If a try-on result doesn't look quite right, there's usually a simple reason and a few easy things you can adjust. Because try-on uses generative AI, results can vary from one image to the next, but the tips below reliably give your shoppers the best, most realistic previews.
Start with your product photos
The image the AI works from matters most, so this is the best place to begin.
Use clear, high-quality images: sharp, well-lit photos give the AI the most to work with.
Show the product on its own: a clean flat-lay, hanger, or plain-background shot reads more clearly than a heavily styled or layered photo.
Capture the whole item: make sure the full product is in frame, top to bottom.
Keep colours true to life: natural, even lighting helps the preview match the real product.
By default, try-on uses the first product image. If that photo shows a model in a full outfit, several items, or an angled shot, pick a better one on Virtual Try-On → Try-on reference photo. AI Frame can also suggest a stronger image from the product gallery. See Product Photo Tips for the Best Try-On Results.
Guidance to share with your shoppers
A big part of a good result comes down to the photo your shopper uploads. The widget already guides them, but it helps to know what works best:
A full-body photo in good lighting, with just them in the shot
A similar outfit type: clothing with a similar neckline and sleeve length to the item gives the most accurate result
A note on how AI try-on works
Try-on uses generative AI to create a preview, so results can occasionally vary or need a second attempt — that's normal. Trying again, or using a clearer photo, usually produces a better result. The tips above are designed to give your shoppers the best possible preview, most of the time.
Need a hand? If results still aren't looking right after these steps, send us a message, share the product and an example and we will gladly take a look with you!
