Why We Reluctantly Built a Sizing Tool. And How It turned out Anything But Standard

When we started building our technology, we deliberately steered clear of being “a sizing tool.” There were already plenty of those on the market. Our vision was always more ambitious: to offer something closer to a great shop assistant, someone who doesn't just tell you your size, but who considers your body shape, your style preferences, and which colors boost your confidence.

People are more than the numbers on a measure band.

We envisioned a digital stylist that could deliver intelligent fit, color and size guidance, curated style advice, and emotionally aware experiences, at scale. Because people aren’t just a size. They are more than a number, more than a character in a size chart. A good human advisor looks at the whole person: the body proportions, their preferences, and how a person wants to feel in their clothes.

But Then the Market Asked

Despite this clear focus, we kept hearing the same request from our customers: “Can you help us with sizing too?”

Sizing might not have been our original path, but we listened. As demand grew, we made a strategic decision to fast-track the development of a size recommendation algorithm: delaying other projects like our conversational AI prototype.

A Smart, Adaptive Sizing Engine

Our sizing algorithm predicts body measurements using a smart questionnaire: asking for inputs like age, height, weight, and a few simple body-related questions. These predictions are then matched with product-specific size charts to generate tailored advice.

The real challenge? Getting accurate results with limited input, while dealing with wild variations in size charts across brands and regions. One brand’s “M” is another’s “L”: and then there’s the question of fit intention (oversized? fitted?).

So we built two powerful things:

1. A robust Sizing API, capable of mapping shopper data to size charts.

2. A conversion layer, which automatically detects and corrects inconsistencies across different brand or regional sizing standards.

   
   

But even that wasn’t enough for us.

Coming from the fashion industry myself, I knew that accurate sizing isn’t possible with just body data and generic size charts. You need more. You need the designer’s original intention. Is the product meant to fit snugly or loosely? Is it cropped, tailored, or relaxed?

So we made two key upgrades:

• We started integrating design sheets and tech packs directly into our system, giving us the real dimensions and fit intentions of each product.

• We enriched our shopper profiles with wearing preferences and more nuanced body data, to provide a fuller, more human understanding of fit.

The Holy Three: Fit, Color and Size

We believe that great shopping experiences come from the combination of body shape, color harmony, and size accuracy: the holy three-way that makes fashion advice actually useful. That’s what we aim to deliver. Because no one should have to settle for a one-dimensional experience.

With our sizing tool now in place beyond the features we already offered, built on our own terms and vision, we’re excited to continue pushing the boundaries of what fashion tech can offer.