AI NewsGo-to-marketAnnouncement
Google Search Console added a filter for multimodal search, so a site can now see how much traffic came from a photo instead of typed words
Google Search Console now reports web multimodal Search performance separately, covering visits from Google Lens, Circle to Search on Android, uploaded images and the Chrome right-click image search. A new Multimodal filter sits inside the Web search type in the Performance report and rolled out globally today.

Image: Google Search Central
Why it mattersA site can now see, and export, the share of its search visits that come from a picture rather than typed words, so it can decide which pages need alt text and image content to stand a chance in that lane.
Search Console can now say what share of a site's Google visits came from a picture instead of typed words. On 24 September 2026, Google Search Central published a post announcing web multimodal Search performance reporting, and the rollout is described as global from today. The change adds a new filter inside the Performance report on Search results and inside the Generative AI features report.
The new filter is called Multimodal. It sits alongside Text-based under the existing Web search type. Google says the multimodal bucket covers visits that come from four surfaces: Google Lens, Circle to Search on Android, image uploads to Google Search, and the Chrome right-click "Search this image" feature. The data is exportable.
What is actually new
Search Console has reported clicks and impressions for typed queries for years. Until today, visits driven by a picture, whether a Lens scan on a phone or an image upload on desktop, sat inside the Web bucket with everything else, so they could not be counted or filtered on their own. The two paths behave differently, because the query on the Google side is an image rather than a string, and the ranking signals lean on what is in the picture and on the alt text, captions and structured data around it.
In Google's words: "This update is designed to give you insights into how your content is surfaced when users search using images (such as with a smartphone camera). The data reported here includes searches with Lens, Circle to Search on Android, image uploads to Google Search, and the Chrome right-click 'Search this image' feature."
The post also says the rollout is happening from today: "The integration of multimodal search data is rolling out globally starting today. You will start seeing these metrics in your Performance report if your site is receiving traffic from these queries."
What a site owner does with it
The Multimodal filter is under the Search type control at the top of the Performance report, next to the existing filters for Web, Image, Video and News. Selecting it narrows the report to visits that started from an image rather than a typed query. Clicking Export downloads the filtered data, so the multimodal slice can be compared against the typed slice for the same page or query group in a spreadsheet.
For a product catalogue, a plant identifier, a recipe or anything a shopper photographs before searching for it, the split is the point. A page that ranks well for typed queries can still be invisible to a Lens visit, because the picture-side signals (a clear product photograph, descriptive alt text, an image sitemap and product schema) carry a different weight in that lane. Reveneau's read is that the value of this report is a per-page baseline: a site owner can now measure how much multimodal traffic each page already earns, and decide which pages are worth improving picture-side before betting time on it.
The post is by Harsh Kharbanda, Product Manager Lead for Google Lens, and Moshe Samet, Product Manager Lead for Search Console.
Source
- Primary source: Announcing web multimodal Search performance reporting in Search Console, Google Search Central, 24 September 2026
This item was written by an AI system from the linked source. Reveneau is responsible for what it publishes.
Get AI News in your inbox
New developer tools, model and agent releases, and how teams are actually using them to release software. Short, and only when there is something worth reading.


