ViSenze Blog

The Economics Of Fashion Retail: Improving The Bottom Line With Machine Learning

Posted by Oliver Tan 03-July-2017

This article was originally published in Forbes on 16th May 2017.

If you are a fashion retailer who is unfamiliar with machine learning, your education needs to start now.

In recent years, we saw the convergence of three important technology trends that will make machine learning indispensable for fashion retailers: 1) Software became widely available to gather and organize massive amounts of data, 2) computers upgraded to offer impressive processing power and 3) the price of storage dropped dramatically. Now you can finally afford to put all those terabytes of data that you collected to work to improve your operational efficiency and save money.


Here are the five areas where machine learning can most improve your bottom line.

Inventory Management

Let’s start with the single most expensive and risky element of any retailer’s business: inventory. According to Mark Jarecke, founder and creative director of agency Four32C, machine learning can optimize purchase flow, provide efficiencies and insights, and create more personalized experiences. A machine will regularly and systematically crawl the internet gathering data so that you have constant visibility into what, when and how much people are buying, competitors’ pricing, and the styles, colors and fabrics that are trending on social media. It will also process demographic characteristics so that you can adjust your ordering by geographic location for your brick-and-mortar locations.

What could it mean for your business if you could reduce waste and out-of-stock delays by 2-3% or more?

Competitive Pricing

A machine’s ability to calculate the right price on a per-customer basis has the potential to change online retail as we know it. What if you knew enough about every shopper’s buying habits that you could understand their pricing tolerance on each item? What if you could offer a bundle of currently trending items on the fly and do so faster than the competition? Both processes would significantly increase the likelihood of a sale. And the more time a shopper spends on your site, the smarter the machine will be in personalizing their shopping experience and improving their satisfaction with your brand.

Reducing Costs For Customer Service And Design

Chatbots powered by machine learning will play a greater role in helping shoppers resolve issues and find exactly what they seek more quickly. Almost 50% of consumers prefer to contact businesses through messaging apps (versus email or calls), further underscoring their desire for immediate resolution. Fashion retailer Tommy Hilfiger has been using chatbots for customer service applications for years, but thanks to language usage advancements in machine learning, we are now seeing “conversational bots” used for more advanced interactions and purchases, reducing the number of requests that require human assistance.

Advancements in machine learning can also significantly lower the cost and time necessary for a design cycle. While much of the traditional design process is trial and error, machines can provide the data that will reduce the uncertainty. Sites like Google, Pinterest, Elle, Polyvore, Vogue and Baidu gather millions of images every month that assist designers in selecting the optimal styles, fabrics and colors for the coming season based on what is trending and what shoppers are buying.

Marketing ROI

Marketing and advertising are among retailers’ largest budget items. Machine learning can offer the information necessary to lower the cost per conversion and increase average purchase amounts. The virtual assistant offered by Shopify, for example, helps its retail customers improve their SEO and email campaigns with recommendations based on previous results. Additionally, Cosabella uses machine learning to determine offers for its email campaigns, what ads to use where and the layout of its web pages. CEO Guido Campello says that the smarter the company becomes with AI, the longer customers remain loyal.

Cost Of Logistics

Retailers should seek logistics partners who are tech-savvy and investing in the connected technologies made smarter by machines. A fleet management system, for example, will improve routing, reduce fuel waste and better maintain vehicle health and safety. Machine learning is also putting driverless vehicles on the road, which will lower the cost of transportation as well as reduce accidents, fuel waste and traffic delays.

What’s Next?

Let me wrap up with one important piece of advice: Don’t go it alone! Most fashion retail companies don’t have a room full of data scientists who are experts in human behavior, marketing analytics and logistics. There is a large (and growing) cadre of technology companies that are focused on artificial intelligence and machine learning applications for e-commerce. Look for one that has technology platforms and specialists necessary to identify the right solutions for your company’s unique needs.

This article was originally published in Forbes on 16th May 2017. The original article can be found here 

Image: Forbes/Shutterstock

  Deep learning,   E-commerce