Quick Answer
As of September 2026, Google Merchant Center still accepts images as small as 100 x 100 pixels for non-apparel and 250 x 250 for apparel, but it is warning merchants about images below 500 x 500. On January 31, 2027, the minimum becomes 500 x 500 for all primary and additional product images. Google recommends images around 1500 x 1500 or larger, with a maximum of 64 megapixels and 16 MB per file. Google documents the transition in its current image-size guidance.
Pixel dimensions are only one Google Shopping check. The file must also show the correct product and variant, remain unobstructed, come from a crawlable URL, avoid prohibited overlays, and carry the required IPTC metadata when generative AI created it. Passing Merchant Center's checks still leaves a shopper unable to inspect the product in person, so the full image set needs to reveal its angles, details, scale, and context. This guide separates current requirements from recommendations so you can fix the right problem.
What are the Google Shopping image size and file requirements in 2026?
Google Shopping's active 2026 minimum is 100 x 100 pixels for non-apparel and 250 x 250 for apparel, but merchants should prepare every primary and additional image at 500 x 500 or larger now. Google is already issuing warnings below that forthcoming threshold, and enforcement begins on January 31, 2027. For YouTube Shopping ads served on TV, 500 x 500 is already required. Google's current image-size guidance explains the transition.
| Specification | Current rule or guidance |
|---|---|
| Minimum through January 30, 2027 | 100 x 100 for non-apparel; 250 x 250 for apparel |
| YouTube Shopping ads on TV | At least 500 x 500 |
| Minimum from January 31, 2027 | 500 x 500 for all primary and additional product images |
| Recommended working size | Around 1500 x 1500 or larger |
| Maximum resolution | 64 megapixels |
| Maximum file size | 16 MB |
| Supported formats | JPEG, WebP, PNG, non-animated GIF, BMP, and TIFF |
| Main image URL | HTTP or HTTPS, ASCII, RFC 3986 compliant, 1-2,000 characters, and crawlable by Google |
The 1500 x 1500 figure is a recommendation, not the approval floor. Google separately considers an image with more than 1,024 pixels to be high resolution and says that multiple high-quality images are strongly correlated with engagement and clicks. Its image-link documentation recommends supplying the largest full-size image available within the 64-megapixel and 16 MB caps. See the official image-link specification and Google's valuable-content guidance.
Google does not prescribe a fixed aspect ratio for the main image. A square master is convenient across many commerce surfaces, but it is not a Merchant Center rule. Google may crop submitted images automatically, so keep the product clear and give important details enough breathing room.
Which main Google Shopping image rules are requirements, and which are best practices?
Google requires the main image to show the actual product accurately and without prohibited content; white backgrounds and 75-90% product fill are recommendations rather than universal approval rules. Treating every recommendation as a policy requirement makes feeds needlessly rigid, while treating a policy rule as optional can lead to disapproval.
| Main image rule | Status | What it means in practice |
|---|---|---|
| Show the entire, correct product and variant | Requirement | Match the color, pattern, material, customization, and other visible variant details submitted in the feed. |
| Use minimal or no staging | Requirement | The main image must keep the sold product clear. Do not add unrelated products unless they are sold together. |
| Show every item in a bundle | Requirement | If the feed identifies the product as a bundle, the image must show the complete bundle. |
| Show one unit of a multipack | Best practice | Use the main image for one unit and additional images for other useful views. |
| No placeholders, generic graphics, or logo-only images | Requirement, with narrow category exceptions | The main image normally needs to be a real view of the product. Hardware, vehicles and parts, and computer software have limited exceptions in Google's specification. |
| No promotional overlays, watermarks, or borders | Requirement | Remove prices, calls to action, shipping messages, barcodes, retailer marks, and other added graphics. A brand mark that is physically part of the product is allowed. |
| Do not submit a thumbnail or merely scale up a small image | Requirement | Supply a genuine full-resolution source. Enlarging weak source pixels does not satisfy the quality rule. |
| Product occupies 75-90% of the frame | Best practice | Keep the item easy to inspect without cropping it or leaving excessive empty space. |
| Solid white or transparent background | Best practice | Google recommends either for broad compatibility, but staged images that clearly show the product are also accepted. Light products can render poorly on transparency. |
The safest main image is therefore accurate, unobstructed, high resolution, and simple. A clean white-background product image is often useful, but Google does not require white for every product. Apparel is also different: Google recommends showing clothing worn by a person, while shoes, bags, and other non-clothing accessories should generally appear alone in the main image and on a person in additional images.
