How Predictive Analytics Mitigates the Inherent Risks of Just-in-Time Inventory
Before the pandemic, lean and just-in-time models helped procurement executives reduce costs and improve efficiency. Now, inflation, shortages of critical materials, and labor shortages force procurement teams into reactive overspending. This article proposes that through predictive pricing, AI-driven supplier discovery, and institutional knowledge reuse, procurement can move away from firefighting purchasing and become a key lever for CFOs to enhance performance.
Editor's Note:Edmund Zagorin is Chief Strategy Officer at Arkestro, a San Francisco-based company that provides an AI and machine learning-powered procurement platform. The views expressed in this article are solely those of the author.
In the years before the pandemic, rising productivity and falling inventory costs—often achieved through lean and just-in-time models—allowed many procurement executives to achieve business success while controlling costs. Today, however, inflationary pressures, frequent shortages of critical materials, and labor shortfalls are causing many procurement teams to overspend, remain reactive, and generate significant value leakage. As a result, a growing number of CFOs and finance management teams view procurement purely as a cost center rather than an opportunity for profit creation.
In this new environment, procurement teams—the function within finance responsible for finding and managing suppliers and executing purchases—need to fundamentally change how they operate, and one of the biggest opportunities lies in the application of predictive technology. For years, predictive technology has been used in other areas of the enterprise to improve efficiency, productivity, and revenue-generation potential—for example, using predictive maintenance to replace machine parts before they fail, thereby avoiding downtime and increasing output. But traditionally, predictive models have not been brought into procurement. Why?
Historically, CFOs and procurement departments have optimized sourcing by consolidating the supplier base and negotiating more favorable contracts with a few key suppliers. But there is far more potential to be unlocked. Here are several areas where predictive technology can help:
Predictive pricing: anchoring at fair levels to avoid paying premiums
When negotiating, procurement teams often lack a clear understanding of the actual fair market price for a given supply item. For suppliers they have worked with before, buyers may know historical transaction prices, but given current price volatility—especially in certain commodity sectors—historical prices may not be a sufficient baseline reference.
Predictive models can provide procurement teams with target pricing references, enabling them to analyze a wide range of market variables and intelligently suggest a reasonable offer that suppliers are more likely to accept. This has a twofold effect: First,research showsthat the first offer (the "anchor" price) becomes the baseline, and the final transaction price will not deviate too far from that anchor—a phenomenon known as anchoring bias. Predictive pricing ensures procurement teams understand the market and do not submit excessively high initial offers, thereby reducing the likelihood of paying a premium in the end. Second, the negotiation process is accelerated, allowing resource-constrained procurement teams to use their time more efficiently and move away from the reactive "panic buying" mode that tends to cause overspending.
Supplier discovery and game theory: expanding the competitive pool to break single-source dependency
Predictive technologies such as AI can help procurement teams discover new suppliers, avoiding the "single-source" trap that can lead to price gouging. By expanding the supplier pool and making suppliers aware of the presence of other competitors and their relative position in the bidding process, procurement teams can foster healthy competition, thereby increasing the likelihood of securing better prices. As the industry saying goes: when suppliers compete against each other, procurement teams win.
Furthermore, just as a Google search returns a list of results, AI can generate a list of suppliers and rank them based on a wide range of variables—not only price, but also attributes such as availability, owner diversity, and quality—making it easier for procurement teams to make informed decisions based on current business needs.
Extending and reusing institutional knowledge: reducing quality risk and protecting the financial bottom line
In supplier sourcing, many factors beyond price must be considered—for example, a supplier's quality history is particularly important. Today, AI can incorporate suppliers' quality records into recommendations, drawing not only on interaction data from current employees but even on feedback from former or retired team members. This enables teams to verify whether suppliers meet required quality standards, avoiding exposing the company to the risk of costly recalls—a risk that ultimately often falls on the CFO's shoulders.
It is often said that there is a huge difference between "working smart" and "working hard." Procurement teams facing challenges today are being given the opportunity to work smart, and this could indeed be the dividing line between companies meeting or missing their performance targets. As we all know, the pandemic has fully exposed the fragility of lean and just-in-time models, and these approaches can no longer be relied upon as core cost-saving measures. Like their counterparts in other business functions, procurement teams must equip themselves with modern tools to regain their rightful place as the CFO's "best partner" and become a cornerstone of organizational performance and success.