What the Stitch Fix IPO tells us about the future of fashion
TL;DR: Companies like Stitch Fix are bringing personalized buying to the masses via artificial intelligence. This efficiency is the future of fashion.
Dear fashion industry, AI is coming for you. Here’s why it matters and why you should be paying attention to the Stitch Fix IPO…
Let’s look at a business model in the fashion industry that seems to be working surprisingly well: Independent fashion retailers, AKA fashion boutiques. They are succeeding because fashion boutique owners often develop incredibly rich relationships with their top customers. When they buy, they buy for Alex, Jackie, and Megan — real individuals. The store owner and stylists know exactly what sizes, styles, and brands their customers will buy. This makes their business incredibly efficient, letting them keep unsold inventory to a minimum, and profit margins high. It’s also great for brands, because boutiques sell through nearly everything at full price and never ask for markdown allowances. But the human relationships that power this model come with a high time and energy cost. A store owner can only lavish attention on so many customers. That means boutique retail, despite its low waste and high efficiency, has remained small scale.
The alternative approach — mass market fashion — often takes a ‘spray and pray’ approach to buying and merchandising. Retailers buy huge amounts of varied styles, based on last season’s sales data, and hope that they connect with the right customers. And the rise of e-commerce has made the shopping experience even less personal, as face-to-face interactions between customers and store employees become a thing of the past. Large retailers would like to treat Jackie, Alex, and Megan as individuals — but until now, that depth of intelligence hasn’t been remotely possible.
AI is about to change that forever. The rise of machine learning means human intelligence will no longer be a limiting factor in matching customers with the right product. Over the coming decade, we’ll see a huge shift in the way fashion retailers buy and stock merchandise as they develop increasingly sophisticated profiles of their shoppers and use AI to buy and serve up the most relevant products. These profiles will incorporate aesthetic preferences, brand tendencies, spending patterns, and a thousand other factors even the most diligent human couldn’t keep track of. As covered on The Jason & Scott show EP 105 ‘Stitch Fix IPO Hot Take’, AI will never need to start over and relearn a fashion customer’s preferences like a new personal stylist who joins the team. And it will draw on a gigantic database of past successes to predict the future with increasing accuracy.
The rise of machine learning means human intelligence will no longer be a limiting factor in matching customers with the right product.
Stitch Fix, the subscription fashion retailer that’s preparing to IPO, is winning with exactly this approach. They’re building customer profiles and using machine learning to deliver the right product to each customer with ever-greater efficiency. As their CEO, Katrina Lake, recently put it in their S-1 filing:
“Our success as a business is inextricably linked to our personalization capabilities. We strongly believe that most existing retail constructs are insufficient and out of date. We also believe our model of personalization, getting to know every client, each product, and generating relevant and actionable recommendations, to be superior and enduring… I believe we are the best in the world at personalizing in apparel at scale, but I also know that we are just at the beginning of how powerful personalization can be.”
Stitch Fix is already rumored to offer great partnership opportunities for brands, and traditional retailers should be watching carefully to see how much market share they’re able to steal.
Some of my burning questions are, will we be able to train AI to mimic the sensibilities of a tastemaker like Rachel Zoe, so the masses could access her styling and tastes, rather than only celebrities? And how deep can we go with personalization? Will AI be able to learn an individual’s tastes and design fashion from scratch that matches your unique style? A team of researchers from UCSD is working on it. Stitch Fix offers high-margin private-label products informed by their customer’s preferences. Will heavy hitters like Amazon also move in this direction, given that they too offer private label fashion and seem to be moving towards on-demand manufacturing?
How deep can we go with personalization? Will AI be able to learn an individual’s tastes and design fashion from scratch that matches your unique style?
I’m hopeful that the rest of the fashion industry will jump on the AI train, quickly adopting it to solve business problems. And late adopters that are being burned by the rise of e-commerce won’t make the same mistake this time. Sophisticated, open-source frameworks like Google’s TensorFlow will make this adoption possible for any business that’s willing, and the established fashion companies will have a large advantage with existing customer bases and data on customer buying behaviors that can be used to train AI models.
Furthermore, established fashion companies have creative talent on their side. We’re not about to see the human touch disappear from fashion. Instead, this is the beginning of “man-machine symbiosis” — a term borrowed from Peter Thiel’s Zero to One. Katrina Lake writes:
“Our business is built with the belief that data and technology make us humans better. Our algorithmic recommendations are a powerful partner to our stylists. Our stylists are freed from rote calculations and tasks and are able to focus on what they are uniquely, extraordinarily good at — creating human connections, being creative, providing valuable context to their choices. This partnership can power more meaningful, more human jobs.”
In other words, our role is changing. As AI begins to handle the mind-bogglingly complex quantification of consumer behavior, humans will be freed to do what we do best — exercise our creativity and explore new aesthetics.
The future of fashion belongs to humans and companies that are the best at interacting with machines. So buckle up, folks. If you think the fashion industry has changed a lot on the past 10 years, it’s only going to accelerate from here.
If you think the fashion industry has changed a lot on the past 10 years, it’s only going to accelerate from here.
PS: If you’re interested in this subject, I highly recommend listening to The Jason & Scott show EP 105 ‘Stitch Fix IPO Hot Take’ — it’s packed with interesting insights. And stay tuned for more thoughts from me on the wild collision of fashion and technology.