How do additional_image_link and lifestyle_image_link differ?
Google accepts up to 10 additional product images and up to 5 lifestyle images, but the two fields serve different purposes and have different content rules. Together with one main image, a feed can submit up to 16 image URLs for one product, although Google does not promise to display every submitted image together.
| Attribute | Limit | Technical rules | Best use |
|---|---|---|---|
image_link | 1 | Main-image rules; required | The clearest accurate view of the sold product |
additional_image_link | Up to 10 | Follows main-image rules, except staging, in-use views, graphics, and illustrations are allowed | Other angles, detail views, scale, packaging, and product use |
lifestyle_image_link | Up to 5 | At least 600 x 600; Google's stated aspect-ratio range is 2:0 to 2:3; no promotional text, overlays, borders, or padding | The product worn, used, or presented in a real-world context on browsy free-listing surfaces |
Additional images are optional, but they let shoppers inspect details that do not belong in the main view. Google says multiple high-quality images are strongly correlated with engagement and clicks, yet quantity alone is not the goal. A second image should answer a question about the product, not repeat the hero with a tiny crop change. The official product-data specification defines the 10-image limit and the exceptions for staging and illustrations.
Lifestyle images are also optional. Google's dedicated lifestyle_image_link documentation says they may appear in free listings and can show clothing on a person, furniture in a room, a complete product set, or a product against a vivid setting. They still need to focus on the product and cannot contain sales copy, retailer logos, borders, or padding added to force the ratio. If you create contextual scenes with AI, use a workflow designed for placing a real product into a new scene and inspect the product closely before publishing.
Why does Merchant Center reject or warn about product images?
Merchant Center flags product images when the file, content, product match, or URL fails a documented check. Do not guess from a generic “image problem” label: open Products > Needs attention, filter by the image issue, and fix the underlying cause shown there.
| Issue | Likely cause | Fix |
|---|---|---|
| Image too small or upcoming-size warning | The image is below today's category minimum or below the 500 x 500 threshold due in 2027 | Replace it with a genuine 500 x 500 or larger source; aim around 1500 x 1500 or above. |
| Promotional overlay on image | Added text, price, shipping claim, watermark, retailer mark, barcode, or border | Export an unobstructed product image without the added element. |
| Low image quality | Blur, weak resolution, distracting treatment, or a thumbnail enlarged beyond its useful detail | Return to the best source file or create a new high-resolution image. The low-resolution product-image guide covers source-side fixes. |
| Generic or placeholder image | The main image is a stock substitute, illustration, logo, or “image coming soon” graphic | Submit the actual product, unless one of Google's narrow category exceptions applies. |
| Incorrect or incomplete product | The visible variant does not match the feed, the product is cropped, or bundle contents are missing | Match the exact offer and show the complete sold product. For AI images, also review the risk of misleading product details. |
| Unable to show or process image | Googlebot cannot crawl the URL, the URL redirects or returns a page, robots.txt blocks access, or the server returns invalid headers or errors | Serve a direct, public image URL with the correct file type and allow Googlebot and Googlebot-image. |
When you replace an image, publish it at a new, unique URL and update image_link. Google says a new URL is typically recrawled within about three days, while changing the bytes behind the same URL can take up to six weeks to detect. Existing approved images may continue to serve during review. Google's image-processing troubleshooting guide covers longer delays and server checks.
What AI metadata does Google require for Shopping product images?
Google allows generative-AI images in image_link, additional_image_link, and lifestyle_image_link, but requires images created with generative AI to carry IPTC metadata identifying that origin. The current policy specifically tells merchants to use the IPTC DigitalSourceType value TrainedAlgorithmicMedia and not to remove embedded source metadata during export, optimization, or delivery.
