Beyond Fit: Why We Integrated Color and Print into Our Recommendation Engine

When we first launched, our recommendation engine was built around one key innovation: personalized fit advice based on body shape. At the time, this was a fresh approach in the market: where most fashion tech tools only guessed size or offered generic suggestions. We focused on fit first, because we knew that when something fits well, confidence follows.

But we also knew that style is more than just fit.

In real life, when you work with a personal stylist, they don’t stop at how something fits your body. They also consider which colors suit you best, based on your natural features like skin tone, eye color, and hair. This is where seasonal color typing comes in. And this is exactly where we’ve been innovating behind the scenes.

From Fit to Full Styling

​

We’ve been expanding our technology to go beyond fit. We’ve integrated colors and prints into our algorithm, linking them to the unique profile data of each shopper. This means our model doesn’t just recommend what fits you, but also what suits you.

The system currently uses the shopper’s skin, hair, and eye color to recommend colors that align with their seasonal type (think: summer, autumn, etc.). And we’re going to go even deeper. We're adding an extra layer of analysis to skin tone, focusing on undertones and warmth: cool, neutral, or warm. To do this, we will introduce a new survey question that allows us to map the seasonal profile even more precisely.

The goal? To match every shopper with a color palette that enhances their natural features, just like a stylist would.

And what’s even better about this? The typical "no-sale" that often follows a traditional color analysis: where a customer is told which colors suit them best, but finds nothing matching in-store, is non-existent in our case. Why? Because we work with the current collections of fashion brands. Our algorithm doesn’t show abstract advice; it boosts the pieces a brand already sells that are the best match for each individual. That means shoppers get real-time, actionable styling support, and brands see increased conversion on the items they already stock.

Smarter Algorithms, Constant Improvement

​

Under the hood, our algorithm is constantly being evaluated for performance and accuracy. We monitor bugs and scoring issues closely so we can keep improving the quality of every recommendation.

As always, our mission is to make it easier for shoppers to find pieces they love,

and feel great in. That’s why this is just the beginning. Colors and prints are only the first of many features we've added to bring truly personalized styling into the digital age.

Stay tuned for more updates on how we're rethinking fashion tech from the inside out!