"Cross-fertilization between tech and fashion is the strength of Heuritech"
1/2 - On the occasion of the announcement, on September 3, 2019, by Heuritech of a € 4 million fundraising, his co-founder Tony Pinville tells more to Le Boudoir Numérique on how he is considering the strategic development of his company, which specializes in artificial intelligence applied to fashion.
By Ludmilla Intravaia
Le Boudoir Numérique : You co-founded Heuritech, in 2013, with your associate Charles Ollion, who is, like you, a PHD in machine learning (field of study of artificial intelligence allowing computer systems to learn how to solve specific problems, based on the data transmitted to them, without being explicitly programmed to perform the task, AN). Can you come back to the milestones that have marked the evolution of your company?
Tony Pinville, CEO of Heuritech : At the end of our thesis in artificial intelligence, we were at a pivotal time of its development, with the advent of deep learning (methods of machine learning to model data with a high level of abstraction, AN). Aware of the disruptive potential of this technology, we wanted to apply it to the business world, by creating Heuritech, a startup halfway between the problems of applied research and business issues. We started to think about how to adapt it to the needs of the market, at the beginning, without a specific area of predilection. We worked with finance, agriculture sectors and postal services. In 2015, following a meeting with Louis Vuitton, we realized that both luxury and fashion had high expectations in terms of trend predictions. We thus made a strategic shift towards the luxury industry in 2016, deciding to focus on the analysis of images on social networks. In 2017, after a first round of fundraising, we expanded our team and launched our first analytics platform for iconic products on social media to help brands measure the success of their iconic models around the world. It was a fundamental moment for us, where our technology has become a solution that can be used by the actors of fashion. In 2019, we created a new trend detection platform for luxury and fashion brands. The goal is to anticipate trends, by analyzing the content of social networks and to allow brands creating collections to obtain data, in real time, on the constant evolution of consumer preferences, three, six, nine months, one year in advance. Our algorithms allow us to determine that the trend, for example, of the asymmetrical blue skirt will show, in one year, a rise of 35%, in Europe.
So you're collaborating now also with fashion brands...
Having started with luxury brands like Louis Vuitton, Dior or Paco Rabanne, our team of forty people lends its services to sportswear and fast fashion industries, with customers like Adidas, Kontoor Brands or Fashion Cube. We collaborate with brands worldwide, in Europe, the United States and Asia.
You have announced, on September 3, 2019, a fundraising of 4 million euros. How is this new milestone important in the strategic development of your business ?
This fundraising will allow us to strengthen our competitive advantage by investing in research and development of machine learning and predictive analytics technologies. We will also recruit new staff and open two offices in New York and Singapore to continue our international expansion. In parallel, we will attack new industries, starting with beauty and then lifestyle. In recent years, many cosmetics brands have asked us to access our trend detection tool because the beauty sector is waiting for informations on what is happening on social networks, where trends are changing very swiftly, especially in Asia. That's why we want to launch a solution taillored for this market.
Why two new offices ?
Next year, we will inaugurate a first office in New York to continue to support our existing customers and be more present on the spot, in order to win new ones. Then, concerning Singapore, we observed that our customers in Europe, in the United States and even in Asia want to know more about the Asian market and, in particular Chinese. Some Asian brands know their market very well but less well the European market and vice versa. And things are happening in China, often difficult to apprehend. Our technology has this stimulating effect that it offers the possibility of bringing out, with precision, the specificities of each market, thus making it possible to confirm or refute certain beliefs and to distinguish between the idea that one have intuitively about a market and its reality, as such. That's why we add other social networks like Weibo (chinese microblogging website, AN) to our predictive tool, to provide our customers with a global coverage of information, with syntheses adapted to each market.
In terms of recruitment, what have you planned ?
We will hire fashion and beauty experts, marketing and sales executives, while expanding our research teams. At the end of the year, we should be fifty and next year, seventy. Initially, our team consisted of experts in pure research on artificial intelligence and computing. In a second step, we hired people with product skills to negotiate this shift between technology and business solution to then hire fashion experts to identify the relevant trends in the eyes of our customers. From now on, we combine technological expertise with business and fashion skills. This cross-fertilization between two generally unconnected worlds, fashion and technology, is our strength and opens up huge exploratory fields.
* Read the second part of Tony Pinville's interview on Le Boudoir Numerique : "AI can contribute to the virtuous circle of sustainable fashion".
* Continue reading on Heuritech with "Will leopard print pleated skirts be trending any time soon ? AI already knows it !".