Fashion purchases are heavily influenced by human emotions, past acquisitions of similar items, and shifts in trends. Simultaneously, technology has always played a pivotal role in both the business sphere and daily life—ranging from CRM tools in the 1980s, the emergence of e-commerce in the early 1990s, to programmatic advertising in the 2010s, and more recently, the dominance of Artificial Intelligence (AI) in the fashion industry, particularly in retail.

According to the Boston Consulting Group, retailers traditionally stocked shelves and ordered products 10 months in advance, with an approximate lead time of 37 to 45 weeks, depending on the style, design, colour, and size preferred by customers. However, inaccuracies in demand forecasting could lead to substantial losses. To mitigate this, retailers are increasingly leveraging new technologies to keep pace with emerging trends.

At a recent conference hosted by the National Retail Federation in New York, discussions centred around the future of retail, which is expected to be shaped by a combination of AI, robots, and holograms. The fashion retail sector has already seen the introduction of robots for warehouse logistics and augmented reality mirrors in fitting rooms to assist with fit and inventory management.

The fashion industry is beginning to harness new technologies. McKinsey & Company predicts that over the next three to five years, the industry could see profits ranging from $150 billion to $275 billion, primarily within the fashion, apparel, and luxury sectors. AI’s incorporation is increasingly evident in areas such as design, marketing, and customer service.

AI has recently attracted significant attention from fashion retail leaders, with many experimenting with the technology in various capacities. Innovations include the creation of digital product passports like Mugler and software solutions such as Replika for LVMH’s startup accelerator, enhancing retail operations. The primary goal for retailers is to engage the right customer with the appropriate product at the optimal time, utilising valuable insights from potential customers. AI significantly enhances the accuracy of these predictive processes through data-driven technology.

Four key strategic areas identified are product design, copywriting, visual content creation, and customer experience. Technology providers like Amazon, Shopify, Salesforce, and other AI specialists are facilitating the learning process for fashion retailers such as Levi’s, Frame, Snipes, and Casablanca. The adoption of new technology presents opportunities, challenges, and risks for both employees and brands. Nonetheless, technology continues to drive rapid advancements across the fashion industry’s major sectors.

Capabilities of Generative AI

Generative AI, which employs machine learning algorithms to create new content, is significantly impacting fashion brands. These algorithms, trained on numerous examples, can autonomously recognise various patterns and structures. Tools like Stable Diffusion, Midjourney, and DALL-E 2 are capable of generating images through these algorithms.

Furthermore, AI plays a crucial role in analysing store foot traffic to identify patterns that lead to higher conversion rates. It aids in understanding customers’ psychographic profiles and tailor’s product recommendations to meet individual demands.

Generative AI has become a highly discussed technology, offering styling advice to customers. It can produce text and image content for marketing campaigns, personal blogs, and e-commerce websites. Additionally, advanced chatbots facilitate interactions between online shoppers and AI technology, providing data-driven insights.

AI also enables retailers to efficiently manage their inventory, ensuring stores are well-stocked with items customers are likely to purchase. Psykhe, for instance, utilises AI to recommend the most suitable products to potential customers based on data-driven analysis. This technology significantly reduces the time required to complete labour-intensive tasks.

Typeface.ai focuses on creating personalised content, aiding retailers in generating text and images for various purposes, including blog posts, e-commerce landing pages, email marketing campaigns, and Google display ads. Another AI tool, Booth.ai, concentrates on photography, capable of capturing product images and rendering them on models against selected backgrounds.

Sparkbox, an AI-centric retail planning platform, enables retailers to forecast demand and manage in-season reorders effectively. It also assists in optimising product pricing. Shopify, a widely used retail point-of-sale system, provides access to real-time retail data and features its own AI capabilities, such as Shopify Magic, which helps users generate product descriptions. These technologies assist retailers in determining production and sales volumes for their stores, although understanding consumer behaviour still requires technological intervention.

Challenges of Artificial Intelligence

Retailers may encounter challenges when implementing AI, as it can be both expensive and time-consuming. AI predictions are probabilistic, with around 40 per cent possibly coinciding with historical data. However, AI may struggle to predict new data or styles of products to be stocked in stores and might not sync with real-time data. Additionally, it is crucial that employees in retail stores receive appropriate training over time. Jaded London highlights difficulties in initial buy planning due to trends significantly influenced by celebrities and influencers on social media. These factors make demand prediction complex, presenting a challenge in effectively leveraging AI in retail.

New Phase (Face) of RFID

RFID (Radio-Frequency Identification) is a modern technique increasingly adopted by retailers to monitor their inventory. This decades-old technology has the potential to revolutionise retail by allowing companies to track items, such as socks or jumpers, quickly and accurately, whether they are on the shop floor or in a warehouse. This eliminates the need for manual counting of items, a concept that gained attention in the early 2000s. However, only a few retailers initially adopted RFID, deterred by implementation challenges and high costs.

