Image Sitemap

An image sitemap is an XML file that tells Google about images it might otherwise miss, using <image:image> tags attached to the pages those images live on. It matters most for images loaded by JavaScript, served from a CDN on a different host, or displayed via CSS backgrounds, since a crawler often cannot see those on its own.

Unlike video or news sitemaps, the image extension is deliberately thin. Modern Google guidance is that the only tag that still carries weight is <image:loc>, the image URL itself. The older caption, title, geo, and license tags were deprecated years ago and are ignored, so do not waste time populating them.

What a real image sitemap looks like

One page can carry multiple images. You nest several <image:image> blocks under a single <url>, up to 1,000 images per page entry.

<?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://example.com/products/trail-boot</loc>
    <image:image>
      <image:loc>https://cdn.example.com/img/trail-boot-side.jpg</image:loc>
    </image:image>
    <image:image>
      <image:loc>https://cdn.example.com/img/trail-boot-sole.jpg</image:loc>
    </image:image>
  </url>
</urlset>

Image sitemap tags: what still works

Limits are the same as any sitemap file: up to 50,000 URL entries and 50 MB uncompressed per file, plus the 1,000-images-per-page cap noted above. Blow past those and split into multiple files behind a sitemap index.

When an image sitemap is actually worth it

Be honest about whether you need one. For a plain content site where every image sits in a server-rendered <img> tag, Google finds them during a normal crawl and the sitemap adds little. The file starts paying off in specific situations: single-page apps and heavily JavaScript-driven galleries where images are injected client-side, product catalogs serving images from a separate CDN host, sites leaning on CSS background-image for hero visuals, and lazy-loading setups where images only load on scroll. In all four, a crawler can visit the page and still come away thinking there are no images on it.

The other honest case is speed. On a large site pushing thousands of new product images a month, an image sitemap gives Google a tidy, direct list rather than making it re-discover everything through rendering. That is a crawl-efficiency win, not a ranking one, but on a big catalog the efficiency compounds.

How to check it on your own site

1. Open the sitemap URL and confirm it returns valid XML with the xmlns:image namespace on the <urlset> line.
2. Copy a few image:loc URLs and load them directly in a browser. Every one must return a live image with a 200 status, not a redirect or a 404.
3. Confirm each image is referenced from the page in its parent <loc>. Google wants the image to genuinely appear on that page.
4. Submit the sitemap in Google Search Console and check the Sitemaps report for a Success status.
5. Spot-check indexing by searching site:example.com in Google Images, or inspect a page with the URL Inspection tool and view the rendered HTML to confirm the image loads.

Common mistakes and how to fix them

• Listing images not on the page. The image in image:loc should actually appear on the URL in the parent <loc>. Orphan listings get ignored.
• Skipping the sitemap because you rely on alt text. Alt text helps, but it cannot expose an image the crawler never sees. Lazy-loaded and CSS-background images are exactly why this file exists.
• Blocking the image host in robots.txt. If your CDN path is disallowed, Google cannot fetch the images no matter how clean the sitemap is. Check your robots rules against the image:loc domain.
• Padding it with deprecated tags. Caption, title, and geo tags do nothing now. They add bytes and a false sense of completeness.
• Forgetting the sitemap after a redesign. Rename image paths and every image:loc 404s. Regenerate the file whenever image URLs change.

FAQ

Do I still need an image sitemap if my images are in the regular HTML?

Often not. If images sit in plain <img> tags in server-rendered HTML, Google usually finds them during a normal crawl. The sitemap earns its keep when images are lazy-loaded, injected by JavaScript, or served as CSS backgrounds.

Can I put images and pages in the same sitemap?

Yes. Add the image namespace to your existing sitemap and nest <image:image> blocks under the relevant <url> entries. A separate image sitemap is a convenience, not a requirement.

Why are the caption and title tags in my generator if Google ignores them?

Legacy support. Many plugins still emit those tags because the schema technically allows them. They do no harm, but they do no good either. Alt text and on-page context are where image relevance actually comes from now.

How many images can one page entry hold?

Up to 1,000 <image:image> blocks under a single <url>. The file itself still obeys the 50,000-entry and 50 MB limits, so very large galleries may need a sitemap index.

How is this different from a video or news sitemap?

Same base XML, different namespace. A video sitemap uses <video:video> with rich metadata, and a news sitemap uses <news:news> with strict recency rules. The image extension is the leanest of the three.

Related: XML Sitemap, Image SEO: The Complete Guide, 4XX Error URL in XML Sitemaps