Huge ROI. Short Demo.
6 minutes
.jpeg)
When the Covid-19 pandemic struck, CPG brands were left scrambling to respond to skyrocketing demand and other dramatic shifts in consumer behavior.
While its competitors were eating massive capital losses and struggling to make sense of the new market reality, Our Client (whose anonymity we’ve protected upon request) was able to align internally, respond quickly, and confidently raise service levels with retailers.
This case study lays out exactly how Unioncrate’s AI made this possible—and how we could do the same for you.
For the 12-month period prior to Covid-19, Unioncrate’s AI achieved forecast accuracy up to 91.3% (Jul ‘19) and a yearly accuracy mean of 84.3% (derived from WMAPE) at the SKU level across all warehouses. Unioncrate’s forecasts bested CPG industry benchmarks with a yearly deviation of +\- 3.69%.
Unioncrate AI is designed to accommodate all changes in consumer purchasing behavior into its models, resulting in high levels of forecast accuracy. As a result our Client, a category leader in perishable products, was prepared for these constant shifts in consumer purchasing patterns unique to the CPG industry.
A perfect storm of unpredictable external factors took hold in early-to-mid-March as a result of Covid-19. Each of them had some level of impact on FMCG/CPG brands in the U.S. They are:
Nobody could have predicted the Covid-19 pandemic nor the devastating impact it would have on countless businesses, communities, and individuals—or the speed with which it threw the entire retail world into disarray.
Covid-19 had an enormous impact on consumption purchasing patterns as shoppers, many of whom were under stay-at-home orders or bound by other similar restrictions, began to stock up on essential goods.
According to Nielsen data, for the two weeks ended March 21, total U.S. CPG sales (in-store and online) increased $8.5 billion from the two weeks prior. For example, for the week ending March 21, online sales grew +91% and in-store sales rose + 45% vs. the same period year ago.
Similarly, retail sales for Our Client’s products grew +73% the week of Mar 16 (W12) and +170% the week of Mar 23 (W13) vs. the week of March 9 (W11).
Like many companies in the CPG industry that experienced unexpected spikes in sales, Our Client faced immediate supply shortages and other supply chain issues that impeded its ability to meet demand. After a demand peak the week of March 23 (W13), Our Client started to face inventory shortages and was forced to cut orders. The situation worsened over the next two weeks until the week of April 13 (W17 in the graph below).
From March 16 (W12) through March 30 (W14), Our Client relied on Unioncrate’s AI to find the most efficient and profitable way to bridge the quantifiable gap between supply and demand. As a result, Our Client was able to quickly act on the following insights:
With this knowledge in hand, Our Client hatched and executed a plan—for Apr 6 (W15) up until Apr 20 (beg. of W17)—to optimize manufacturing output by category and reallocate inventory across all warehouses based on demand by their respective ship-to locations. Our Client’s actions minimized the impact of the supply shortage, optimized transfer costs, improved service levels and maximized sales in a remarkably fast turnaround.
Unlike another AI, Unioncrate’s neural network models are designed with a deep understanding of the complexities and ever-evolving dynamics of the CPG industry. Unioncrate’s AI efficiently analyzes quintillions of internal (historical shipments, promotions, NPDs, e.g.) and external datasets (point-of-sale data, seasonality, competition e.g.) to find dynamic, multi-layered relationships between variables that uncover quantifiable demand signals, resulting in highly accurate forecasts across multiple time horizons.
Covid-19 is an outlier event that brought about unprecedented changes to consumer purchasing behavior. Ultimately, though the impact of Covid-19 on demand is a vital data point, it is only one of countless variables the deep learning models take into account to deliver weekly SKU forecasts. Put simply, Unioncrate’s AI corrected itself.
Unioncrate’s AI was able to achieve 78.5% monthly forecast accuracy for the month of May. The ability of our AI to quickly adapt to unforeseen variables helped Our Client efficiently plan production, adjust stock levels, minimize working capital expenditures, reduce overstock, and realize sales that it would have otherwise lost.
As a result, Unioncrate helped Our Client maintain its leadership position in the marketplace by allowing them to anticipate risks, identify sales opportunities, and realize a number of concrete efficiencies, including:
Because leadership can trust Unioncrate’s demand forecasts, they can confidently allocate resources in areas that would maximize profits and operational efficiencies, such as investment in products with higher profit margins.
Our Client now uses Unioncrate’s weekly forecasts as the starting point for all internal demand planning conversations, helping their team align from the get-go. This makes cross-functional team dynamics stronger and more efficient and pushes the decision-making process in the right direction, faster.
Unioncrate provided higher accuracy forecasts by ship-to location, empowering Our Client to transfer inventory efficiently from warehouse locations with lower demand.
Unioncrate’s automated demand forecasts enable sales teams to focus on growing the business rather than spending their time tabulating forecasts manually.
Because Unioncrate’s AI consistently delivers highly accurate forecasts, Our Client can confidently reduce safety stock levels.
Unioncrate is an AI-powered Integrated Business Planning (IBP) platform that delivers demand forecasts with unmatched accuracy, collaborative visibility, and actionable intelligence—enabling CPG brands to plan and execute agile supply chain strategies at the click of a button.
.jpeg)
