With all the talk of emerging technologies, many organizations are still making consequential inventory and replenishment decisions with tools and processes that haven't fundamentally changed in years.
They’re still relying on spreadsheets, lagging indicators, and gut instinct layered over historical data. In an environment where a small blip in consumer demand can trigger a bullwhip effect across a supply chain, that's a serious vulnerability.
The good news is that the path forward is clearer than it's ever been. Modernizing forecasting is a disciplined, achievable shift that delivers measurable improvements in inventory performance, supply chain resilience and operational efficiency.
Most organizations have more data than they know what to do with. The problem is that the data isn't being used in a connected, forward-looking way.
Consider what happens when a major retailer runs a promotion. Consumers pantry-load, causing demand spikes that can cannibalize other products or lift sales of complementary item. The orders flow upstream to manufacturers and distributors, but once the promotion ends, demand collapses, and the upstream partners are left holding excess inventory they didn't anticipate. The demand signal was visible — the challenge was translating it quickly and precisely into the right operational response based on sell-through signals.
This dynamic plays out across retail, grocery, fashion, food and beverage, and consumer packaged goods. Millions of dollars of inventory sit in stores across thousands of locations. Too much, and you're tying up working capital and sacrificing shelf space that could carry higher-performing products. Too little, and customers face empty shelves and walk out the door, possibly for good.
Traditional forecasting struggles with these dynamics because it's primarily backward-looking. It asks, "What did we sell last week?" The more important question is, "What will we sell next month, and why?" This is why boundaryless planning is becoming increasingly relevant. It’s where supply chain, merchandizing, logistics and store ops are able to continuously see, analyze, decide and act.
What Modern Forecasting Looks Like
Effective forecasting today is about connecting the forecast to the full planning and execution environment, as well as leveraging causal factors —weather, product placement, price, promotional tactics, macro and micro economic factors (such as tax or SNAP programs), demographics, housing starts and discretionary income.
The starting point is getting the assortment right. Category-management analytics, driven by point-of-sale signals and local demographic data, give planners the insight to rationalize assortments before the forecast is even complete. Which products should be on the shelf? In which stores? With how many facings? Which product is on the landing page of an e-commerce site? These decisions upstream of the forecast have an enormous impact on its accuracy and utility.
That same logic now has to extend beyond the four walls of the store. E-commerce and direct-to-consumer demand are no longer separate channels sitting beside the retail business; they’re part of the same demand signal. Buy online, pick up in store and buy online, ship from store models are turning stores into fulfillment nodes, which means that every forecast has to account for where demand is created, where inventory is positioned, and how quickly product can move across channels. This is where omnichannel sales strategy becomes inseparable from supply chain planning: If retailers can’t see and forecast demand across stores, digital channels, pickup orders, and ship-from-store activity in one connected view, they risk solving one channel’s problem by creating another channel’s inventory gap.
From there, forecasting needs to incorporate external signals. Weather has a direct effect on what sells and when. A cold snap changes beverage and food demand in ways that historical averages can't capture. An approaching hurricane creates a predictable surge in generators, plywood and cleaning supplies. Organizations that integrate these signals into their planning are able to position inventory ahead of demand rather than react after the fact.
Promotions, competitor activity and local events all behave similarly. They're knowable in advance, and they have a measurable effect on demand patterns. The question is whether your forecasting infrastructure can incorporate them systematically, or whether planners are still making manual adjustments based on memory and experience.
The Gap Between Plan and Execution
One of the most persistent and costly disconnects in supply chain operations is the gap between what the plan says and what actually happens on the floor, in the store or within other channels. Planning systems and execution environments often operate on different data sets, updated on different cadences, with different assumptions baked in.
When these systems don't talk to each other, decisions made in planning don't reflect operational reality, and the supply chain or merchandising teams don’t see plan changes in time to act on them. The result is a slow, reactive organization that's always catching up.
Closing this gap requires data harmonization, where the same signals inform both planning and execution. It also requires moving from insight to action. Companies today want systems that surface recommendations proactively, telling teams what to do and when, rather than requiring them to dig through data after a problem has already cost them.
Artificial intelligence and machine learning have earned their place in this work as practical tools for scaling proactive decision-making across large, complex operations. The shift is from business intelligence that answers questions to intelligence that surfaces decisions.
Forecasting modernization is also inseparable from supply chain resilience. The organizations that weathered recent disruptions were the ones with the operational discipline to act on what they knew, and the structural flexibility to absorb what they didn't.
Scenario planning is a core component of this. What happens if a key supplier can't deliver? What if a port closes due to labor disruption or geopolitics? What if a competitor runs a promotion that pulls volume? These are operational capabilities that allow leadership to evaluate trade-offs in advance rather than improvising under pressure.
Diversifying the supplier base is a direct output of this thinking. Organizations that depend on a single source for critical inputs are exposed in ways they often don't fully appreciate until something goes wrong. Spreading that risk across two, three, or four suppliers, even at a slightly higher cost, is a form of forecasting, an explicit bet that the cost of disruption outweighs the cost of redundancy.
Inventory strategy also needs to reflect this reality. Forecasts will never be perfectly accurate; lead times won’t be perfectly reliable, and trucks won't always arrive on schedule. The goal is building enough buffer to absorb variability without over-investing in inventory that ties up working capital and generates obsolescence risk. Finding that optimal stocking level — not too much and not too little — is where the real work happens.
Addressing the Counterargument
Some organizations push back on modernizing forecasting on the grounds that their demand environment is too unpredictable, too promotional, or too fragmented to forecast accurately. That argument has it backwards. The more volatile and complex the demand environment, the more valuable a disciplined, data-connected forecasting capability becomes.
Others point to the cost and complexity of implementation. These are legitimate concerns, and they deserve honest answers. Not every organization needs the same solution. The right starting point depends on the current state of planning maturity, the quality of available data, and where the biggest inventory and service-level gaps exist. But the cost of not modernizing, measured in excess inventory, missed orders and lost margin, tends to far exceed that of making the investment.
Supply chain leaders are being asked to do more with less, to carry less inventory while improving service levels, to respond faster while planning further ahead, and to manage risk in an environment that keeps generating new ones.
The organizations that will meet that challenge are the ones that treat forecasting as a strategic capability rather than a back-office function. That means connecting demand signals to assortment decisions, integrating external factors into the planning environment, closing the gap between plan and execution, and building resilience into the inventory strategy from the ground up.
The tools needed to do this are available. And the case for doing it is stronger than it's ever been. The decision is whether to treat forecasting as a constraint or an enabler.
Bob Patel is managing partner and practice leader, supply chain management and retail planning at Highspring.
