How ecommerce brands can use Google Virtual Try-On demand to build on-brand virtual try-on product imagery and shoppable campaign creative
Use Google Virtual Try-On demand to improve Shopping discovery, then create owned on-brand try-on imagery for PDPs, paid social, and campaigns.
Lamina Team
Product Team @ Lamina
What is Google Virtual Try-On for ecommerce brands?
Google Virtual Try-On is a Google-hosted shopping feature that lets eligible shoppers see how an apparel item might look on them, using a merchant’s product image and the shopper’s uploaded photo. It sits in Google Shopping, Search, and Images at the discovery and consideration stage. It is not a merchant-controlled studio for producing campaigns.
That split matters to ecommerce teams. Merchant Center imagery may help an item qualify for try-on exposure, while your PDP hero, paid-social creative, email modules, and display units still require assets you own, approve, and resize for each placement. Google also says the visualization is not an exact representation of fit or appearance. Never let product copy turn a try-on image into a sizing promise.
| Metric | Value | Source |
|---|---|---|
| Recommended source-image resolution for merchant apparel imagery | 1024 pixels or higher | support.google.comas of 2026-08-02 |
| Supported apparel categories in Google’s merchant try-on experience | 4: shoes, tops, bottoms and dresses | support.google.comas of 2026-08-02 |
| Add-to-cart rate reported for sessions using try-on in one vendor observational dataset | 29.2% | genlook.appas of 2026-07-17 |
| Try-on usage occurring on phones in that same vendor observational dataset | 90% | genlook.appas of 2026-07-17 |
How does Google Shopping virtual try-on work?
Google Shopping virtual try-on begins when a logged-in shopper finds an eligible item, taps “Try it on,” and uploads an image they have permission to use. Google’s AI then combines that photo with the merchant’s product image, producing a visualization of the garment on the shopper.
There are real guardrails. Google limits merchant eligibility and shopper availability by category, market, age, login state, and relevant personalization settings; lingerie, swimwear, and accessories are excluded from the listed merchant experience. Under the earlier model-based version, generative AI put Merchant Center garment images on professionally photographed models across body types. Eligible products could carry a try-on badge in free listings and Shopping ads.
Google has cut photo friction for some U.S. shoppers with a studio-like, full-body digital version generated from a selfie and their selected usual clothing size. That could widen participation. It still does not hand the retailer a downloadable set of branded campaign images.
How should brands prepare product imagery for Google Virtual Try-On?
Treat the Merchant Center hero image as product infrastructure. Submit a clean, complete, high-resolution image that lets Google’s system read the exact garment; Google recommends at least 512 × 512 pixels and prefers 1024 pixels or more. Show one complete listed garment, either front-facing on a simply posed model or mannequin, or laid flat.
Clear the clutter first. Hands, bags, accessories, cropped garment edges, and heavy wrinkles in flat lays can hide the details the system needs to render, making cleanup a feed-quality job rather than cosmetic polish. Keep one canonical, product-accurate Shopping source asset so every downstream variation begins with the same verified color, silhouette, trim, and logo treatment.
Start with impression volume and apparel eligibility. Replacing weak hero images for high-impression tops, bottoms, dresses, and shoes is a controlled way to expand VTO-ready coverage without reworking the entire catalog.
Our new AI Mode experience is built for every part of shopping — from finding inspiration to buying at the right moment. Plus, our virtual try-on tool now works with your own photos.
Can ecommerce brands create their own on-brand virtual try-on campaign creative?
Yes. Brands can make owned virtual try-on and on-model campaign creative, though that runs as a separate workflow from Google Shopping’s consumer try-on feature. Build from approved product references, approved model or talent rights, defined body-size representation, poses, styling, lighting, background rules, logo rules, disclosure requirements, and channel export ratios.
Control comes back to the brand here. A VTO-ready feed image serves accurate product discovery; a 1:1 paid-social unit, 1.91:1 display banner, and PDP module each need composition and copy space built for that placement. Start with an approved product/model output. Then make channel variants instead of forcing one catalog frame to do every job.
Google Cloud’s Breuninger example shows a workable progression: the retailer first used a VTO API to dress professional models in different outfits, adding variety to user tests, then moved toward selfie-based try-on. The lesson is operational, not vendor-specific. Get product fidelity and art direction in place before you widen the experience.
Now if you don’t have a full body photo of yourself you can use a selfie and Nano Banana, our Gemini 2.5 Flash Image model, will generate a full body digital version of you for virtual try on.