Google's AI-generated content policy also defines the following IPTC NewsCodes. These definitions describe the source type; they do not remove the policy's explicit instruction for generative-AI images.
IPTC DigitalSourceType value | Google's definition |
|---|---|
TrainedAlgorithmicMedia | Created with a model derived from sampled content |
CompositeSynthetic | A composite that includes synthetic elements |
AlgorithmicMedia | Created purely by an algorithm that is not based on sampled training data |
Check the file served from the final feed URL, not only the original export. Image editors, storefront optimizers, CDNs, and feed transformations can re-encode a file and strip metadata after it leaves your production workflow. The dedicated Help-Desk Article on whether Google Shopping accepts AI-generated product images covers this publishing check, while the broader guide to AI-generated product-image rules across platforms explains why one export is rarely safe for every channel.
Embedded IPTC metadata is separate from a visible disclosure. Google also provides AI label settings for AI-created or edited advertising assets and says those approved labels do not violate its usual overlay ban. Google cautions that using its label setting does not by itself guarantee compliance with applicable law. See Google's AI content label guidance for the current distinction.
Do better product images lower Google Shopping CPC or improve ranking?
High-quality product images can support Google Shopping visibility and click performance, but Google publishes no fixed CPC reduction for improving an image. Google's product-data guidance says high-quality images are more likely to appear for relevant queries, and its valuable-content guidance says multiple high-quality images are strongly correlated with engagement and clicks. Those are useful reasons to improve imagery, but they are not a promise that a particular edit will lower CPC by a set percentage.
Google Ads documentation is explicit: “Shopping campaigns don't use Quality Score.” Overall offer quality and relevance still factor into Shopping eligibility, but the hidden-Quality-Score formula often repeated in third-party guides does not apply. This distinction matters because it changes the recommendation from “improve images and assume savings” to “improve images, then measure what happened.”
Track impressions, clicks, and CTR in Merchant Center, segmented by ads and free listings where useful. Compare a defined group of changed products with a stable control group, and avoid changing titles, prices, bids, and images at the same time. Google provides the Merchant Center product-performance metrics; the Help-Desk Article on product photography and conversion evidence explains why image performance should be tested rather than assumed.
How should a brand plan a Google Shopping image set?
A useful Google Shopping image set assigns one shopper question to each image instead of filling every available field by quota. Start with an accurate main view, add inspection and use-case views through additional_image_link, then reserve lifestyle_image_link for genuine context.
| Image role | Feed attribute | Shopper question answered |
|---|---|---|
| Clean hero or on-model apparel view | image_link | What exactly is being sold? |
| Alternate angle | additional_image_link | What does the product look like from another side? |
| Detail view | additional_image_link | What are the material, finish, controls, or construction details? |
| Scale or fit view | additional_image_link | How large is it, or how does it sit on a person? |
| Packaging or included-parts view | additional_image_link | What arrives with the purchase? |
| In-use scene | additional_image_link or lifestyle_image_link | How is the product used? |
| Contextual brand scene | lifestyle_image_link | What does the product look like in a believable setting? |
Not every product needs every row. A simple item may need four strong views; a technical bundle may need several detail and contents images. For a wider channel plan, see how to take professional product photos and the guide to consistent catalog photography. If the same assets also go to Amazon, use the separate Amazon product-photography requirements guide because its main-image rules differ.
How can brands produce consistent Google Shopping images at catalog scale?
Google evaluates primary, additional, and lifestyle images under different field rules, while brands have to produce them as one catalog. The practical way to scale is to separate product identity from photographic direction: keep the item's defining evidence stable, then reuse the production choices that determine how it appears in each feed field.
Nightjar gives those two jobs separate places. A Product groups the real item's defining Product Photos, factual description, and physical dimensions. A Recipe saves the reusable Product Photography setup without saving the Product itself. For Google Shopping, one Recipe can preserve a flat-color background, product-only Framing, Shadow, square aspect ratio, resolution, and output format for listing views; another can carry a chosen Background and Photography Style into contextual images. Applying each setup to the next Product keeps the photographic direction stable, while the Product keeps each image tied to the item actually sold.
