Quick answer: To show up when people search with a photo, your images must be easy for Google to find and understand. Use real HTML image tags, descriptive file names and alt text, original high-quality photos placed next to relevant text, and complete product and Business Profile data. Then check the new multimodal filter in Search Console to see how much Lens and Circle to Search traffic you already get.
- You can now measure photo searches. Since September 24, 2026, Search Console reports traffic from Google Lens, Circle to Search, image uploads and Chrome's "Search this image" as a separate multimodal search type.
- India is the biggest market for it. Google says more people use Lens in India every month than in any other country, and that India is its largest user base for visual search.
- There's no special trick. Google says its AI features are rooted in its core ranking systems. Good image SEO, indexable pages and complete product and business data are what matter.
- Google can't see some images. Google's own image guide says "Google doesn't index CSS images", so photos added as background images may be invisible to search.
What is multimodal search, and why should a business owner care?
Multimodal search means searching with more than typed words. A customer points their phone camera at a product, circles something on their screen, uploads a screenshot or right-clicks an image in Chrome, and Google finds matching results.
This is no longer a small habit. In October 2024, Google said Lens was being used for "nearly 20 billion visual searches every month", and that Lens queries were one of the fastest growing query types on Search. In December 2025, Google India said visual searches had increased 70% globally year over year.
For Indian businesses, the numbers are even more relevant. Google India says "more people use Google Lens here every month than in any other country worldwide." When it launched Search Live in AI Mode in India in October 2025, Google said people in India form its "largest user base for both voice and visual search globally." Search Live lets people talk to Search and use their camera to get real-time help, in English and Hindi.
So when a shopper in Jaipur photographs a block-print bedsheet, or a homeowner in Kochi snaps a broken tap fitting, Google has to decide which websites to show. If your images and pages are hard for Google to understand, you won't be one of them.
What did Google add to Search Console in September 2026?
On September 24, 2026, Google announced multimodal reporting in Search Console. It appears in the Performance report for Search results and in the report for Generative AI features. Google says the data covers searches made with Lens, Circle to Search on Android, image uploads to Google Search and Chrome's right-click "Search this image" option.
In the Performance report, the old "Web" search type is now split into two options: text-based and multimodal. Google's help page describes multimodal search as traffic "when users search using images (such as with a smartphone camera)."
Two details matter when you read the numbers:
- There's no query list for this traffic. As Search Engine Journal and Search Engine Roundtable reported from Google's help documentation, specific text query data isn't available because these searches mostly use images, not words. You'll see clicks, impressions and pages, but not what people typed.
- This is new data. Search Engine Roundtable reported Google's John Mueller confirming the data wasn't in the reports before. So you're not losing text traffic. You're seeing traffic that was previously hidden.
You can download the data with the Export button.
How does Google decide which website to show for a photo search?
Google hasn't published a separate rulebook for Lens rankings. What it has said is that its generative AI features "are rooted in our core Search ranking and quality systems." To be a supporting link in AI Overviews or AI Mode, Google says a page "must be indexed and eligible to be shown in Google Search with a snippet."
For images, Google's Image SEO guide explains where it gets its understanding: "Google extracts information about the subject matter of the image from the content of the page, including captions and image titles." It also uses the page title and meta tags.
In simple terms, Google looks at three things:
- Can it find the image? The image must be in a real HTML image tag on a page Google can crawl and index.
- Does it understand the image? File name, alt text, caption and the text around the image all describe it.
- Is the page a good answer? The page itself has to be useful, trustworthy and relevant to what the person is looking for, just like in normal search.
Google's July 2026 guide to its AI features also says "When it makes sense, look for ways to support your textual content with high-quality, relevant images and videos." Images help, but they work alongside good text, not instead of it.
How do I make my images easy for Google Lens to understand?
These steps come straight from Google's image guidelines, in the order we fix them for clients.
Step 1: Use real HTML image tags, not background images
Google's guide says "Using standard HTML image elements helps crawlers find and process images," and is blunt about the alternative: "Google doesn't index CSS images." Many website themes show product and banner photos as CSS backgrounds. Ask your developer to use <img> or <picture> tags for any photo you want found.
Step 2: Give every image a descriptive file name
Google recommends file names that are "short, but descriptive". IMG_4521.jpg tells Google nothing. jaipuri-block-print-double-bedsheet-indigo.jpg tells it exactly what's in the photo.
Step 3: Write alt text that describes the image
Google calls alt text "the most important attribute when it comes to providing more metadata for an image." Describe what's actually in the picture, the way you'd explain it to someone on the phone. Don't stuff keywords. Google warns this "results in a negative user experience and may cause your site to be seen as spam." The free Pagelens Accessibility Checker scans a page for missing alt text and other basic issues, with no signup.
Step 4: Place images next to the text that explains them
Google says to make sure "images are placed near relevant text and on pages that are relevant to the image subject matter." A product photo belongs on its product page, with the name, price and details right beside it, not only in a gallery page with no words.
Step 5: Use original, sharp photos
Google's guide notes that high-quality photos appeal to users more than blurry, unclear images. Original photos also help Google connect your image to your page. If you use the same supplier photo as fifty other sellers, Lens has fifty equal matches to choose from. Photograph your own products, shop, team and work.
