The Supply Chain Challenge in CPG
For years, supply chain experts have pursued the dream of seamless, end-to-end information flow—from the consumer to the manufacturer. The goal? Ensuring products don’t sit on shelves too long or, worse, become dead stock.
This is especially critical in the consumer packaged goods (CPG) industry, where efficient product flow and inventory management drive already thin margins. In established retail channels like supermarkets and big-box stores, businesses have historically relied on past sales data to predict demand and supply with reasonable accuracy. Longer shelf-life packaging also helps prevent product expiration before purchase.
However, poor demand forecasting can lead to underperforming revenue per square foot—a major issue in a business with only 2-3% margins. Smaller retailers, such as convenience and specialty stores, face even greater risks when making inventory decisions, as missteps can quickly impact their bottom line.
The Twofold Problem: Maximizing Revenue and Managing Inventory
The core issue for retailers is twofold:
-
Maximizing revenue per square foot – If a product slows down, identifying the cause is crucial. Could a discount revive demand? Is a shortage of complementary goods (e.g., an egg shortage causing a decline in cupcake mix sales) affecting sales? Has consumer preference shifted toward a competitor’s product?
-
Managing inventory effectively – Large supermarket chains and brands like P&G and Unilever have spent hundreds of millions on big data solutions designed to track product movement, identify patterns, and optimize shelf stocking. These enterprises are now using AI to enhance their efficiencies further.
What About Small and Medium Retailers?
Smaller retailers lack the IT budgets of large corporations but still need access to data-driven inventory decisions. Instead, they often rely on distributors and their data, which may be biased, incomplete, or skewed toward supply-side influences rather than consumer-driven demand trends.
This leads to inefficient inventory management, with distributors influencing what gets stocked—not necessarily what will sell best.
CRSTBL’s AI-Driven Solution
At CRSTBL, we approach demand forecasting by focusing on where consumers and retailers interact: at the point of sale. Our goal is to connect insights from retailers to manufacturers, optimizing the entire supply chain.
Take alternative CPG products (e.g., THC, CBD, e-cigarettes, mushrooms). The supply and demand flow in this category is slow and unpredictable, with products often launched with minimal consumer research. Despite this, the market has grown to $50B+ annually, but 15-25% of inventory at distributor and retail levels remains dead or slow-moving.
Why? Because the supply chain follows a complex flow:
-
Manufacturers (“vendors”) sell to master distributors
-
Master distributors sell to local distributors
-
Local distributors sell to 150,000 convenience stores and 50,000 vape/smoke shops
Without data-driven insights, products are stocked based on outdated assumptions rather than real consumer demand.
AI: A Game Changer for Small and Medium Retailers
AI can solve inventory inefficiencies at a fraction of the cost of traditional big data solutions. However, its success depends on collaborative industry participation across the entire supply chain.
At CRSTBL, we’re leveraging CRSTA, our AI platform, to connect:
-
800-1,000 manufacturers
-
300-400 distributors
-
200,000+ retail outlets
This approach eliminates reliance on expensive, centralized big data solutions, replacing them with a distributed data model that is faster and more cost-effective. Instead of an army of data scientists, we use neural networks and decision-emulation models to create a demand forecasting system tailored to alternative CPG.
The Impact: Smarter Retail, Less Waste
Unlike traditional CPG categories, alternative products are heavily regulated, with local, state, and federal laws influencing demand. Our AI models can identify regulatory-driven trends that impact inventory performance at a hyper-local level.
We’re currently onboarding $1B+ in combined revenue across supply chain participants to test how well AI can predict demand without costly big data infrastructure.
The Bottom Line: AI Can Empower Smaller Retailers
Data-sharing is already standard among large retailers like Walmart, Amazon, Target, and Kroger, who sell POS and register data to optimize inventory. Our mission is to bring this same data-driven advantage to smaller retailers—but at a fraction of the cost, with AI-powered insights that reduce dead and slow stock.
Want to see how CRSTA can optimize your inventory and revenue? Let’s talk.