Nightjar's built-in visual review compares outputs with the request and reference images and can retry obvious failures at no extra Credit cost. That extra check matters when a substituted product, broken brand mark, or missing item would break the match between the image and feed.
Nightjar outputs JPEG, PNG, or WebP at 1K, 2K, or 4K where supported. Those settings choose the production size and format. Before submission, verify dimensions, file size, product accuracy, crawlability, and IPTC metadata on the file served from the final URL. The recommended ecommerce file-output settings are a useful starting point.
Google's own Product Studio is a free Merchant Center suite that can create images, change or remove backgrounds, increase resolution, and generate product videos. For repeated catalog production, Nightjar lets a Team reuse Products, visual controls, and Recipes before the files reach the feed.
Frequently Asked Questions
What size should Google Shopping images be in 2026? Use at least 500 x 500 now, even though the lower 100 x 100 non-apparel and 250 x 250 apparel limits remain enforceable until January 31, 2027. Google recommends around 1500 x 1500 or larger.
Does a Google Shopping main image have to be square or white? No fixed main-image aspect ratio is published, and white or transparent is a Google best practice rather than a universal requirement. The main image must still show the correct product clearly, without prohibited overlays or unrelated items; this white-background photography guide explains when a clean background is useful.
How many product images can I submit to Google Merchant Center? You can submit one main image, up to 10 additional images, and up to 5 lifestyle images. These are separate feed fields, and submission does not guarantee that all 16 images will appear together.
Does Google Shopping accept AI-generated product images? Google Shopping accepts AI-generated product images when they remain accurate, meet the normal image rules, and carry Google's required IPTC metadata. Read the focused answer on Google Shopping's AI-image policy before publishing.
Why did Merchant Center warn about an image that still meets the old minimum? Google is warning merchants about primary and additional images below 500 x 500 ahead of the January 31, 2027 enforcement date. Replace warned images with genuine higher-resolution sources rather than stretching a thumbnail.
Should I reuse the same image set for Google Shopping and Amazon? Only after checking both policies. A clean Google main image may also work on Amazon, but the platforms have different main-image and category rules; use the Amazon requirements guide before syndicating the same file.
Google Shopping image checklist
- Primary and additional images are at least 500 x 500 before the 2027 enforcement date.
- Working masters are around 1500 x 1500 or larger where the source supports it.
- Every file stays below 64 megapixels and 16 MB.
- File extensions match JPEG, WebP, PNG, non-animated GIF, BMP, or TIFF content.
- The main image shows the complete, correct product and visible variant.
- No added promotional text, watermarks, retailer marks, barcodes, or borders remain.
- Product fill is reviewed against Google's 75-90% recommendation.
- URLs are direct, stable, public, correctly encoded, and crawlable by Googlebot-image.
- AI-created files retain the required IPTC
DigitalSourceTypemetadata at the final served URL. - Additional and lifestyle images add a distinct angle, detail, scale cue, or context.
- Products with changed images use a new image URL for faster recrawling.
- Merchant Center's Needs attention view is clear after Google reprocesses the feed.
- Repeated catalog imagery follows a consistent background and lighting system.
References
- Google Merchant Center: 2026 product data specification update - 500 x 500 warnings and the January 31, 2027 enforcement date
- Google Merchant Center: Image link - Main image format, URL, content, and quality rules
- Google Merchant Center: Product data specification - Primary and additional image fields
- Google Merchant Center: Lifestyle image link - Lifestyle image limits and content rules
- Google Merchant Center: AI-generated content - IPTC metadata requirements
- Google Merchant Center: AI content labels - Visible AI label settings and disclosure caveat
- Google Merchant Center: Valuable content coverage - Image resolution, quantity, engagement, and clicks
- Google Merchant Center: Product data optimization - High-quality image and product-performance guidance
- Google Merchant Center: Image processing delays - Crawl timing and image-hosting checks
- Google Ads: Auction insights - Shopping campaign quality and eligibility
- Google Merchant Center: Product performance - Impressions, clicks, and CTR reporting
- Google Merchant Center: Product Studio - Current Product Studio capabilities
- Nightjar - AI product photography for repeat catalog production