Step 6: Help Google discover every image
Google supports BMP, GIF, JPEG, PNG, WebP, SVG and AVIF. Use responsive images with srcset or <picture> so phones load smaller files, and submit an image sitemap if Google might otherwise miss some images. Fast pages matter too, because most visual searches happen on mobile data. You can test speed with the free Pagelens Page Speed tool.
Image SEO checklist: what to fix first
Use this table to check your most important pages.
| Check | Weak setup | Better setup |
|---|---|---|
| How the image is added | CSS background image | HTML <img> or <picture> tag |
| File name | IMG_4521.jpg | steel-modular-kitchen-basket-2-tier.jpg |
| Alt text | Empty, or "kitchen basket buy best price India" | "Two-tier stainless steel pull-out basket fitted in a modular kitchen" |
| Placement | Gallery page with no text | Product or service page, next to the relevant description |
| Photo | Stock or supplier photo used by many sites | Your own sharp, well-lit photo |
| Product data | Price and stock only in the image | Product structured data and a Merchant Center feed |
| Business photos | Two old photos on Google Business Profile | Fresh, real photos of your shop, products and work |
| Indexing | Page blocked or not indexed | Page indexed and eligible for a snippet |
What extra steps help online stores and local businesses?
Google's AI guide says "Using products like Merchant Center and Google Business Profiles can help your products and services be visible in both AI responses and other Google Search results." Many visual searches are shopping or local, so these two matter.
For online stores: product data and Merchant Center
Add Product structured data to product pages. Google says it can make a page eligible to show as a product snippet with "ratings, review information, price, and availability." The free Pagelens Schema Generator creates the JSON-LD code for you to paste in.
If you list products in Google Merchant Center, follow its image rules. Google asks for images of at least 500 x 500 pixels and recommends around 1500 x 1500 pixels or above. It also says not to use images with promotional overlays, text, watermarks, logos or borders. An "Upto 50% OFF" banner on your product photo can hurt more than it helps.
For local businesses: Google Business Profile photos
If someone photographs your shopfront, signboard or a dish from your restaurant, your Business Profile photos help Google make the match. Google's photo guidelines ask for JPG or PNG files between 10 KB and 5 MB, recommend 720 x 720 pixels, and say photos should be in focus, well lit and without heavy filters. Add exterior photos so customers can recognise your business, plus interior, product and team photos. Our guide on getting your business ready for Google's AI calling covers the rest of your Business Profile.
Is this different from optimising for ChatGPT or Perplexity?
Partly. ChatGPT and Gemini apps also let people upload photos and ask questions. But when those tools look up businesses or products, they still lean on web pages they can find and read. Clear text beside your images, descriptive alt text and pages crawlers can access help across every AI tool, not just Google.
Don't buy into special "AI image markup" either. Google's July 2026 guide says "You don't need to create new machine readable files, AI text files, markup, or Markdown to appear in Google Search."
Content still decides who gets cited. The GEO research paper (KDD 2024) found that adding citations, quotations and statistics to content could raise visibility in generative engine responses by up to 40%. Our guide on how to get your business cited in AI search explains how to apply that.
How do I check whether visual search is bringing me customers?
Here's a simple monthly routine:
- Open Search Console's Performance report. Change the search type to multimodal and set the date range to the last 3 months. Note clicks and impressions.
- Look at the Pages tab. Since there are no queries, pages are your best clue. If product pages appear, Lens shoppers are finding you. If none do, start with the checklist above.
- Compare with text-based search. Check whether the same pages perform in normal search. A page that does well in text but never in multimodal search usually has an image problem.
- Test it yourself. Photograph one of your own products with Google Lens and see whose site appears. Then check whether Google's AI answers mention you for the questions customers ask, using the free Pagelens AI Overview Checker and the other free Pagelens tools for ChatGPT, Gemini and Perplexity.
- Fix indexing gaps first. If important pages aren't indexed, no image fix will help. Work through our indexing checklist for Google and AI search.
What results can AI visibility work bring?
Image SEO is one part of the AI visibility work we do, alongside technical fixes, content and brand mentions. Here's what that combined work delivered for two OffBuddy clients (share of tracked buyer questions where AI tools named the brand):
| Client | AI visibility before | AI visibility after | Time taken |
|---|---|---|---|
| iProledge | 3.8% | 24% | 3 months |
| Deepsolv | 0% | 6.8% | 2 months |
Results depend on your industry, competition and starting point, so treat these as examples, not promises. The bedsheet, kitchen basket and tap examples in this guide are illustrative.
Sources
- Google Search Central Blog: Announcing web multimodal Search performance reporting in Search Console (September 24, 2026)
- Google Search Console Help: Performance report (Search results)
- Google Search Central: Image SEO best practices
- Google Search Central: Optimizing for generative AI features in Google Search (updated July 2026)
- Google Search Central: AI features and your website
- Google Search Central: Product snippet structured data
- Google Merchant Center Help: Image link requirements
- Google Business Profile Help: Photo and video guidelines
- Google Blog: New ways to search with Lens and AI-organized results (October 2024)
- Google India Blog: India's Year in Search 2025, new ways to search (December 2025)
- Google India Blog: New languages in AI Mode and Search Live debuts (October 2025)
- Search Engine Journal: Google Search Console adds multimodal filter for image-based searches
- Search Engine Roundtable: Search Console Performance report gains multimodal search type filter
- Aggarwal et al.: GEO: Generative Engine Optimization (KDD 2024)