Google Just Turned Camera Search Into an Optimization Channel
The opportunity: sell ecommerce brands a $500–$1,500 visual-search photography audit now, then turn the workflow into software once Google opens API access to automate it.
On September 24, 2026, Google gave merchants a new marketing channel without launching a new product.
Google Lens has existed for years, and Circle to Search ships on Android. People already photograph lamps, sneakers, plants, car parts and furniture because they don't know what words to type. What changed on September 24 is that merchants can finally see some of that behavior.
Search Console's Performance report now splits Web traffic into two search types: text-based and multimodal. The multimodal bucket covers Google Lens, Circle to Search on Android, images uploaded to Google Search and Chrome's right-click "Search this image." Google's John Mueller clarified that this is new data that was never included in previous performance reporting. For most stores, a slice of traffic that used to be invisible now shows up in the dashboard.
With it comes the first clean version of a loop merchants have never been able to run:
camera search → product page → sale → image change → measurement → repeat
Whenever a platform exposes a new measurement surface, a short window opens where everyone has the data and almost nobody has a process for acting on it. The heist is to become the person standing next to the ecommerce manager during that window, saying: these products are already being found by camera, these similar products aren't, and here are the photographs worth creating next.
Here's the business in brief:
The money: Ten $950 audits a month is $9,500, and 20 agencies on a $399 workspace add $7,980 MRR. Google Lens already fields roughly five billion commercial-intent searches a month.
Inside:
• Four-part audit with $500–$1,500 pricing
• Gap-scoring formula for reshoot backlogs
• Cold email template and agency white-label
• Six-week MVP and the matched-test moat
The behavior is already enormous
Nobody has to invent this demand. In October 2024, Google said Lens handled nearly 20 billion visual searches a month, with 20% of them shopping-related. On March 12, 2026, Google's VP of Global Advertising, Dan Taylor, put the number above 25 billion a month, with 20% carrying commercial intent. Twenty percent of 25 billion is roughly five billion monthly camera searches from people who want to buy something, and Google already sells Shopping ads beside Lens results.
The intent behind those searches differs from typed search. Someone who types "brass mid-century wall sconce" knows what the object is called. Someone who photographs a light fixture in a restaurant may not know "sconce," "aged brass" or the model number; they just want that thing. The advantage shifts toward catalogs where the look of a product matters as much as its name: lighting, furniture, décor, tools, replacement parts, collectibles, fashion and plants.
Google already tells merchants that product photography matters here. Product structured data makes listings eligible for shopping experiences in Google Images and Lens, and Merchant Center accepts up to 10 additional images per product, with Google encouraging different angles and in-use shots. What merchants have never had is a way to connect their own visual-search performance to their own photography decisions. Half of that connection now exists. The other half, visual search optimization tied to what the studio shoots, is a service business.
The catch: the data supports a service before it supports SaaS
Search Console tells a merchant which pages earn multimodal impressions and clicks, broken out by country, device and date. It doesn't show the photo the shopper took, the object they circled or any text query. Google says plainly that because these searches mostly use images, query data isn't available.
The API is the bigger problem for a founder. The Search Analytics API's `type` parameter still accepts web, image, video, news, Discover and Google News. "Multimodal" appears nowhere, and it isn't a search-appearance value either, so every ecommerce SEO tool that reads Search Console through the API gets text and camera traffic blended together. The only way to get the multimodal slice today is the Export button in the interface.
The history is also short. Early reports show multimodal data starting around September 10, 2026, and Google hasn't confirmed a start date. As of September 29, a merchant has a few weeks of camera-search data, and Search Console only retains 16 months of anything.
So the obvious first product is off the table for now: connect Search Console, connect Shopify, get automated visual-search recommendations every Monday. Nobody can build that cleanly today. The entry point is a productized service that works from merchant exports, sells a baseline now, and sells the follow-up read when the data matures.
The product is a gap detector
Image search itself is commoditized. Shopify already has customer-facing visual search apps priced like impulse buys. BIRSE runs from free to $50 a month. Snapseek starts at $9.90 and tops out at $299.90 a month for 100,000 products, with revenue attribution arriving at its $99.90 tier. Pitch a merchant on "we use AI to understand your images" and you're competing with a $9.90 plugin.
Those tools help shoppers search a store. None of them answers the question a merchandising team actually has: where should the next $2,000 of photography budget go? Framed that way, visual similarity becomes plumbing and the product becomes resource allocation. The merchant who pays for it is the one about to book a studio day.
How the engine works
Unlock the Vault.
Join founders who spot opportunities ahead of the crowd. Actionable insights. Zero fluff.
“Intelligent, bold, minus the pretense.”
“Like discovering the cheat codes of the startup world.”
“SH is off-Broadway for founders — weird, sharp, and ahead of the curve.”
