Google Search Console now lets site owners filter searches that combine images and text. The multimodal filter, highlighted in Yoast's October 2026 SEO update recap, helps publishers understand how people discover their content through visual searches that mix an image with a text query.
Multimodal search is a search that combines more than one input type: a photo plus a question, a screenshot plus a refinement, or a visual query with text constraints. Google has said one in every six searches is multimodal, which makes this a mainstream behaviour rather than an edge case.
Until now, Search Console reporting blended these searches with ordinary text searches, so a site could not tell how much of its discovery came through visual entry points. The new filter separates that traffic, making visual discovery measurable for the first time in Search Console.
With the filter, a site owner can see which queries, pages, and images drive multimodal discovery, and how that traffic behaves compared with text-only search. A page that performs well for image-led queries may deserve different optimisation than a page that wins on text alone.
The filter also helps diagnose image problems. If a product or article should be discoverable visually but shows no multimodal impressions, the likely culprits are missing or poor images, weak alt text, or pages where the image is not the clear subject of the content.
Start by comparing multimodal impressions and clicks against your overall Search Console performance for the same period. A high share of multimodal discovery on product, recipe, travel, or how-to content suggests visual optimisation will pay, while a low share on visually driven content suggests missed opportunity.
Then inspect the top multimodal queries and pages. Look for patterns: are certain image types, subjects, or page templates overrepresented? Those patterns are a brief for new visual content, better image metadata, and pages designed to answer the visual query's implied question.
The fundamentals are image quality, relevance, and context. Use original, high-resolution images that genuinely show the subject, write descriptive alt text and filenames, and surround the image with text that answers the question the image raises.
Structured data helps too. Product, recipe, and article markup gives Google machine-readable context for what an image depicts, which improves matching when the query is part visual. Treat image SEO as a first-class optimisation, not an afterthought to text SEO.
Visually discoverable pages earn a distinct kind of link: other writers embed or cite the image and its source page when they illustrate their own stories. A page that is the canonical visual source for a topic accumulates links from writers who need that image, not just that information.
Invest in original visuals for your most linkable pages: charts, diagrams, photos, and infographics that other writers cannot easily reproduce. Multimodal discovery will surface those visuals to more writers, and each surfacing is a chance to earn a link.
The multimodal filter covers searches that combine image and text, but visual discovery also flows through Google Lens, image search, and visual results inside ordinary search. A complete visual audit checks all of them, not just the new filter, and compares which pages win in each surface.
Also check that images are not blocked by robots rules, lazy-loading, or JavaScript that hides them from crawlers. A beautiful image that Googlebot cannot fetch is invisible to every visual surface, and the fix is usually technical rather than creative. Crawl a sample of key pages as Googlebot and confirm the images load. Repeat that crawl check after every template change, since redesigns are the most common cause of newly blocked images.
Sources: Yoast, October 2026 SEO update recap.
Linkbaits helps you create content that earns citations and backlinks. Start with the free tools and build your first linkable asset this week.
Start free trial