Or listen in your favorite podcast app
Apple Podcasts / Google Podcasts / Spotify
For many, the idea of walking into a store and finding new clothes that actually fit is non-existent. The designs are either too bulky, or the lengths of the sleeves and pants are too long. Frustrated, many leave and do their shopping online instead. Except the experience isn’t much better there. Savitude co-founders Nick Clayton and Camilla Olson are on a mission to help turn the situation around. The idea behind the company is a radical one: to recreate a designer’s brain by using code. Nick and Camilla joined IT Visionaries to discuss what that means and how they are using A.I. and machine learning to find clothes for everybody.
3 Key Takeaways
- Too many clothes are made for body types that aren’t realistic because designers typically only design for one or two body types
- Thanks to a backlog of data and through the use of A.I., Savitude utilizes nine different body types, and predicts how clothes will fit for customers
- One-in-six online clothing purchasing ends up being returned due to fit, which means there is a huge opportunity to offer a better, more frictionless experience using technology
For a more in-depth look at this episode, check out the article below.
Camilla Olson said it best, “We’ve recreated a fashion designer’s brain. We’ve put in the code of how she thinks. So when we make recommendations, it goes with the same process that I would in my studio trying to fit somebody.”
Olson is the CEO and founder of Savitude, a company on a mission to uphold that manifest by delivering form-fitting fashion designs while eliminating waste. Olson, along with co-founder and CTO Nick Clayton, joined IT Visionaries and discussed why eCommerce has put a strain on the fashion industry, and how they are using A.I. and machine learning to develop clothes that work for everybody.
Founded in 2016, Savitude is an integrated A.I. retail technology company that utilizes predictive learning and analysis to assure designs cover the full spectrum of body types. Savitude accomplishes this mission by optimizing line planning with enhanced sketching and predictive trends. The idea was generated once Olson — who holds a degree in fashion design — realized that retailers were tailoring their products to an unrealistic look.
“The industry is focused on the hourglass body shape, not the actual bodies,” Olson said. “There are only 20% of women who have hourglass body shapes, so we realized we could solve this problem with predictive modeling techniques.”
Olson explained that she learned in design school that the focus in the fashion industry has never been on designing for real women with real bodies. Instead, designers create looks geared toward manufacturing and efficiency.
Olson and Clayton ideated for five years, workshopping the issue of one-size-fits-all design while trying to figure out a solution from a customer perspective.
“A lot of A.I. solutions out there take a cookie-cutter approach,” Clayton said. “What we did from the start was when we actually designed the structure of the algorithms we were using, we brought design philosophy into the process. That gave us a system that actually is able to make inferences a lot quicker and with a lot less data because it is structured in such a way that the structure flows naturally into the types of inferences that it needs to make.”
To streamline the flow, Savitude takes inspiration from images, utilizing a designers’ archival history to help them predict how future designs might fit, and predict upcoming trends. According to Olson, those things are helping Savitude ideate trillions of different designs to fit their customer base.
“A fashion designer can only design for one or two body shapes at a given time,” she said. “In our taxonomy, we have seven to nine different body shapes that we consider. So in our system, we can test for all seven to nine to make sure that the collection you’re producing matches your customer population.”
Part of Savitude’s goal is for designers and retailers to have an optimal assortment of products, the other half is to eliminate waste. Clayton went on to explain that one-in-six online clothing purchases ultimately gets returned, which often gets sent to a landfill. The solution is not easy, but it’s worth it.
“What it requires is that you’re more intentional about the way that styles are distributed,” Clayton said. “You can create an assortment that serves everyone with the same number of skews, just with a set of skews that are better distributed so that there is something for everyone. Then you’re not overproducing dresses for one body shape when they represent only 20% of the population.”
So what are the inner-workings of the A.I. and how does it ideate based on various factors? According to Clayton, they look at two sources of input — the first is body shape and proportion and the second is the design of the clothes.
Savitude deploys a visual recognition system that allows them to look at photos of clothes from retailers or from social media, and then use those pictures to measure design details. The clothes are then assigned a numerical score that correlates to the person shopping.
“Most people understand images, and designers particularly are fantastic at understanding images,” Clayton said.
Clayton and Olson reiterated that their technology is in the infancy of the design process, with beta testing occurring this year. But the progress so far has been exciting.
. —
