When you see someone wearing something nice, how many times have you been tempted to ask questions like “from where did you get this? which brand?" I am sure it happens all the time. The social element in fashion is so high that recommendations are becoming important in making the right choices. This stands even more true for GenZ and Millennials.
In the online fashion world, recommendations generically mean options like “Shop Similar” — which is basically pulling out similar styles from the catalog. OR the ones based on image search wherein image recognition techniques scan through zillions of images to make relevant recommendations.
However, ecommerce companies need to step up their game by making it more personalized. Although, CTR (click through rate) of email-based recommendations is quite high, but it does nothing for enhancing the customer experience.
All the major tech companies like Google (YouTube, google news) Apple music, Netflix, etc. use recommendation algorithms because it provides huge scope for personalization by processing vast amounts of data at minimal costs. Such, artificial intelligence (AI)-powered recommendation engines are quite common in the areas of fashion too and are largely based on users’ browsing history/mapping similar users, etc. But the question is can fashion recommendations be truly automated and eliminate human curation?
And, what could be the future for fashion recommender modules — Voice? Machine learning is already at an advanced stage in the text and image format, and combined with natural language processing if it can get similar command over voice then an efficient "emotional commerce" experience can truly be achieved. Say, I need a new dress for an office event or a girls night out and if I can interact back and forth with the technology platform through voice and emotions to get good recommendations then voila! I need not scan through massive catalogs of fashion e-commerce companies.
As per a Wall Street article, voice could become the next fingerprint and technology should be able to detect fine nuances in the voice which are inaudible to humans. So, it can determine a lot about personality/physical features, etc., but it’s still an evolving technology and has a long way to go. Until then we can only rely on data and stylists both working hand in hand to scale up our fashion quotient.