local seo
Google rolled out visual-search reporting in Search Console last September. Now you can see which of your images people are finding through Google Lens, Circle to Search, and other visual-discovery tools.
For a roofing contractor, that might be photos of completed jobs. For a nonprofit, event images or program highlights. For a landscaper, before-and-after shots.
The question: Are people actually finding you this way? And if not, what needs to change?
Most local businesses think about search as typed queries. Someone searches “HVAC repair near me” and clicks a result.
Visual search works differently. Someone points their phone at a neighbor’s new deck, circles it, and asks “Who builds this?” Or they snap a photo of a cool storefront sign and search for similar work.
If your project photos are optimized, you show up. If not, someone else does.
The new Search Console reports tell you which images are pulling traffic and which ones are invisible.
Navigate to Search Console, then Performance, then the Search Appearance filter. You’ll see options for “Google Lens” and “Circle to Search” alongside the standard web-search views.
The data breaks down:
It’s the same structure as regular search reporting, just filtered for visual discovery.
Standard monthly reporting shows trends. Traffic up, traffic down, which pages performed.
That’s useful. But it’s not enough.
We’ve learned to run a separate diagnostic layer on top of routine reporting. Instead of summarizing what happened, we interrogate the entire digital surface to find what’s broken or missing.
For visual search, that means:
Pull Search Console data for image impressions and clicks. Then cross-reference with your analytics platform to see which images actually convert visitors into leads or donations. Add a manual crawl of your image library to identify photos that should rank but don’t appear in the reports at all.
You’re combining three data sources—console, analytics, live audit—to build a complete picture. No single tool shows the gaps.
Some patterns show up consistently when we dig into visual-search performance:
High-quality project photos get impressions but zero clicks. The images show up in search, but something about the context or metadata doesn’t compel people to visit. Usually it’s generic file names or missing alt text that fails to communicate what the photo actually shows.
Event photos from nonprofits rank well but send traffic to outdated pages. The image performs, but the linked page is from last year’s event or a general “programs” page instead of current information. People bounce immediately.
Service pages with stock photography get ignored entirely. Meanwhile, a single authentic job-site photo on a blog post pulls steady visual-search traffic. Authenticity matters more than polish.
We maintain two separate workflows now.
The reporting pipeline generates client-facing dashboards. Traffic trends, goal completions, top pages. Clean summaries that show whether things are moving in the right direction.
The interrogation workflow questions everything. Why did this image stop getting impressions? What’s missing from pages that should rank but don’t? Which photos need better metadata or different context?
Same data sources. Different questions. Different outputs.
Reporting tells you what happened. Interrogation tells you what to fix.
Start with the Search Console visual-search filter. Sort by impressions to see which images are already getting discovered.
Look for patterns:
Then pull your analytics data. Which visual-search visitors actually convert? That tells you which types of images to prioritize.
Finally, audit your image library manually. What should be showing up in visual search but isn’t? Those are your optimization targets.
When we run these deeper diagnostic passes, we flag them separately from standard monitoring.
It’s not part of the monthly report cycle. It’s a distinct process with its own documentation and scope.
That clarity matters. Teams need to understand when we’re summarizing performance versus when we’re tearing apart the entire site to find improvement opportunities.
Both are valuable. They’re just different jobs.
If you’re heading into planning season, visual-search data gives you something concrete to work with.
You’re not guessing which images matter. You’re looking at actual discovery patterns.
Optimize the photos that already get impressions. Fix the landing pages for images that generate clicks. Add better metadata to project photos that should rank but don’t.
It’s not complicated. Just methodical.
The data’s there now. Might as well use it.
Let's have a real conversation about what's getting in the way — and how we fix it.
Schedule a call →