 30-Minute Recovery Workflow
Use this when Merchant Center starts showing image disapprovals.
Step 1: Export the Affected Items
Pull product ID, title, variant attributes, current image_link, disapproval reason, and landing page URL.
Step 2: Group by Cause
Do not work item by item. Group by pattern:
| Pattern | Typical Root Cause |
|---|---|
| Same image URL rejected across many SKUs | Reused overlay or broken URL |
| One color family rejected | Variant image mismatch |
| Only mobile landing pages affected | Theme or CDN issue |
| Recently edited images still rejected | Same URL, cached old content |
Step 3: Replace the Source Image
Create a clean product image that meets the spec, upload it as a new file, and update the feed URL.
Step 4: Validate the Landing Page
Open the product page in an incognito browser. Confirm the main image matches the feed image, loads fast, and appears in normal HTML.
Step 5: Resubmit or Wait for Feed Reprocessing
After the feed is updated, request review or wait for the next Merchant Center processing cycle, depending on the issue type.
FAQ
What image size should I use for Google Shopping in 2026?
Use at least 500 x 500 px because that is the minimum stated in Merchant Center guidance. For real catalog work, aim around 1500 x 1500 px or larger when possible, staying under 64 megapixels and 16 MB.
Can Google Shopping images have text?
The main image_link should not contain promotional text, calls to action, prices, free shipping claims, borders, watermarks, or logos that are not part of the product itself. Use additional images or landing page content for richer explanations.
Should I submit lifestyle images or white-background images?
Use the image that shows the actual product clearly with minimal staging for the main feed image. Lifestyle images can work when they do not obscure the product, but heavily staged scenes often create cropping and clarity problems. Put lifestyle and scale images in additional slots.
Do product annotation images help Google Shopping?
Use annotation images as secondary assets, not as the primary image_link. For landing pages and additional images, clean measurements and callouts can reduce buyer uncertainty. Tools such as SizeMarker are useful when you need dimension graphics that stay readable without turning the main feed image into an ad banner.
Why did Google keep rejecting my image after I fixed it?
The most common reason is that the image URL did not change, so Google may still be seeing the old cached image. Upload the corrected image as a new file, update the feed URL, confirm Googlebot can fetch it, then request review or wait for processing.
Sources & References
- Google Merchant Center image hosting and upload guide — Img.vision
- Google's 1500×1500 image recommendation for product feeds — ALM Corp
- Products resource — Content API for Shopping
- Google Image SEO best practices — Google Search Central
- Where ecommerce content can appear on Google — Google Search Central