Tony Pinville is the CEO of Heuritech, a trend forecasting company based in artificial intelligence for the fashion industry. Tony co-founded Heuritech in 2013 following his thesis in Machine Learning at the Pierre and Marie Curie University in Paris, drawing from his AI background. Finally, in 2019, he was appointed Young Leaders by the French-American Foundation. In this interview, he talks about what his solution has to offer and what’s next for the company.
E-3 Magazine: What exactly does Heuritech offer?
Tony Pinville: Heuritech provides solutions to fashion brands with our AI-based trend forecasting approach. We like to say that the collaboration between our data science teams and fashion experts is our secret sauce: we offer data-driven trend forecasts to brands in the fashion, luxury, and sportswear industries, including brands ranging from Louis Vuitton to Adidas to Havaianas. It is with these insights that Heuritech provides fashion brands the power to forecast demand and trends more accurately, produce more sustainably, and achieve unprecedented competitive advantage.
How does your solution work?
Pinville: Heuritech is able to accurately forecast trends up to 12 months in advance with 90 percent precision through an image recognition technology which was conceived and developed in-house. Daily, this technology analyzes millions of shared images and videos on social media to pinpoint over 2,000 attributes, from silhouettes to colors to textures. These data are then collected to provide an inside-look on a number of useful insights for our clients, such as trend growth between particular seasons, optimal launch times, consumer segmentation, apparel category, and geography. For example, Heuritech’s most recent trend report predicts that cable knit will rise 20 percent in visibility in summer 2021 compared to last summer for womenswear in Europe. We can even tell you that this fabric will be most desirable in the tops category, and that it will be most appealing to edgy and mainstream consumers. This is the sort of precision Heuritech provides to its clients.
What are the customer-side requirements? Do they need special devices to use your service?
Pinville: The only requirement on the customer side is a computer or smartphone. Heuritech has two online platforms: one for trend forecasting, meaning for colors, textures, and the like, and the other for product intelligence, meaning for shoes, bags, and so on. We personalize the respective data to each client: for instance, a client looking to expand their menswear in China will be given access to menswear trends in China across particular seasons. Our in-depth data accompanied with images on each platform truly allow our clients to align their teams who may speak a more data-driven language with those who speak creativity.
Why did you start Heuritech to begin with?
Pinville: Following my thesis in Machine Learning at La Sorbonne University in 2013, I launched Heuritech. Why? Because at the time, no one was talking about deep or machine learning. So with Heuritech’s co-founder, we wanted to find a way to apply this technology in a meaningful way, but we weren’t yet sure in what capacity. As such, we created Heuritech with first the technology in mind, developing cutting-edge machine learning algorithms capable of recognizing any attribute you can think of in images. We then tested our technology in a range of different industries, before finally landing where we are today: fashion. It clicked when we met with Louis Vuitton in 2016: they saw the power and the potential of the technology, and together, we co-constructed our first product with a use case in mind. And the rest is history. We’re lucky to say that we now work with major brands all around the world.
How could your solution potentially help customers navigate the COVID-19 pandemic and the different challenges of reopening?
Pinville: During the initial pandemic during spring of this year, Heuritech developed a program called New Data for Fashion to help smaller brands stay on their feet. We provided personalized trend reports and free resources to these applicants because we felt that emerging and independent brands needed all the help they could get, particularly because many of them have focus in sustainability and are truly changing the industry. Globally, what we’ve noticed as the year has progressed amidst this pandemic is that brands – big and small – can no longer rely solely on their intuition. Data have found a much more significant space in the fashion industry this year, and Heuritech is in the right frame to provide the industry with data-driven solutions. COVID-19 has accelerated the digitalization of brands, and we are helping our clients navigate shifting consumer behaviors and evolving trends through trend forecasting. Our market trend reports, for instance, have been of great assistance to our clients, who can plan collections and communicate with their customers with the assurance that they have a clear line of sight into the future.
How are you connected to SAP?
Pinville: Heuritech was part of SAP.io on the 2019 Spring Cohort in Paris, where our HQ is located. The program led by Sébastien Gibier focused on retail, consumer products, and CX. During six months, we got the opportunity to get a deeper knowledge of SAP as well as its clients and industry knowledge from the experts, which helped us a lot in our development. SAP integration for us is a long-term goal which we are working towards to become part of the App Center.
What’s next for Heuritech?
Pinville: Artificial intelligence will only continue to grow in importance in the fashion industry: According to MarketsandMarkets’ ‘AI in Fashion Market’ study, AI is expected to grow by 40 percent every year by 2024. It’s fortunate for Heuritech, because we’ve been firm believers of the benefits that AI can bring to this industry for over 7 years now. At Heuritech, we are vying for a fashion industry that is not only more streamlined, but more sustainable. Our data-driven trend forecasting can achieve this: brands infuse our insights into their strategies to better plan their collections in a way that avoids overstock and delivers desirable products, every time. For one, I know I’m optimistic about the future of fashion. It’s more green and more data-driven.