When it comes to building product taxonomies for Fortune 500 and specialty retailers, the sheer breadth of products, styles, cuts and fits ensures that even the most well-crafted automation solution will often miss important details that consumers actually care about. At Lily AI, we went about building our customer intent platform differently.
The depth of Lily AI’s taxonomy is driven by a dedicated in-house styling team that bring deep backgrounds in fashion, technology and retailing, and who are experts in helping retailers control product accuracy and depth. The team even taps their extensive expertise to advise Lily’s retailing partners on core trends and upcoming market shifts. This, combined with the deepest image recognition platform in retail, is what provides a depth and scale of attribution that no other solution can match, and turns qualitative product attributes into a universal mathematical language at a high volume with unprecedented accuracy.
We caught up with Kathy Lee, Lily AI's Director of Styling, to understand how she and her team help to drive product discovery.
It's said that online product discovery is the main battleground right now for retailers. What brought us to a place where this has become so important?
As an online and avid shopper myself, whether I’m just browsing, aka “window shopping”, or looking to purchase an item I didn’t even know I wanted or needed, product discovery is extremely important to increase sales revenue and perhaps customer loyalty.
The fashion market has always been a competitive space, and even more so now, especially during the pandemic. More consumers are shopping online, which is why personalization has become even more important. If I’m shopping around for a cutout dress with all-over silver sequins for an New Year's Eve cocktail party, I’d love to see an endless and diverse selection of similar items, and maybe even a whole outfit to complete the look. Just like any other online shopper, if I don’t see what I’m looking for within 2-3 minutes, I’m going to search somewhere else. This realization is something that retailers have really woken up to.
Lily AI built out an in-house styling team right when the company was founded. Why was this such a key part of how we built our platform, and what are your team's core responsibilities?
Our in-house styling team’s core responsibilities include in-depth research and development of product attribution for multiple categories, across all genders and unisex/genderless. To name just a few, those categories include apparel, handbags, shoes, swimwear, intimates, to even socks & hosiery - with more on the way. Creating our core taxonomy for each category is essential to train our automation “to think and behave like a stylist”. Our in-house styling team continues to grow and develop our existing categories to ensure attributes are updated to the latest terminology used in the fashion and retail spaces.
What does having an in-house team provide Lily AI customers with, as opposed to retailers who rely on pure automation to build their product taxonomies?
Having an in-house team with an expansive range of backgrounds and experiences really allows our customers to have a source of experts right at their door. It's part of the engagement we have with any of our customers.
Our styling team brings a diverse range of experiences, having come from various brands and retailers such as Nordstrom, Paige, Saks 5th, Stitchfix, Macy’s, Urban Outfitters, and more. The team consists of visual merchandisers, buyers, personal stylists, clothing designers, make-up artists, and even interior home design experts. Another reason Lily AI stands out by having an in-house team is analyzing the importance of each attribute and its relevance to the user based on dressing style or preferences. As I'd said, our team continues to evolve our taxonomy every day so that Lily AI customers will receive the latest product attributions based on seasonal trends and changes.
How are your team's stylists integrated with our AI-powered automation? How do you two "work together" on a daily basis?
Aside from researching, developing, and updating our taxonomy for multiple categories, our team of experts plays a prominent role in auditing our product attribution that comes from automation. As we continue to expand to even new verticals, such as home and beauty, accuracy plays a crucial role in the automation process. It's a real competitive advantage for Lily AI, having the AI and human experts work together in this way.
Finally, tell us a little bit about the core trends reports and market conditions research that you provide to Lily AI customers.
I’m happy to announce that we just launched our trends team this past quarter! As of today, we’re able to elevate our customer's experience by offering a trend analysis report each year. Not only will we be able to provide our customers with upcoming and future seasonal trends, but we’ll be able to enhance their customers' experience with the latest attribution or ‘trend’ terms based on market and cultural shifts.