Research indicates that RFID can significantly enhance inventory accuracy, positively impacting sales even with reduced stock levels. Yet, its integration remains complex. Companies are exploring new uses for RFID, such as self-checkout, combining it with AI, and gaining deeper insights into customer interactions with products in-store, aiming to derive greater value from the technology. Despite sparking considerable debate, RFID continues to gain traction as costs decrease, accuracy improves, and the need for precise cross-channel inventory tracking grows.

Each year, an increasing number of businesses join the ranks of RFID users. Notable examples include Zara, Uniqlo, and American Eagle, which leverage the technology for enhanced inventory management and self-checkout capabilities, among others.

How do Big Brands do it?

Inditex

The parent company of Zara, Inditex, began rolling out RFID technology across all its brands in 2014, embedding chips within the plastic security tags of its products. This move has since become a cornerstone of the company’s operations. The adoption of RFID at Zara has not only freed up employee hours but also simplified in-store replenishment and maximised sales at full price.

In March of this year, Inditex announced a shift away from traditional hard security tags in favour of embedding RFID chips directly into the garments. This initiative aims to alleviate one of Zara’s major pain points: long waiting times for customers. By facilitating self-scan checkouts, this change is expected to significantly reduce customer friction. Despite Inditex’s full commitment to RFID technology, other retailers have encountered difficulties with its adoption, citing the labour and resources required to tag merchandise and integrate RFID data with existing systems.

At Zara’s store in New York’s SoHo district, a long counter on the second floor features large bins for customers to drop off their purchases. Above each bin, screens welcome users to the self-checkout process. Traditionally, customers would scan the barcode on each item’s tag to check out, but these bins automatically detect the clothing and calculate the total cost. Customers simply follow the on-screen instructions to pay, remove security tags, and leave with their purchases. Sales assistants are on hand to offer help, particularly with detaching security tags.

For item identification, customers use a wand to scan tagged items, enabling rapid and frequent updates. This system provides real-time insights into stock levels, identifying items that need restocking, those that are out of stock, and those not selling well. This technology has the potential to replace traditional inventory audits, which required hours of manual counting every six or twelve months, with more efficient and accurate stock management.

Uniqlo
Uniqlo’s self-checkout system has been instrumental in reducing instances of out-of-stock items on the sales floor. During recent earnings calls, companies such as Nordstrom, Macy’s, and American Eagle have highlighted the benefits of this technology, ranging from enhanced inventory management to theft prevention. The adoption of RFID technology has increased in recent years, driven by the reduced cost and improved accuracy of chips and readers, as well as businesses’ growing reliance on omnichannel strategies, necessitating accurate inventory tracking across various channels.

McKinsey & Co reports that RFID technology can improve inventory accuracy by more than 25 per cent. A study conducted by the ECR Retail Loss Group, which consists of manufacturers and retailers, observed similar accuracy improvements across ten retailers. The study highlighted benefits including reduced labour costs and increased sales with lower stock levels, with companies like Adidas, C&A, Lululemon, and Marks & Spencer among those reaping these advantages.

However, ECR found that implementing and integrating RFID technology presented significant challenges. The most formidable obstacle was integrating RFID-generated data with existing retail systems. Companies also encountered issues if not every item was tagged immediately, resulting in multiple data streams. Ensuring the technology delivered results required a sustained commitment and resources. The time savings gained from employees passing items over an RFID reader, as opposed to manually scanning a barcode on each item, have made a notable difference. Additionally, the data collected has enabled companies to optimise product routing, further enhancing operational efficiency.

RFID’s Future Frontiers

Retailers are exploring new uses for RFID technology, such as the self-checkout options implemented by Zara and Uniqlo, to maximise its value. American Eagle announced plans to deploy a new system across its stores, integrating computer vision—a form of artificial intelligence—with RFID, following a pilot that demonstrated over 99 per cent accuracy in tracking product location and availability.

American Eagle employs radar sensors, roughly the size of a toilet seat, mounted on the store’s ceiling at regular intervals, equipped with cameras. These sensors, in conjunction with RFID tags, provide a real-time inventory snapshot accessible via a 3D interface. This setup allows staff to pinpoint items’ locations without needing an RFID wand. The technology furnishing this capability to American Eagle underscores the innovative approaches being adopted.

Companies like Macy’s are utilising RFID scanners at store exits to assist with theft management—not to prevent it, but to collect data on stolen items. This allows stores to promptly remove stolen items from their inventory system, helping avoid order fulfilment errors or misleading customers about item availability. Additionally, analysing theft data helps identify frequently targeted products, enabling retailers to enhance security measures for those items.

Another emerging trend involves installing readers at the entrance of fitting rooms to monitor which items are tried on and, subsequently, purchased or not. However, the true advantage of these insights depends on their application. RFID technology, in essence, is a data generation tool; it does not yield benefits independently. The value derived from the data hinges on how effectively companies utilise it. Retailers that understand the real-time dynamics of the industry and recognise how technology can enhance business efficiency stand to gain significant profits.