Thomas Scott pilots two GenAI tools for fashion retailing

Thomas Scott (India) Ltd. is piloting two Generative Artificial Intelligence (GenAI) tools, thread.ai and catalog.ai, after building a centralized data-driven backend over the past two years; the tools extend a retail process that can test new styles at about 120 minimum order quantity (MOQ) units before inventory is scaled.

What GenAI tools is Thomas Scott piloting?

Thomas Scott is piloting thread.ai for fashion and lifestyle trend intelligence and catalog.ai for e-commerce visuals and product-listing work. The presentation describes thread.ai as an intelligent GenAI co-pilot that identifies trending and high-demand styles, providing data-driven insights for design, pricing and merchandising strategy.

Catalog.ai is the second pilot-stage component of Thomas Scott's proposed formalized application. It analyses high-conversion fashion and lifestyle styles, automates product shoots with artificial intelligence (AI) models and backgrounds, streamlines editing, and manages e-commerce catalog tasks. The presentation does not disclose a rollout date, user count, spending amount, or revenue target for either pilot.

How does Thomas Scott identify fashion demand before scaling inventory?

Thomas Scott identifies demand by collecting high-volume keyword data, high-traffic browsing-page data and competitive information on major marketplaces. Its algorithms are intended to find high-demand trends and online micro-markets with supply-demand gaps before the company launches products across brands.

The process begins with minimal inventory rather than a large initial production commitment. Thomas Scott launches multiple products, monitors customer demand and satisfaction, and then increases inventory bets on best-performing products with high customer satisfaction. The presentation says algorithms guide decisions including demand, inventory, pricing and marketing, linking the two GenAI pilots to the existing data-led process.

What manufacturing model supports Thomas Scott's trend-led retail process?

Thomas Scott uses a high-width, low-depth small-batch manufacturing strategy, under which a new style can be piloted at about 120 MOQ units. The company says successful pilots can be scaled up, making the initial batch a test of demand before a larger inventory commitment.

The model combines in-house manufacturing with direct-to-consumer (D2C) distribution. Thomas Scott says the D2C structure lowers costs, improves competitiveness and reduces markdown losses, while its pan-India fulfilment centres support two-day deliveries. The presentation also cites manufacturing expertise as a mechanism for efficient production, product quality and fewer returns.

What operating base supports Thomas Scott's technology plan?

Thomas Scott says it built a centralized, data-driven backend over the past two years to support operations, design, cataloguing, brand management and merchandising across multiple brands. Management's stated focus is to formalize that technology into a more robust application with an enhanced user experience and integrated GenAI capabilities.

The centralized operating structure covers five functions: operations, logistics and fulfilment; creative, design and catalog; brand management and analytics; finance and control; and production and sourcing. Thomas Scott has more than 35,800 stock-keeping units across more than 15 brands and more than nine online and offline channels, providing the operational scope for this shared backend.

Thomas Scott's online-focused retail channels include Myntra, Ajio, Tata Cliq, Amazon, Flipkart and its own website. These marketplaces are part of the company's data-collection process, alongside its four manufacturing units and four fulfilment centres. Total fulfilment capacity is listed at 15,000 pieces per day.

How has Thomas Scott's income changed as its retail model expanded?

Thomas Scott reported operational income of Rs 254.9 crore in FY26, compared with Rs 161 crore in FY25 and Rs 91.1 crore in FY24. Operational income increased by Rs 93.9 crore from FY25 to FY26, following a Rs 69.9 crore increase from FY24 to FY25. The presentation does not attribute those changes specifically to thread.ai or catalog.ai.

Earnings before interest, tax, depreciation and amortisation (EBITDA) margin fell by 2 percentage points from 14.05% in FY24 to 12.05% in FY25, then rose by 1.05 percentage points to 13.10% in FY26. In H1-FY27, the reported EBITDA margin was 13.13%, while business-to-consumer (B2C) revenue accounted for 96% of revenue and business-to-business (B2B) accounted for 4%.

Thomas Scott lists monthly manufacturing capacity of 60,000 bottoms, 60,000 shirts and 20,000 bags. The company says its distribution network is designed for the majority of orders to arrive within a maximum of two days, with same-day or next-day delivery in major urban areas. This fulfilment mechanism supports the D2C model alongside the data-led product-testing process.

Conclusion

Thomas Scott is moving from a centralized data-driven retail backend to two named GenAI pilots: thread.ai for trend identification and design, pricing and merchandising inputs, and catalog.ai for e-commerce visual and catalog workflows. The pilots fit a model that collects marketplace data, tests styles at about 120 MOQ units, monitors demand and satisfaction, and scales inventory for products that perform.

The next disclosed development to watch is management's plan to formalize its technology into a more robust application with an enhanced user experience and GenAI integration. The presentation gives no launch timetable, adoption measure or financial target for thread.ai or catalog.ai, leaving future disclosures on pilot deployment, catalog use and inventory outcomes as the relevant indicators of progress.