Retail Insight, the AI-powered retail operations platform, trusted by the world’s leading retailers, today announced new breakthroughs in machine learning algorithms that enable retailers to improve profitability on food products likely to end up being wasted.
An enhancement to its WasteInsight platform, the Predictive Waste feature proactively manages waste risk in-store, improving sell-through and sustainability. A new set of forward looking signals now include delivery schedules, item perishability data, store-level demand patterns and customer basket data. The solution helps retailers flag at-risk stock as soon as sales and inventory signals suggest a product will not sell through in time, enabling them to intervene with discounts days earlier than traditional waste management processes.
Research from 3,500 retailers estimates the cost of food waste is now over £400 billion ($540 billion) a year and it remains one of the biggest challenges facing retailers today. Most retailers run on fixed markdown schedules, applying the same number of markdowns to the same categories at the same point before expiry, regardless of actual trading conditions. The process has historically been manual and slow, requiring staff to walk the aisles and audit expiry dates by hand.
However, this very challenge represents an opportunity, as Alex Considine Tong, Chief Product Officer at Retail Insight explains, “Our Predictive Waste solution helps retailers address their waste and sell-through challenges earlier in the process, when the margin impact is larger.”
Store staff typically identify products nearing expiration the night before the final day they can be sold, at which point most of the margin opportunity is already gone, particularly on lines that were already overstocked or underselling. New machine learning techniques, combined with more granular enterprise data, now make it possible to identify which products are likely to become waste well before they reach that point. With the Predictive Waste solution, staff receive predictive markdown prompts up to a week before expiration. Retailers can now catch sell-through risk earlier, avoiding overly aggressive, last-minute discounting.
The technology also supports better assortment decisions. With Predictive Waste, assortment optimisation can happen in near real time, driven directly from the shelf.
“Instead of treating markdowns as an isolated event, sell-through signals can feed back into buying decisions, so teams can see where they have been consistently over ordering,” continued Considine Tong. “Retailers can pull back before the problem repeats and see real shifts in margin and sustainability.”
The Predictive Waste feature builds on WasteInsight’s dynamic markdown capabilities, which already help retailers find the ideal price to achieve their target margin and waste balance. WasteInsight pairs AI insights with clear, in-app prompts within a pre-defined process that directs store associates when to prioritise a markdown and by how much, to ensure optimal sell-through.
The Predictive Waste solution is part of a continued investment into Retail Insight’s AI-powered retail operations platform, which today processes 15% of the world’s grocery data across more than 68,000 stores across the world. Retail Insight’s platform processed $1.4 trillion of grocery revenue last year, delivering $795 million of gross profit improvement while saving 856 million meals from landfill.
