How to Check If Your Images Rank in Multimodal Search

Google Search Console's new "Multimodal" filter is the fastest way to find out whether your images surface in Google Lens, Circle to Search and image-upload queries — and most site owners haven't looked at it yet.

That filter sits under "Search type" in the Performance report, and it's the diagnostic starting point Helen Pollitt outlines in her Search Engine Journal column on multimodal image optimization. Multimodal search, she explains, covers inputs well beyond keywords: photos via Google Lens, voice, and hybrid queries such as photographing a green dress and asking Google for it in blue.

The stakes have shifted. "Whereas in years past, images were just an augmentation of a page's content, now they are an important entry point in a search journey," Pollitt writes.

Are your images findable at all?

Start in Google Search Console. Filter the "Search results" report under Performance by "Search type: Image" to see which pages earn impressions and clicks from image search. No impressions doesn't automatically mean an indexing problem, but it warrants investigation. Use "Inspect any URL" on the page hosting the image, and run a live test on the image URL itself — though URL Inspection can't reliably confirm individual image indexing.

Then test manually. Take a photo of your own product and search with it via Lens or Circle to Search to see whether Google connects your picture with your company's product shots. You can also use the src: operator in Google Images, typing src: followed by the image's own URL, to find pages referencing it.

What's blocking the crawlers?

Check server log files for Googlebot-Image activity, and run a crawl in Screaming Frog with the user-agent set to "Googlebot Image" to catch 3XX, 4XX and 5XX responses. Pollitt lists the usual technical suspects:

• robots.txt rules blocking image URLs
• x-robots headers or the
• CDN or botwall restrictions, especially on separate subdomains like media.example.com
• JavaScript-dependent images, or images set as CSS backgrounds
• lazy loading that only fires on scroll — Googlebot doesn't scroll or swipe
• broken or redirecting image URLs, and session parameters creating duplicate URLs for the same image

Use Google Search Console's Inspect tool on the rendered HTML: images should appear as elements with valid URLs in the src attribute, loadable without JavaScript.

Can the bots actually understand the image?

Discoverability is only half the job. Bots read surrounding context — titles, alt attributes and on-page copy — to interpret what an image shows. Structured data adds precision: schema.org's ImageObject communicates licensing and other metadata, making images eligible for the "licensable" badge in results, while primaryImageOfPage markup tells Google which image leads the page.

Supported formats matter too: JPEG, PNG, WebP and AVIF. Use

One nuance deserves attention. Google's image metadata guidelines tell publishers: "Make sure people can access and view your pages that contain images without needing an account or logging in." Paywall structured data protects gated text, but it may not extend the same protection to images.

Pollitt's closing checklist: verify image visibility in Search Console and Bing Webmaster Tools, confirm crawler access and HTTP responses, review JavaScript and lazy loading, test CDN bot restrictions, check image sitemaps, and audit page context. Her bottom line is blunt — the foundations of multimodal optimization are the same technical standards image SEO has always demanded, and sites that skipped them now have AI search bots to worry about too.

By James Calloway, Marketing Herald