Why Every Fashion Store Needs Complete the Look

Think about the last time you bought a shirt you really liked. You probably weren't thinking "I need a shirt." You were thinking about Friday dinner, or a trip, or the trousers already in your closet that never had anything to go with them.

Nobody wants a shirt on its own. They want an outfit, and the shirt is how they get into it.

Most online stores don't sell that way. The product page shows one item, very well. There's the zoom, the size guide, the fabric details and the reviews. Then the shopper adds it to the bag and the page has nothing left to tell them. What would they wear it with? Which shoes? Does the store even sell something that works? Figuring that out is left to them.

Some shoppers will open more tabs and work it out. Others won't bother, and the order goes through with one item in it.

What is "complete the look"?

Complete the look is the part of a fashion store that answers "what do I wear with this?" It suggests complementary pieces, such as trousers, shoes or a bag, that go with the item the shopper is looking at, so they can buy a whole outfit instead of a single piece.

It's different from "you might also like." That row shows alternatives: other shirts, in case this one isn't right. Complete the look shows complements: the things that go with this shirt once the shopper has decided on it.

Shoppers already look for the rest of the outfit

This isn't a guess. Baymard Institute is an independent ecommerce UX research firm that has run more than 4,400 moderated usability sessions. In its product page research, it describes test participants hunting around product pages for supplementary products after adding an item to their cart. One example was a shopper looking for trousers and shoes to go with a shirt they had just found. Participants found these suggestions very helpful when they were on the product page, and often expected them to be there.

That's a shopper who has just said yes and wants to say yes again, and many product pages don't help.

In a 2014 review of 50 major ecommerce sites, Baymard found that 58% offered only one type of suggestion, either alternatives or complementary products, or mixed both into the same row. In later analysis, it reports that 47% of sites don't have a section on the product page dedicated only to supplementary products.

The two jobs are different. Alternatives help a shopper who hasn't decided yet. Complementary pieces help a shopper who already has. If you put them in one carousel, neither job gets done well.

Why complete the look matters for average order value

Here's the commercial side, with the caveats that come with each number.

The most-cited study on this is Salesforce's Personalization in Shopping report, which analysed more than 150 million shoppers and 250 million visits. Visits where a shopper clicked a product recommendation made up just 7% of visits but drove 24% of orders and 26% of revenue. Purchases where a recommendation was clicked had a 10% higher average order value.

Two honest caveats. First, the study was published in 2017, so it's nearly a decade old. Treat it as a long-standing benchmark, not a measurement of today's tools. Second, it compares shoppers who clicked with shoppers who didn't, and people who click recommendations may simply be keener buyers to begin with. It shows that recommendations and bigger baskets go together. It doesn't prove how much of the difference the recommendations caused.

The wider personalization research points the same way. McKinsey's 2021 Next in Personalization report found that 71% of consumers expect companies to deliver personalized interactions, and 76% get frustrated when that doesn't happen. It also found that personalization most often drives a 10–15% revenue lift, with company-specific results ranging from 5% to 25% depending on the sector and how well it's executed.

That last point matters. Personalization done badly doesn't earn much. Doing it well is what pays off.

For context on order size, Shopify's guide to average order value says apparel and accessories orders typically fall in the $40 to $170 range. Whatever your number is, an order that could have included the trousers and shoes but didn't is revenue your store missed.

Why most complete-the-look rows fall flat

If this is so valuable, why do so many stores get it wrong? It usually comes down to one of four problems.

It's built by hand. A merchandiser styles a few hero products for the lookbook, and every other product page gets nothing. Hand-styling doesn't scale past the products your team has time for, and it goes out of date as the catalogue changes.

It's built on "people also bought." Purchase history is useful, but it doesn't understand style. It will happily put a second black T-shirt next to the first because plenty of people bought two. That isn't an outfit.

It mixes alternatives and complements. As Baymard's research shows, many sites put "instead of this" and "as well as this" in the same row, which leaves the shopper to sort one from the other.

It only shows products. Four thumbnails in a row aren't a look. Shoppers need to see the pieces together to picture themselves wearing them.

How Stylor helps shoppers complete the look

Stylor is an AI stylist that sits on your storefront and answers "what do I wear with this?" using the products you sell.

It works from your catalogue. Connect your Shopify store and Stylor imports your products, then keeps them in sync on a schedule. New products are picked up on the next sync, and products you remove stop appearing. Every look is built from items in your catalogue.

It understands what it's styling. Stylor's AI reads each product's photos and description to tag details such as colour, fabric, fit, category, formality and season. That's what lets it put pieces together in a way that makes sense, rather than just grouping things other people bought.

It builds complete outfits. Stylor puts together full looks: a top, a bottom and shoes, plus accessories where they fit. It aims for colours that work together and suit the occasion and season.

It styles the product the shopper is looking at. Add Stylor to your store with a single embed snippet. When a shopper opens it on a product page, it knows which item they're viewing, so asking "what do I wear with this?" gets a complete look built around that product.

Shoppers can ask for what they need. Not everyone wants the default suggestion. A shopper can ask for "something for a summer wedding" or "a way to dress this down for the weekend" and get complete looks back, still from your catalogue.

It shows the look, not just the pieces. Each look comes with an AI-generated, full-body image of a model wearing the outfit, so shoppers can see how the pieces work together.

It keeps sold-out sizes out of the cart. When a shopper adds a look's items to their cart, Stylor won't add a size that's out of stock.

You can measure it yourself. Stylor's analytics show how many sessions used the stylist, how many looks were generated, and how many cart adds, and how much cart value, came from Stylor. The figures come from your own storefront. We'd rather you judge Stylor on your own numbers than on anyone else's.

How to test whether complete the look is working

Whether you use Stylor or something else, test it properly. A common mistake is comparing shoppers who clicked the feature with shoppers who didn't. As with the Salesforce study, that mostly tells you who was already keen to buy.

A fairer test looks like this:

1. Split visitors randomly. Show the feature to half your visitors and hide it from the other half, and keep each visitor in the same group across visits.
2. Measure everyone in each group, including people who never touched the feature.
3. Track revenue per visitor, not just order value. Revenue per visitor captures both "did they buy?" and "how much?", so a bigger basket can't hide a drop in conversions.
4. Count returns. Measure the revenue you keep, not just what goes through checkout.
5. Run it long enough. Cover several full weeks, and avoid a big sale that could skew one group.

A test like this gives you a result you can trust and defend. It also tells you whether the feature is earning its space on the page.

Stop selling just the shirt

Your shoppers are already trying to put an outfit together. Baymard's research shows them looking for the trousers and shoes. The question is whether your store helps them or leaves them to work it out alone.

Completing the look is good merchandising. It's what a great sales associate does: hands you the jacket that finishes the outfit before you think to ask. Stylor brings that to your product pages, styled from your own catalogue, with numbers you can check for yourself.

Want to see what Stylor would put together from your store? Try Stylor with your catalogue.

---

Sources: Baymard Institute, "Product Page Usability: Recommend Both Alternative & Supplementary Products" (2014) and "7 Product Page UX Implementations that Make REI Best-in-Class." Salesforce, "Personalized Product Recommendations Drive Just 7% of Visits but 26% of Revenue," November 2017. McKinsey & Company, "The value of getting personalization right—or wrong—is multiplying," November 2021. Shopify, "Average Order Value: Formula and 7 Ways." For the academic background, see Kang et al., "Complete the Look: Scene-based Complementary Product Recommendation," CVPR 2019. Statistics in this post are third-party industry research, not Stylor results. Images are AI-generated illustrations, not screenshots of Stylor.