How Marks & Spencer uses AI to optimise Food and Clothing range planning

Marks & Spencer has integrated artificial intelligence into its merchandising operations to sharpen demand forecasting, optimise range planning and reduce waste across both its food and clothing divisions. The British retailer is using machine learning to predict customer preferences, allocate stock more precisely and support sustainability targets by cutting overproduction and markdowns.

What AI tools does M&S use for merchandising?

Marks & Spencer has deployed AI and machine learning across its supply chain and merchandising functions, focusing on demand forecasting and inventory management. The company has worked with technology partners to embed predictive analytics into its planning processes, enabling merchandisers to anticipate customer demand with greater accuracy and adjust product assortments accordingly.

The retailer has invested in data science capabilities to support its Reshape for Growth transformation programme, which aims to modernise store estate, digital platforms and operational efficiency. AI-driven tools analyse historical sales data, weather patterns, seasonal trends and emerging customer behaviour to inform buying decisions and stock allocation at store and regional level.

How does AI improve food range planning at M&S?

In its food business, M&S uses machine learning to forecast demand for fresh and ambient products, helping to reduce waste whilst maintaining availability. The company has focused on reducing food waste as part of its Plan A sustainability commitments, and AI plays a role in matching supply with customer demand more closely.

Predictive models assess factors including local demographics, store footfall, promotional activity and external events to determine optimal stock levels for each location. This granular approach helps the retailer avoid both out-of-stock situations and excess inventory that risks spoilage. The system continuously learns from sales outcomes, refining forecasts as new data becomes available.

How does AI support clothing and home merchandising?

For clothing and home products, M&S applies AI to range planning, size allocation and markdown optimisation. Machine learning algorithms analyse past performance of similar styles, fabric types and price points to guide buying teams on assortment depth and breadth. The technology also supports dynamic pricing and promotional planning, identifying which products are likely to require markdown support and when.

M&S has invested in improving its clothing ranges and reducing excess stock as part of efforts to restore profitability in its general merchandise division. AI-driven insights help merchandisers make faster, more informed decisions about product mix, reducing the risk of overbuying slow-selling lines whilst ensuring adequate stock of high-demand items.

The retailer uses predictive analytics to optimise size curves by location, recognising that customer demographics and preferences vary across its store network. This localised approach reduces the likelihood of size-specific stockouts or surplus, improving both sales conversion and inventory efficiency.

What are the operational benefits of AI in range planning?

AI-driven merchandising delivers several measurable benefits for M&S:

• Improved forecast accuracy: Machine learning models reduce forecasting error by identifying patterns human planners might miss, leading to better alignment between supply and demand.
• Reduced markdowns: More precise buying decisions and dynamic pricing strategies minimise the volume of end-of-season stock requiring heavy discounting.
• Lower waste: In food, tighter demand forecasting reduces spoilage and waste, supporting environmental targets and protecting margins.
• Enhanced availability: Optimised stock allocation ensures popular products remain in stock, reducing lost sales and improving customer satisfaction.
• Faster decision-making: Automated data analysis frees merchandising teams to focus on strategic choices rather than manual number-crunching.

These efficiencies contribute to improved working capital management, as the retailer holds less excess inventory and turns stock more quickly. The operational improvements also support M&S's efforts to compete more effectively with digitally native retailers and fast-fashion competitors that have built data-driven merchandising into their business models from the outset.

How does AI merchandising support sustainability goals?

M&S has committed to reducing waste and environmental impact through its Plan A programme, and AI-enabled range planning contributes directly to these objectives. By producing and stocking only what is likely to sell, the retailer reduces overproduction, unsold inventory and the associated carbon footprint of manufacturing, transporting and disposing of unwanted goods.

In clothing, more accurate demand forecasting helps avoid the industry-wide problem of excess stock that ends up in landfill or incineration. In food, reducing waste aligns with broader efforts across the grocery sector to minimise the environmental and economic costs of spoilage. AI tools also enable M&S to optimise logistics and distribution, reducing fuel consumption and emissions by ensuring the right products reach the right locations without unnecessary redistribution.

The integration of AI into merchandising represents part of a wider digital transformation at M&S, as the retailer seeks to modernise legacy systems and processes. The company's Reshape for Growth programme has prioritised investment in technology and data capabilities, recognising that data-driven decision-making is essential for competing in a rapidly evolving retail environment.

While M&S has made progress in deploying AI for range planning and demand forecasting, the retailer continues to refine these systems and expand their application across the business. The effectiveness of machine learning models depends on data quality, integration across systems and ongoing human oversight to ensure algorithms reflect business strategy and customer needs. As the technology matures and M&S accumulates more transactional data, the precision and impact of AI-driven merchandising are expected to increase, supporting both commercial performance and sustainability outcomes.