In the ebook Survive or Thrive: Retail Imperatives for 2020, we covered how the only way to stay hyper-focused on inventory is with AI-enabled technologies that can sense and respond to demand-changing factors in real time. But, how does the growth in shopper demand for fresh change the forecasting requirement and how do you get it right?
Food wastage – volume and cost, environment
Roughly one-third of the food produced in the world for human consumption every year, approximately 1.3 billion tonnes, gets lost or wasted. These losses amount to approximately US$ 310 billion in developing countries and US$ 680 billion in industrialized countries, and it’s clear to see that this is not a sustainable model for anyone. Statistics like this show that getting smart with fresh isn’t just about meeting consumer demand; fresh food forecasting is vital and it’s on every grocery retailer CEO’s radar. However, beyond just dollars and cents, reducing waste through accurate forecasting has additional societal value, as today’s consumers are increasingly more eco-conscious.
AI powers data for fresh forecasts
To support the growth in fresh and short-life, retailers need to get a lot right in an area where so much can go wrong. The shopper wants the best product, the best shopping experience and the best price and the retailer is trying to get the balance right between shopper needs and their own requirements – maintain margin, improve availability, reduce waste, customer satisfaction – the list seems endless. Systems address some of the challenges, but legacy systems were never designed to process the volume and variety of data needed to produce an accurate demand forecast for fresh. But AI-enabled systems are designed for exactly that purpose. (To learn more about the role of AI in improving demand forecasting effectiveness read our new viewpoint paper).
One of the key requirements in fresh is for retailers to understand the movement of items, not just at a broad level but down to store-item-level detail. With so many factors affecting demand, like weather, price, promotions, social media sentiment, ultra-fresh items and store level sales – a lot of data needs to be input to create the demand forecast. But because demand can change so rapidly, the system needs to be able to account for and adjust to daily, intra-daily, and at times, hourly changes in demand at store level. A five-degree variance in forecast temperature can change the assortment in a basket from soup and bread to salads. Legacy systems can’t do anything with that information, AI systems can.
Dealing with product markdown in fresh food forecasting
Another challenge faced by retailers as they try to grow in fresh, is product markdown. The nature of the category means that products are often marked down to avoid wastage. While this solves that problem, it also creates a risk of training shoppers to wait for the discount and training forecasting systems to buy. Legacy systems don’t take price elasticity into account and they don’t record the actual price being paid for a product when it has one day of shelf life left. Instead, a reduced-price item must be recorded as such and not in the same way as a full-price item. Retailers do not want to replenish goods that are being marked down at large discounts, just to avoid wastage. AI systems account for this, as they are designed and built to drive success in fresh and short-life demand forecasting. They can process the high level of data needed to improve accuracy in demand forecasting to reduce wastage and out-of-stocks and keep the new shoppers coming back for more.
Retail imperatives for success
Working on the “success equation” today entails more than addressing waste and its financial and environmental impact. Physical stores must evolve to enable the best experiences possible for customers. Promotions must become ultra-personalized. Inventory levels have to walk the delicate line between out-of-stocks and overstocks.
To explore these and other factors, we recently published our new ebook, Survive or Thrive: Retail Imperatives for 2020. I encourage you to take a read.
If you’re just considering the role of AI, read our paper on Demand Forecasting.
If you’re ready to starting to looking at solutions, why not check out our Buyers’ Guide to Demand Forecasting.