Shop the look beyond fashion: which categories actually work
“Complete the look” sounds like an apparel term, and it isn’t — but the thing that decides whether it works for your store is not your category. It’s your photography. The detector only reads a photo as an outfit when a real person is wearing or carrying the items, and that single rule explains every case where it works and every case where it doesn’t.
The rule that actually decides
A look is built from a photo of a person. When the app analyses your images it looks for a real model wearing or carrying the products; a flat lay, a packshot, or a product sitting alone on a plain background is skipped and returns nothing at all.
That skip is deliberate, not a gap. Given a lone product on white, an image model will happily “detect” the garment’s own straps, trim and shadow as separate items and assemble a nonsense outfit out of one product. Returning nothing is the honest answer, so that is what it does.
So the useful question isn’t “is my category supported?” — it’s “does my photography put a person in the frame?”
Beauty: on a model, not on marble
A makeup look photographed on a model works exactly like an outfit: the detector reads what the person is wearing and matches each product to your catalogue.
The routine flat lay — cleanser, serum and moisturiser arranged on a marble slab — does not. It is the single most common mismatch in beauty, because it is also the most common way beauty brands shoot. If your PDP hero is the flat lay and your model shot is image four, the fix is usually reordering rather than reshooting.
Sportswear and gear: worn, not laid out
A runner photographed in the full kit — shoes, leggings, top — produces a complete look, and the accessories being carried or worn come with it.
The same kit arranged neatly on a studio floor is a flat lay, and returns nothing. Same products, same catalogue, entirely different outcome, decided purely by whether someone is in the shot.
Where it genuinely doesn’t work
Two cases are worth being blunt about, because no amount of configuration changes them:
- Home and furniture room sets. A styled room with nobody in it has no one wearing or carrying anything, so no look is built from it — however beautifully merchandised the room is.
- Catalogues shot entirely in packshots. If every image is a product on white, there is no photo for a look to come from.
That is not the end of the road for those stores. The “you may also like” related-products widget is included on every plan and works from catalogue data rather than image content, so it runs perfectly well on a packshot-only catalogue. It is the right tool there; a styled look is not.
What doesn’t change
Where a person is in frame, the mechanics are identical whatever the category. It works from the photos you already have, with no reshoot required. It links to the exact product where you sell it and a close alternative where you don’t. Shoppers add to cart inside the widget, and sold-out items are swapped for the next best match automatically, so a look never quietly breaks when one item sells out.
One practical detail worth knowing: photos are read in gallery order and merged until the outfit looks complete. Leading with a full-length shot of a person is both the best result and the cheapest one.
Do it with Shop The Look
Merch - Shop The Look detects the outfit in the product photos you already have and matches each item to your own catalogue — the exact product where you sell it, a close alternative where you don’t — with add to cart inside the widget and sold-out items swapped automatically. Where a styled look isn’t the right fit, the “you may also like” related-products widget is included on every plan. It installs as a no-code theme app block, stores no shopper personal data, and runs its analysis ahead of time so the storefront only loads pre-computed results. Live on the Shopify App Store, with plans from a free tier up to Enterprise.
Install on the Shopify App Store →
Frequently asked questions
Does complete-the-look only work for apparel?
No, but the deciding factor is your photography rather than your category. Any category works when the photo shows a real person wearing or carrying the items — beauty on a model, sportswear on an athlete, accessories being carried.
Will a flat lay or packshot produce a look?
No. A flat lay, a packshot, or a product shown alone on a plain background is deliberately skipped and returns no items, so no look is built from it — it stops the app inventing an outfit out of one product’s own straps and trim.
Does it work for furniture or home products?
Not from a styled room photo with nobody in it, because there is no person wearing or carrying anything. For catalogues like that the included “you may also like” widget is the better fit, since it works from catalogue data rather than image content.
Do I need to reshoot my catalogue?
Only if none of your photos show a person. If you already have model shots further down the gallery, reordering so a full-length one leads is usually enough.
Is there a free plan?
Yes. A free tier is available, with paid plans priced by how many AI looks can be live on your store at once, up to Enterprise.