While his taxi sped away from the airport in Bangalore, India, Ranjan Sharma’s mind was also racing to solve a riddle. Bestseller India, where he serves as CIO and Head of Supply Chain, was the fastest growing fashion retailer in the country—so why were the sales of his top label slowing down?
It was the summer of 2019 and the overall performance of Bestseller India was impressive. Ever since launching 11 years ago, the company grew at an average rate of over 50% a year. At more than 1,500 stores across India, customers snapped up popular clothing brands such as Jack & Jones, Only, Selected and Vero Moda.
But then, Bestseller India’s top-selling Only brand, designed for the burgeoning youth market, encountered headwinds. “Only had suddenly gone down and there was nothing that we could figure out in terms of what went wrong,” says Sharma. “We were puzzled and had to find out why it was going down.”
Sharma and his team traveled from Mumbai to Bangalore, India’s IT capital, for answers. “We were looking for a partner,” Sharma says, “who could not only bring in the new technologies, but also understood the fashion space well.”
Technology notwithstanding, understanding the fashion industry in India is notoriously difficult. Within its population of almost 1.4 billion people, India encompasses hundreds of dialects, religions and ethnic groups. Consumer preferences change markedly from town to town. Tracking micro-segmented markets at this scale challenges even the best and most experienced product planners and fashion merchandisers.
Bestseller India is a subsidiary of Bestseller, a worldwide retailer based in Denmark that is a leader in “fast fashion”—a dynamic business model that moves trendy clothes from the runway to the rack in a matter of days or weeks. When fast fashion hits the mark, fresh clothing designs fly off the shelves. But if the new designs miss with consumers, inventory gets marked down—and some of it ends up in landfills. In fact, for the global fashion industry as a whole, 20% of clothing produced eventually goes unsold.
The apparel manufacturing industry is also a major consumer of raw materials, water and energy. For example, the production of a single cotton t-shirt requires up to four liters of water. By more closely aligning consumer demand with design and production, the fashion industry can improve profits while also supporting environmental sustainability.
Historically, most fashion retailers tend to rely upon past experience and “gut” instinct to decide which designs to manufacture, what quantities to produce, and where to market their wares. But the diversity and dynamism of the Indian market revealed the limitations of a gut strategy, especially when Bestseller India looked closely into why its top-selling label was no longer performing as expected.
As Sharma went into his next meeting in Bangalore, he knew he needed a technological advantage for his designers and buyers to develop better forecasts and deliver the right product at the right time. And when he finally sat down with the experts at the IBM Research Laboratory he got right to the point: “Could IBM help us solve these problems by bringing in insights from data with AI?”
1st in Industry
AI Powered fashion planning and forecasting tool 1st in India fashion industry
Enhanced Presence
AI Solution now being extended to 3 more brands
Confident that IBM offered the right combination of IT experience as well as retail industry acumen, Sharma and his Bestseller India team soon expanded the scope of what they were looking for. In addition to turning around sales for the Only brand, Sharma recognized an opportunity to transform business processes by tapping the competitive advantages of AI-powered decision-making.
For its initial IBM® Cognitive Enterprise innovation project with IBM Consulting™, Bestseller India set a very ambitious goal: develop a totally new, bespoke platform with AI capabilities to support preseason design, planning, production and forecasting—essentially modernizing processes by empowering employees with technology. “This was a big change for us, for the planning teams to move from a gut feeling to AI-delivered information that we wanted to implement,” says Sameer Ambalkar, Head of Business Solutions at Bestseller India.
To support such a big change for its company and employees, Bestseller India was eager to collaborate with experts who could introduce innovative practices and new ways of working. So, the company chose to work with the IBM Garage™—a proven development framework that integrates people, processes and technology to transform business and culture. During the IBM Garage Enterprise Design Thinking™ Workshop, IBM and Bestseller India experts created a roadmap for co-creation, beginning with research on how users would interact with the technology.
“Collaboration started right from the outset in terms of getting inputs from designers, merchandisers, buyers, brand heads and IT professionals—the team that would actually be using this tool,” says Zian Lakdawalla, Business Transformation Manager at Bestseller India. “There were personas created for designers and buyers, and all of these ideas led to the development of the project.”
IBM Watson® AI tools were used to help predict the best products for new offerings, determine the right product mix for each store and improve the efficiency of the supply chain. “IBM Garage is a fantastic model to work with,” says Sharma. “It cuts across technologies and domains to get a better solution and leverage what is ready.”
The project focused on creating intelligent workflows for key business processes, empowering employees to work smarter and use their time more efficiently with access to real-time data and insights through AI-powered tools. “When IBM said that we needed to bring technology into the design process, it was a big ‘aha’ moment,” says Mukta Srivastava, Only Brand Product Manager at Bestseller India. “Now, designers will spend more time on higher value work instead of managing files and data.”
The enormous size of the Indian economy and large differences between individual markets was a huge challenge for the development team, but a challenge that could be addressed within the IBM Garage framework. “We have a lot of demand that varies from one region to another and understanding that was very key for us,” says Lakdawalla. “We wanted to see how good the tool could be to predict better sell-through rates and see what is really trending to maximize our investment.”