How to turn Google VTO demand into shoppable brand creative
Audit the eligible assortment
Pull high-impression apparel SKUs from Shopping, then separate shoes, tops, bottoms, and dresses from unsupported categories. For every priority SKU, check that the garment is complete, the silhouette is clear, the color is accurate, and the source image is 1024 pixels or higher where possible.
Protect a canonical product asset
Approve one product-accurate image per SKU for Merchant Center and internal generation. Before it goes live, verify garment construction, labels, prints, hardware, logos, and color; that reference keeps channel variants from drifting away from the item the shopper receives.
Write a production brief for owned try-on assets
Specify approved model or talent rights, size range, pose, styling, environment, lighting, color treatment, logo and text placement, required disclosures, and output ratios. A vague prompt gets vague creative. A bounded brief gives reviewers something they can actually check.
Generate variants after product approval
Create PDP, social, display, and email versions from approved product and model inputs. Google’s advertiser guidance describes making on-brand variations by editing an uploaded asset, including background swaps and seasonal additions. Review every result for garment fidelity, anatomy, labels, claims, and landing-page consistency.
Run a measured activation test
Test VTO-ready coverage and Shopping CTR at the catalog layer, then test PDP hero imagery and channel-specific creative against existing controls. Track add-to-cart, conversion, return reasons, production time, and approved-asset reuse. That separates a good-looking image from one that earns its keep commercially.
A practical two-track operating model for an ecommerce apparel team responding to Google Virtual Try-On demand.
Google Shopping asset role
over Before priority-SKU feed activation
Campaign asset ownership
over During campaign production
Performance measurement
over During campaign evaluation
Does Google Virtual Try-On increase ecommerce conversion rates?
Google Virtual Try-On may improve shopper consideration, though ecommerce brands should not promise a universal conversion lift without their own controlled test. A vendor-reported dataset across 577 Shopify stores linked try-on sessions to a 29.2% add-to-cart rate, versus 9.4% without try-on. Those shoppers may already be more ready to buy, so the comparison does not establish causation.
The phone-heavy usage signal gives you a practical placement call: if try-on interest clusters on mobile, put the test where shoppers can act right away—on a mobile PDP, in a vertical social unit, or in a Shopping flow that reaches the product page cleanly. Measure the whole chain, returns and return reasons included. A visualization can win clicks, create fit expectations, and still hurt the account.
Use a SKU- and audience-specific holdout. Compare VTO-ready or on-brand, on-model treatments against existing imagery, hold price and offer constant, and review results by category rather than rolling dresses, shoes, and tops into one verdict.
Now people can try billions of clothing products on themselves virtually. Early results and engagement have been extremely positive, particularly with Gen Z users, and we’ll be bringing this functionality to all U.S. users imminently.
What guardrails keep virtual try-on creative trustworthy?
Virtual try-on creative stays trustworthy when presented as a product visualization, rather than proof of exact size, fit, or appearance. Google says its fashion model is designed to understand garments and bodies, including drape, stretch, and folds. Google also makes clear that the output is not exact.
Put human review at the last gate. Reviewers need to check generated imagery against the approved product reference for logos, labels, fabric detail, color, anatomy, crop, and every claim shown alongside the product; brand-critical hero moments need a closer pass. You can produce a large asset family fast. A published asset still needs an accountable owner.
That discipline keeps each system in its lane: Google captures demand around eligible product listings, while your creative workflow turns approved inputs into shoppable, on-brand assets that can run across owned channels.
Continue reading
Virtual try-on for ecommerce: turning digital dressing rooms into on-brand product and campaign creative
Virtual try-on can help shoppers evaluate products on a PDP and give brands a reusable, on-brand creative mechanic for campaigns, email and events.
Lamina Team
Product Team @ Lamina
Data report: AI virtual try-on for ecommerce—what it is, how Google Shopping virtual try-on differs from merchant-ready try-on creative, and the product-accuracy, brand-control, and cost criteria brands should benchmark before choosing a tool.
AI virtual try-on can improve visual product discovery, but Google Shopping Try-On and merchant-owned try-on solve different jobs. Benchmark fidelity, control, coverage and total cost before you buy.
Lamina Team
Product Team @ Lamina
How fashion brands can launch on-brand AI virtual try-on that helps shoppers visualize garments accurately—and turn the resulting looks into product-page, social, and campaign creative.
Launch AI virtual try-on as a visual-confidence layer, then turn approved looks into governed PDP, social, and campaign assets without promising exact fit.
Lamina Team
Product Team @ Lamina