amazon app update stylesnap

The days of fashion retail being a side hobby for major technology companies are over. And experts might someday reflect on this week as one of the major thresholds in the transition.

Instagram and Amazon tipped their hands thanks to new updates that aim to boost fashion discovery. The photo-sharing network just expanded brands’ ability to promote influencer content, while the e-commerce juggernaut juiced up intelligent style identification and recommendations with a new feature called StyleSnap.

The two companies are often uttered in the same breath, as examples of tech giants going after fashion commerce and discovery through a variety of routes.

Instagram has gone from mere inspiration destination to e-commerce enabler, complete with shopping carts and checkout features. The company recently allowed influencers to share shoppable posts, and with its just-announced branded content ads, brands will be able to promote influencer content as advertisements in feeds and Stories.

Ultimately, the latest change means users will see content from people or companies they don’t follow, and marketers see vast opportunity in the update. “Paid media tools from a platform, like Instagram, are revolutionary, and for us, they help to produce better return on ad spend, foot traffic into stores and sales,” said Ryan Detert, chief executive officer of Influential, an artificial intelligence-driven influencer marketing platform.

The latest change might even give Instagram a leg up on Amazon — at least on the premium end.

“With this move Instagram is becoming the higher-end, ‘luxury Amazon,’ which Amazon could never achieve because the IG platform is so visually stimulating and elevated,” said Permele Doyle, president and founder of Billion Dollar Boy, an influencer marketing firm that serves clients like Estée Lauder, Armani Exchange and L’Oréal, among others.

“[The update] positions Instagram as a major player in the e-commerce space, shifting the original focus of the platform from visual discovery to a place for direct conversion,” she said. “It wants to become the shop-front.”

But Doyle also believes the tactic is risky. The change moves the emphasis even further away from organic content. “This has already happened, as Instagram has been moving toward favoring paid-for media, and this move will only increase that,” she said.

While some Insta users might be pleased with more influencer-driven messaging and shopping, the changes could alienate others that look to the platform strictly for inspiration or entertainment. And, Doyle added, It could also compromise how followers perceive the authenticity of their idols, which is the lifeblood of social influence.

Indeed, mixing experiences and intentions on social media is tricky business. Snap Inc. discovered that last year when it fought off a severe backlash to an app redesign, which blurred the lines between Stories, communications from friends and the professional content in Discover.

Instagram also contends with the extra scrutiny of being a Facebook company. The update could funnel more data and insights into user behavior during a time when its parent company sits on the brink of more oversight and investigation on privacy and anticompetitive practices.

Regulators have a sharp eye on other giants, like Amazon, as well, though its fashion ambitions are nothing new. Amazon has been working for years on numerous ways to build up its style cred, from launching private labels to the Echo Look selfie fashion camera to the Prime Wardrobe service to its influencer-driven The Drop distribution model and its latest StyleSnap app update.

StyleSnap, revealed Wednesday morning at Amazon’s first Re:Mars conference on artificial intelligence and machine learning, lets users snap or upload images and receive fashion recommendations, thanks to machine learning.

The goal is to help shoppers who “struggle to find styles they can’t describe in words,” said Jeff Wilke, Amazon’s chief executive officer of consumer business, at the Re:Mars conference in Las Vegas.

Amazon’s app already offers a camera feature that lets users snap and upload photos of books and other products, and its tech does an impressive job of analyzing the images and searching the items in the marketplace.

Now pointed directly at apparel, accessories and footwear, the move is an overt attempt to make Amazon a major fashion destination for shoppers looking to identify styles and buy them.

“When a customer uploads an image, we use deep learning for object detection to identify the various apparel items in the image and categorize them into classes like dresses or shirts,” Wilke explained onstage. “We then find the most similar items that are available on Amazon.”

Users simply click the camera icon in the Amazon app to take a photo or upload a screen grab of an outfit seen online, and it pulls up recommendations based on factors like price, reviews and brands.

The premise seems simple, and it might be for human beings. But for machines, it takes a lot of work to master.

Here’s how Amazon explained the process in a blog post on Wednesday: “To have neural networks identify a greater number of classes, we can stack a greater number of layers on top of each other,” Amazon’s Arun Krishnan wrote. “The first few layers typically learn concepts such as edges and colors, while the middle layers identify patterns such as ‘floral’ or ‘denim.’ After having passed through all of the layers, the algorithm can accurately identify concepts like fit and outfit style in an image.”

The two strategies — people-powered and machine-driven — looks like two different vectors of approach with the same goal: To reach more fashion consumers and energize product discovery on these platforms.

Whether these particular updates take off and become major successes matters perhaps less than the fact that these massive platforms are aiming to fundamentally change the way consumers find and purchase the objects of their desire.

load comments
blog comments powered by Disqus