To support software development, Bestseller India chose the IBM Cloud® Kubernetes Service, a managed platform as a service (PaaS) offering that met Bestseller India’s requirements for maximum availability and flexibility.
After months of work and iteration, the Bestseller India-IBM Garage team brainstormed 61 unique concepts for the platform—called Fabric.ai—which became the first AI-powered tool for the Indian fashion industry.
The first version of Fabric.ai included a comprehensive set of seven core AI modules, along with six specific tools for designers, buyers and merchandisers. Whether users accessed data from previous seasons or updated data for future use, Fabric.ai enabled immediate cognitive analyses of how well products are performing.
The user experience was further enhanced by an interface that displays sales and product information in an easy-to-use visual format. Per the original design brief, Fabric.ai initially focused on the Only line of apparel, but the platform has the scalability to handle Jack & Jones, Vero Moda and other brands in the future.
“The platform is well thought-through and robust in terms of understanding fashion, which helps bring in data in a more predictable and easier format for all users to engage in,” says Ambalkar. “The model has been the crux of this, and IBM helped make it possible.”
Designers appreciated the ability to use a visual similarity tool that compares new products with products from previous seasons. “Fabric.ai will help us pore through information that is relevant and in a pictorial form, which is much easier for designers to consume than looking at spreadsheets,” says Srivastava. “It will become much easier for them to deliver a product that is closer to customer demand and make historical performance information available at the click of a button.”
Fabric.ai also provided a better look at the sales performance of specific products at the retail store level. “The store-view option will help us find the right assortment for each store to help the buying and merchandising team,” says Lakdawalla. “In addition to helping predict which products will be successful in the next season, the tool will help us focus on the supply chain, prioritize the sell-through of current products and help ensure that inventory doesn’t pile up.”
With many stores closed because of the COVID-19 pandemic and large segments of the retail industry being put on hold, having Fabric.ai in place will provide a head start when normal business conditions resume. “At some point, we had to start using AI technology and leverage it to develop better performing product ranges,” says Srivastava. “With IBM, it has been a great journey and it is going to bring in a lot difference in how we’re delivering and how the ranges perform with our customers.”
In 2019, Bestseller launched “Fashion FWD,” a long-term corporate initiative designed to bring sustainable fashion forward. The strategy is based on the belief that sustainability is a prerequisite for ongoing business success.
Sustainability begins at the design phase, where digital design processes can minimize waste at the start of the creative process. With Fabric.ai optimized for designers, Bestseller India has a digital platform to help inform more sustainable material choices up-front in the value chain.
Fabric.ai also provides product planners with data-driven perspectives on producing clothing with a leaner environmental footprint.
Bestseller India’s deployment of Fabric.AI coincides with another important Fashion FWD supply chain initiative in India related to “crop-to-shop” sustainability. In August 2020, Bestseller signed agreements with 2,000 organic cotton farmers in India to source 1,500 tons of organic cotton grown without chemical pesticides and fertilizers. That’s enough cotton to produce three million t-shirts.
Like many nations around the world, India has been hit hard by the COVID-19 disease and the pandemic has disrupted the retail marketplace where Bestseller India serves its customers. However, even as many Bestseller India employees work remotely during the pandemic, development continues on Fabric.ai because IBM’s cloud-based Garage Methodology supports virtual development and collaboration.
“We continue to provide the support that the team needs,” says Lakdawalla. “It will just build up as the days and months progress. To take this process forward, that is what I feel will be the biggest game-changer.”
“Given the current situation, this is not just a need but a must-have today,” says Sharma. “The shops and the warehouses are changing. How will we be able to create new business models which can solve current and future problems and reshape our business? These are issues we will continue to work on with IBM.”
Based in Mumbai, Bestseller India (link resides outside of ibm.com) operates 250 exclusive outlets and also sells its brands through over 1,500 external stores. With over 3,500 employees in India, the company is part of the parent Bestseller organization based in Denmark. Founded in 1975, privately-owned Bestseller operates in 70 markets around the world through more than 2,800 chain stores and 12,000 external multi-brand stores.
Legal
© Copyright IBM Corporation 2020. IBM Corporation, IBM Consulting, New Orchard Road, Armonk, NY 10504
Produced in the United States of America, December 2020.
IBM, the IBM logo, ibm.com, Enterprise Design Thinking, IBM Cloud, IBM Consulting, IBM Garage, and IBM Watson are trademarks of International Business Machines Corp., registered in many jurisdictions worldwide. Other product and service names might be trademarks of IBM or other companies. A current list of IBM trademarks is available on the web at ibm.com/trademark.
This document is current as of the initial date of publication and may be changed by IBM at any time. Not all offerings are available in every country in which IBM operates.
The performance data and client examples cited are presented for illustrative purposes only. Actual performance results may vary depending on specific configurations and operating conditions. THE INFORMATION IN THIS DOCUMENT IS PROVIDED “AS IS” WITHOUT ANY WARRANTY, EXPRESS OR IMPLIED, INCLUDING WITHOUT ANY WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND ANY WARRANTY OR CONDITION OF NON-INFRINGEMENT. IBM products are warranted according to the terms and conditions of the agreements under which they are provided.