By Eleonore Valadji
We’ve heard far too many times that fashion trends are cyclical. In fact, we’ve witnessed it ourselves with the return of the 80s mum jeans, 2000s ultra low-rise, and more recently the UGG boot. Yet there remains a great deal more work to forecasting than the cyclical theory – How long will the trend last? How will it vary from its previous forms? It is clear that the UGGs worn a decade ago couldn’t be further from the ultra-mini platform editions Bella Hadid was infamously spotted wearing with little other than micro shorts and a leather jacket.
Trend prediction has long been the qualitative work of experts in the fashion industry. These people meticulously inspect runway shows, street fashion, and pop culture in order to refine their predictions of the season. These are then sold to designers and trickle down to high street retailers. But times are changing, and Pinterest, Instagram, Tiktok,… have entered the scene and thrown the fashion industry hierarchy. A stint in fashion school is no longer a pre-requisite for contributing to and influencing these sorts of discussion. The power of trendsetting has trickled from the hands of the few fashion ‘experts’ into those of celebrities, influencers, and even the girl from your school who magically blew up on tiktok overnight. In other words, a democratisation of fashion has occurred in such a way that we no longer rely on the spectacles of fashion shows to dictate trends. However, this disruption of power at the same time means that there is much more input for designers to consider when developing new collections.
This is where AI comes in, turning the qualitative task of trend forecasting into a more quantitative process. The motivation has always been to sell, but with a wider range of inspiration sources, you could say that customers have become exceedingly picky about our fashion choices – these can no longer just be dictated to us. However, AI can make use of our comprehensive social media presence to track the evolving popularity of items and identify key trends in real time thanks to image recognition, hashtag detection and collecting like/repost/follow statistics. This gives fashion businesses a clear direction to follow to a degree of thoroughness which is unmatched by traditional qualitative methods. As a result, we can only expect that the fashion industry will become increasingly reliant on AI and data-based methods.
Companies such as Platforme or Heuritch are already making use of this technology to estimate shelf time, demand, and trends for optimized production planning. As both companies equally emphasize, optimizing the process of trend-forecasting through AI and data is particularly relevant to the rising issue of sustainability within the fashion industry. Less waste from inaccurate predictions and trials as companies refine their clothing items to consumer preferences.
With consumers switching preferences just as quickly as new pieces and collections are produced, it is clear that using AI technology that skips the trial and error period will be essential for maintaining business efficiency in a digitizing world, particularly considering the sustainability benefits of doing so.
Komentarze