Interview by Caroline Issa
Portrait courtesy of Janice Wang

CI McKinsey & Company estimates that more than 70% of apparel returns are because of the wrong size and fit, which costs retailers billions and creates huge waste. Alvanon works with body data and consumer analytics to define fit standards. Currently, a person may be a size M at one brand, L at another and S at a third. Where do you see the biggest blind spots in the industry’s data – age, size inclusivity, geography, gender – and how are you trying to close those gaps?

JW The core problem is misalignment. When we started Alvanon 25 years ago as a mannequin company, it was to fix the fact that designers, technical designers, production and merchandising were not comparing apples to apples, even though they were all supposedly working on the same customer. My father saw early on that people would buy clothes on the internet; his logic was that if you don’t first create a shared standard – a ground truth – then every new tool, channel or business model just adds another layer of noise on top of already bad information. Today, maybe the product team and the factory are aligned on a size “medium” because they’ve refined that over multiple seasons, but they’re not truly aligned on small, large or extra‑large. The grading logic is fuzzy, which means the extremes start to behave unpredictably. Merchants are then misaligned with their products because they’re making volume bets without understanding how those sizes actually fit real bodies, and e‑commerce is misaligned with both, often getting fragmented or delayed feedback on returns and reviews. Scale makes it worse: a brand worth $100 million can still put five people around a table and sort it out; a brand worth $7 billion has 5,000 people at 5,000 tables, all working in silos and layering bad information on bad information.

CI If the issue is systemic misalignment, where do you actually start to fix it – and how quickly can brands see impact on returns of sold goods?

JW You always start with standards and grading – what we call “the foundation”. For example, we worked with Bershka in a very straightforward way. First, we agreed on a clear brand standard for the core body, then we defined the graded sizes that flow logically from that standard, and then implemented that consistently across design, technical, sourcing and suppliers. Once that was in place, in just six months they saw a 10% reduction in how many items the consumer returned – no rocket‑science technology, just hard, slightly unglamorous foundation work. When your foundation is right, you get an immediate reduction in returns and visible efficiency gain across the value chain. Factories are happier because “fit” is expressed on a physical form, not as a mysterious model and some vague comments on a tech pack [a document that outlines the design and technical specifications of a garment]; they can literally see what needs to change and where. Merchants can see, for example, that a size‑18 version of a certain style, cut on a particular block, will never perform at a price that makes money, so they either reduce the buy in those sizes or redesign the style instead of blindly repeating the mistake. A lot of the lost art is simply costing, grading and allocation being looked at holistically, rather than outsourced to factories.

CI We tend to treat sizing as an exact science, but bodies are messy and always changing. How do you manage the tension between the fantasy of perfect precision and the realistic goal of predictable fit?

JW The first step is to strip out preference. If you can’t do up the zipper or you can’t pull the garment over your shoulders or hips, you’re not that size – that’s not a styling choice, it’s basic reality. Once you’ve put people into the right “can actually wear it” bucket, then you can talk about preference: oversized, body‑skimming, cropped, boxy – all the nuances that make fashion fun. We’re not working with made‑to‑measure brands; we’re working with brands who are building garments industrially. The job then is to hold a much better living library of bodies and patterns, and to make conscious decisions about which bodies each product is built for. Then you match product, size range and buy allocation to the actual demographic: region, age, build, culture, climate, lifestyle. Historically, you had very focused propositions – Long Tall Sally in the UK for tall women, for instance. The challenge now is that many retailers want to be “for everyone” while their underlying standards are fuzzy, which guarantees disappointment at the edges. Most companies now have some concept of an Asian fit because they recognise it isn’t enough to say, “Let’s make a size 0 and be done with it.” Global sportswear brands, for example, institutionalised Asian fits recently because they are global, performance‑driven and data‑rich; they could see that certain regions had different body shapes and proportions and they scaled that knowledge into product. The deeper question for any brand is: are you truly “for everybody”, or do you know exactly whom you’re designing and grading for – and where they live, shop and return from?

CI Looking ten years ahead, what is the most radical change you’d like Alvanon to have triggered in how the apparel industry thinks about risk?

JW The most radical idea is actually very simple: inventory is cash. Then the core risk question becomes, “How precisely can I predict what will sell, where and to whom – and how fast can my operations pivot when that prediction changes?” Once you think that way, you can’t treat sizing, grading and regional allocation as an afterthought; they are central to your balance sheet. The fast‑fashion players, particularly some Chinese‑based ones, move faster largely because they have a manufacturing backbone. They can hold grey fabric rather than finished goods, cut later, and allocate styles and quantities by region as the data comes in, instead of locking everything in months ahead. Most apparel companies are still driven by financial reporting cycles rather than operational logic, so they don’t invest in the long‑term strategy that would let them be that agile. Releasing those old levers – being willing to question legacy sizing, to re‑platform data, to get genuinely granular about size curves, regional assortments and warehouse allocation – is where the real risk reduction lies. Granularity matters; sizing matters; size matters. In a world of overproduction and tight margins, building around that fact is not a niche concern – it’s the business model. .