Virtual Try-On with Size Recommendation: Why Solving Both Problems at Once Changes the Conversion Equation

Two questions have stood between online clothing shoppers and the purchase button for as long as fashion e-commerce has existed. The first is visual: how will this look on my body, not on the model in the photograph. The second is practical: which size should I choose, given that sizing varies between brands and the label tells me very little about how a specific garment will actually fit. The industry has addressed these questions in sequence – size charts, then size calculators, then virtual try-on – but always through separate tools, in separate places, requiring the customer to assemble the complete answer on their own. Combining both answers in a single interface, at the moment the customer is already engaged with a product they want, is something that has not existed until now.

The two barriers and why they reinforce each other

It is tempting to think of the visual barrier and the sizing barrier as independent problems that happen to coexist. In practice, they interact and amplify each other in ways that matter for conversion. A customer who is uncertain about size is less likely to engage meaningfully with virtual try-on – why spend time seeing how a garment looks on you if you do not know which size to view it in. Conversely, a customer who has used try-on and likes what they see still faces the sizing question at the moment of checkout, and that unresolved uncertainty is enough to produce an abandoned cart, a two-size order with the intention of returning one, or a purchase followed by a return. Both outcomes are costly for the brand, and both trace back to the same underlying problem: the two questions were never answered together.

Standard tools each address one barrier with varying degrees of success. A size chart with garment measurements is genuinely useful, but it requires the customer to know their own measurements, which most people do not have readily available, and to do the work of comparison themselves. A size calculator removes some of that friction but still depends on manual data entry – an additional step that introduces drop-off at the exact moment the customer's attention should be on the product. Virtual try-on without size guidance answers the visual question and does it well, but it says nothing about which size the customer should actually add to their cart. A size recommender without try-on produces a number without any visual context. None of these tools, taken alone, closes both questions – and the gap between them is where purchase intent goes to dissipate.

How the combined widget works – and what makes it different from what came before

The mechanics of combining both functions in a single widget are built around something the customer has already done: upload a photograph. The AI that analyzes that photograph to overlay the garment simultaneously reads the customer's body proportions – the relative measurements and shape characteristics that determine how a specific garment will fit. It compares those proportions against the garment measurements the brand has provided for that item, and produces a size recommendation within the same interface where the try-on result appears. The customer sees how the garment looks on them and sees which size is recommended, in the same session, without navigating away or entering any additional information.

The distinction from existing size tools is meaningful. Standard size calculators ask customers to input their measurements – chest, waist, hip, height – and match those against a brand's size chart. This approach has two weaknesses that are well-documented in e-commerce research: most customers do not know their measurements with useful precision, and the manual input step creates friction at a moment when the customer's attention and intent are at their peak. The approach taken here derives body data from the photograph the customer has already provided for try-on purposes. There is no additional form to fill, no separate tool to navigate to, no interruption in the experience. The size recommendation is a by-product of something the customer was already doing.

First implementation – a brief example

The first brand to implement this combination was Pop'n'Shop, a Russian youth fashion brand whose audience – young, fast-moving shoppers accustomed to making quick decisions – is precisely the segment for whom removing both barriers simultaneously has the most direct conversion impact. This is an audience that will not spend time cross-referencing size charts, but also does not want to deal with the friction of returns. The integration was a natural extension of their existing try-on implementation and required no additional technical work on the brand's side: the size recommendation function was added to the widget they were already running, with the brand's garment measurement data as the only new input.

In practice, the brand provides garment measurement data once, and the widget goes live in one business day with onboarding support from LOOKSY. The customer sees how the garment looks on them and sees which size is recommended, in the same session, without navigating away or entering any additional information.

Why the combination outperforms either tool alone – the conversion mechanics

The mechanism through which combining visual try-on and size recommendation increases fashion e-commerce conversion rate beyond what either achieves separately is not complicated, but it is worth being precise about. When a customer is on a product page looking at a garment they are interested in, they have two open questions. Virtual try-on closes the first: they can see how the item looks on their body. Size recommendation closes the second: they know which size to choose. When both questions are answered simultaneously, in the same interface, at the moment of highest engagement, there is no remaining basis for deferring the decision. The purchase either happens or it does not for reasons unrelated to information gaps – which is the cleanest conversion condition a brand can create.

The effect on returns works through two independent mechanisms, which is why the reduction tends to be larger than either tool produces on its own. The first mechanism is straightforward: a customer who ordered the correct size does not return the item because it does not fit. Wrong size is consistently one of the top two reasons for returns in online clothing retail, and a size recommendation grounded in actual garment measurements and actual body proportions addresses it directly. The second mechanism is about expectation alignment: a customer who saw themselves in the garment before purchasing has a more accurate expectation of how it will look when it arrives, and is less likely to experience the gap between expectation and reality that drives impulse returns. Both mechanisms operate independently, and together they compound.

There is also a customer lifetime value dimension that tends to get less attention in conversion discussions. A customer who ordered the right size and received what they expected has had a better first experience with the brand than one who had to navigate a return. That better experience affects repurchase likelihood – and the difference between a customer who returns once and one who returns three times is not captured in a single-session conversion metric. Reducing sizing errors at the point of first purchase is, in this sense, also a retention investment.

What brands need to make it work – and why the preparation has value beyond this feature

For the size recommendation to function accurately, brands need to provide garment measurements for the items in their catalog: the actual centimeter values for chest, waist, hip, length, and any other dimensions relevant to a given category, broken down by size. Many brands have this data in some form but not in a consistent, accessible format – and preparing it requires an initial investment of time. This is worth acknowledging directly, because understating the effort involved does not help brands make good decisions about whether to pursue the feature.

What is equally worth stating is that this preparation has value that extends well beyond the try-on widget. A brand that publishes accurate garment measurements reduces the volume of customer support inquiries about sizing – a real and measurable operational cost. It reduces sizing-related returns independently of any technology. And it signals a level of product transparency that is increasingly valued by the segment of shoppers who research carefully before buying: people who read specifications, compare measurements, and make considered decisions. Building the infrastructure of detailed garment data is not just a prerequisite for this feature; it is a competitive posture that pays off across multiple channels.

The technical integration is handled by the LOOKSY team: the brand provides the measurement data, and the widget goes live in one business day, with a dedicated manager supporting the onboarding process. Brands that want to see how the combined try-on and size recommendation experience works on their own catalog can book a demo at looksy.tech.

What this means for the broader problem of online clothing retail

The difficulty of selling clothing online has always been a problem of information. A customer in a physical store can touch the fabric, hold the garment against their body, try it on. They can answer both questions – how does this look on me, which size fits – through direct experience. Online retail has spent two decades building approximations of that experience: better photography, richer descriptions, size charts, virtual try-on. Each addition has reduced the information deficit but not eliminated it. The combination of visual try-on and AI-driven size recommendation in a single interface does not replicate the physical fitting room, but it addresses the two most consequential information gaps in the same session, at the same moment, with data derived from the customer's own photograph rather than from their best guess about their measurements.

For brands, this means that the question of how to reduce clothing returns online store-wide and improve fashion e-commerce conversion rate simultaneously now has a more complete answer than it did before. Not because the tools are new in isolation – virtual try-on has existed for some time, and so have size calculators – but because they now operate together, on the same data, at the same moment. The friction points that have been the hardest to close are closing at the same time, for the same customer, in the same interaction.