The recently released report ‘Growing and Sustaining Competitive Advantage in Grocery Retail’ shows that for the companies participating in this survey, the annual value of spoilage was on average around 70 million and up to several hundred million annually for the largest companies offering a wide range of fresh products.
The survey responses also demonstrate that grocery retailers with less advanced store replenishment practices are almost five times more likely to suffer from above-average spoilage levels than their competitors who use forecast-driven store replenishment extensively.
Yet 20% of the respondents haven’t even started implementing any kind of replenishment automation in their stores and only 30% of the companies have implemented forecast-based automated store replenishment extensively in the categories they offer. This means that there is a lot of money on the table, which is good news for the hard-pressed food retailers.
RELEX’s experience is that grocery retailers typically attain spoilage reductions of 10-30% by implementing forecast-based automatic store replenishment. For example, German grocery retailer Bünting attained 24% reduced waste, while significantly increasing product availability and saving hundreds of hours of work in stores. Furthermore, as the scale and complexity of the operations increase, the attainable benefits also increase.
Key Considerations When Automating Replenishment of Fresh Products
The biggest difference in automating replenishment of fresh products in comparison to center-store products is the increased importance of planning accuracy. Below we have listed a few critical requirements when dealing with short shelf-life products. The good news is that modern forecasting and replenishment solutions are able to not only meet these requirements but consistently exceed the performance of store personnel placing replenishment orders manually.
Day-level baseline forecasting: In retail, it is quite common that demand for different products vary within the week. For products with short shelf-lives, it’s crucial to get the day-level forecasts right in order to avoid spoilage while still having great availability. Smart forecasting systems are able to automatically consider weekday-related variation in forecasting on a store and product level as well as to adapt to changes in local demand patterns. In addition, day-level forecasting enables accurately accounting for special weeks, such as the time around Christmas, and local events, making ramp-up and ramp-down of short shelf-life inventory possible.
Accurate forecasting of promotions: For fresh products, the impact of promotions must be forecasted on a store-level to account for local demand patterns and to attain sufficient forecast accuracy. Causal modeling makes it possible to automatically consider all factors that have an impact on promotional uplift – type of promotion, marketing activities, price change, in-store display etc. – per store and product. Furthermore, it is quite typical that a promotional uplift for one product can introduce a reduction in sales for another. Especially for fresh products, this cannibalization effect needs to be considered to minimize excess stock and spoilage.
Balancing spoilage and availability: With fresh items, the short shelf-lives dictate that replenishment needs to follow sales as closely as possible. For the shortest shelf-life items that need to be sold that same day, a 100% on-shelf availability means that there will be waste or markdowns, unless the forecast is consistently nothing less than perfect on a day, store and product level. To achieve the right balance between on-shelf availability and waste for each day, the best replenishment solutions employ dynamic safety stocks that are continually optimized in accordance with how demand fluctuates within the week and on the longer term.
Taking the weather into account: Weather can have a significant impact on demand – both for fresh and center-store products. However, as short shelf-lives limit the opportunities to use buffer stock to reduce the impact of demand fluctuations, it is often even more critical to take changes in weather into account when planning the supply of fresh products. Machine learning makes it possible to automatically detect how weather conditions impact on local demand for different products, and to use this information to attain more accurate forecasts.
If you want to read more about these requirements and the benefits from automating your fresh food replenishment, have a look at our e-book: ‘Fresh Food Forecasting and Replenishment’.