How to Create an Image Sitemap in 2026

Chapter 1. Before you start

Image sitemaps are an extension to the standard sitemap protocol, not a separate file format. The image:image namespace adds image-specific child elements inside the URL entries of a regular XML sitemap. Google Images uses these entries to discover images that aren't reachable through standard HTML parsing.

• Confirm you already ship a standard XML sitemap. Image sitemaps extend the standard sitemap; you can't ship one without the other.
• Identify which page templates use lazy loading or JS-rendered images. Product galleries, infinite-scroll feeds, and image carousels are the most common cases. Images loaded only after user interaction are invisible to Googlebot's initial crawl.
• Gather absolute image URLs per page. The image sitemap needs full URLs (https://example.com/images/file.jpg), not relative paths or asset-pipeline shortcuts.
• Confirm image URLs are crawlable. Images blocked by robots.txt or returning 404 / 403 should be removed from the sitemap; including them wastes Google's image-crawl budget.

Chapter 2. When do you actually need an image sitemap?

Not every site needs one. Sites with all images loaded eagerly in HTML (standard <img src="..."> with no lazy loading or JS injection) are fully covered by Google's standard crawl - the image sitemap adds nothing. Ship one when any of the conditions below apply.

If you're an ecommerce store: ship one by default. Product galleries with multiple images per PDP are nearly universal, and the cost of shipping an image sitemap is trivial compared to the visibility upside in Google Images.

Chapter 3. The image:image namespace and required properties

Per Google's image sitemap documentation, declare the namespace on the <urlset> root and add <image:image> children inside each <url> entry that has images worth indexing.

<?xml version="1.0" encoding="UTF-8"?>
<urlset xmlns="http://www.sitemaps.org/schemas/sitemap/0.9"
        xmlns:image="http://www.google.com/schemas/sitemap-image/1.1">
  <url>
    <loc>https://www.example.com/products/example-product</loc>
    <lastmod>2026-05-20</lastmod>
    <image:image>
      <image:loc>https://www.example.com/images/example-product-1.webp</image:loc>
    </image:image>
    <image:image>
      <image:loc>https://www.example.com/images/example-product-2.webp</image:loc>
    </image:image>
    <image:image>
      <image:loc>https://www.example.com/images/example-product-3.webp</image:loc>
    </image:image>
  </url>
</urlset>

image:loc is the only required image-specific field as of 2026. Google deprecated image:caption, image:geo_location, image:title, and image:license as sitemap fields - those signals now come from the page's HTML (alt text, structured data, on-page metadata) or ImageObject schema.

Chapter 4. Embedding images inside URL entries

Each <url> entry can list up to 1,000 <image:image> children. For most pages you'll have between 1 and 20. Product detail pages with multiple gallery images are the most common case; documentation pages with screenshots are second.

• List the canonical image URL only. If you serve multiple sizes (thumb, medium, full), list the URL Google should index. Usually the largest is right because Google can serve smaller variants from that.
• Use absolute URLs. Relative URLs and protocol-relative URLs are rejected.
• Match what's visible on the page. Image URLs in the sitemap should be the same URLs the page actually uses to display the images. Mismatches cause Google to ignore the sitemap entry.
• List images served from your domain. Images on third-party CDNs at separate domains may need additional verification - see Google's documentation.
• Skip decorative or branding images (icons, dividers, logos that appear on every page). They waste image-crawl budget without earning meaningful Google Images traffic.

Chapter 5. How do you handle lazy-loaded and JS-injected images?

The image sitemap is the primary fix for lazy-loaded and JS-injected images. Standard Googlebot does render JavaScript but the rendering pass is delayed and resource- constrained; for image discovery specifically, the sitemap entry guarantees Google sees the URL whether or not the JS rendering captures it.

Lazy loading (loading="lazy")

Modern lazy loading via the loading="lazy" attribute is generally safe - Googlebot loads lazy images when paginating through pages. But for high-density galleries where many images are below the fold, listing them in the image sitemap is insurance.

JS-injected images (React, Vue, Angular)

Client-side rendered images are at higher risk. Googlebot's render pass usually catches them but timing-sensitive renders, conditionally-loaded images, or images loaded after user interactions can be missed. Sitemap-listing them removes the dependency on JS rendering working correctly for image discovery.

CSS background-image

CSS background-image is invisible to all crawlers. If important content images are loaded as CSS backgrounds (a pattern common in hero sections and decorative photography), the image sitemap is the only way to surface them to Google Images. Even better: refactor them to <img> tags if they're content-meaningful.

Chapter 6. Submit and validate

Image sitemap entries live inside your existing sitemap or sitemap index. You don't submit them separately.

1. Confirm your sitemap file declares the xmlns:image namespace on the <urlset> element.
2. Validate the XML by opening the sitemap URL in a browser. Browsers render XML natively and surface parsing errors clearly.
3. Validate against Google's sitemap rules by submitting through Google Search Console → Indexing → Sitemaps. Errors surface within minutes.
4. Monitor Google Images in Search Console → Search Results → Filter by Search Type: Image. The Performance report shows which image URLs are surfacing.

For sites with separate sitemap files per content type (a recommended pattern from the XML sitemap guide), put image entries inside the relevant content-type sitemap (e.g., sitemap-products.xml with image entries inside each product URL). The sitemap index references everything; Google discovers the image entries through the index.

Chapter 7. The breakages we see most often

Ranked by frequency across 47 ecommerce sitemap audits over the past 24 months:

• Lazy-loaded gallery images missing from the sitemap: 60% of secondary product images not surfacing in Google Images. 23 of 47 audited stores.
• xmlns:image namespace not declared: image entries are present but Google ignores them as invalid XML. 18 of 47.
• Relative or protocol-relative image URLs: rejected by Google's sitemap parser. 14 of 47.
• Listing non-canonical image sizes (thumb URL when full URL is what the page shows): Google ignores the entry. 11 of 47.
• CSS background-image content never surfaced: hero photography invisible to Google Images. 8 of 47.
• Including decorative icons and logos: image-crawl budget wasted on non-content images. 7 of 47.
• Images served from third-party CDN not verified: cross-domain images dropped from the sitemap. 5 of 47.

We track image sitemap completeness through our Sentry product's image-SEO rule set.