The Truth Is: Wholesale Distribution’s Margin Problem Is Really a Planning Problem

authors

Santiago Garcia-Poveda

Retail Digital Transformation Leader

Published on: August 26, 2026

6 read min

Wholesale distribution has always been a game of inches.

A fraction better on purchasing. A little less inventory sitting in the wrong warehouse. One fewer expedited shipment. A few percentage points of improvement in fill rate. None of these moves look transformational by themselves.

But when net margins live in the low single digits, small operational misses do not stay small for long. That squeeze is not easing. McKinsey reported in May 2026 that warehouse wages are up more than 30% since 2020 and that 82% of supply chain leaders are now absorbing direct tariff impacts affecting 20 to 40% of their costs. This is pressure landing squarely on an industry with almost no room to give.

That is what makes the current environment particularly difficult for distributors. Costs are rising, customers expect faster and more reliable service, assortments are expanding, supply remains unpredictable, and planners are being asked to manage millions of combinations across SKUs, locations, customers, and channels.

The real issue is that many companies are still trying to manage that complexity with planning infrastructure built for a much simpler business.

How Wholesale Distributors Can Win with Intelligent, Connected Planning

Margin pressure is intensifying. Consumer expectations are shifting. Supply chains remain fragile. And the data available to distributors has never been richer, but most organizations still cannot act on it quickly enough to make a difference.

This white paper explores the structural challenges facing wholesale distributors, and makes the case for a fundamentally different approach to planning and decision-making: one that is integrated, data-driven, and AI-powered from the ground up.

Complexity is multiplying faster than planning can move

Let’s look at demand planning, for example.

A distributor today is not forecasting one relatively stable stream of customer orders. Demand can be influenced by major-account forecasts, e-commerce activity, local sales intelligence, promotions, new locations, project pipelines, weather, economic conditions, and changing end-consumer behavior.

Now, multiply that across hundreds of thousands of SKUs, dozens of distribution centers, and hundreds of customers.

The number of planning intersections and overlaps quickly stretches into the millions.

And still, many organizations still rely on some combination of ERP systems, standalone applications, spreadsheets, and manual planner intervention to make sense of it all. The consequences are familiar.

Inventory accumulates where demand does not materialize. Another location runs short. A supplier delay creates an urgent purchase. Freight is expedited. A sales team pushes a promotion without a complete view of available supply. Finance discovers the margin impact after the decision has already been made.

Each function may be making a perfectly reasonable decision based on the information it has.

But, problematically, it means that the business is not making one connected decision.

More inventory does not equal more resilience

When supply becomes unpredictable, the instinctive response is often to add buffer stock.

That can buy time. It can also create an expensive illusion of security.

Inventory tied up at the wrong node does little to protect service somewhere else. Excess stock increases working capital and storage requirements, while perishable or time-sensitive products introduce another risk: the inventory intended to protect service can eventually become waste.

The better question is not simply, "How much inventory should we carry?"

It is: "Where should each unit of inventory sit to create the most service protection for the least working capital?"

That requires looking across the network rather than setting safety stock independently at each warehouse.

Multi-echelon inventory optimization does exactly that, taking demand variability, lead times, service targets, storage constraints, and the relationships between stocking locations into account simultaneously.

According to o9 customer results, this approach can contribute to total network inventory reductions of 10–30% without deteriorating service levels.

For a low-margin distributor, that is not simply a supply chain KPI. It is a balance-sheet decision.

The real opportunity is connecting the decisions

Better forecasting matters. Better inventory optimization matters. Better supplier collaboration matters.

But there is a limit to what each can achieve independently.

“The challenge is not that distributors lack awareness of these problems. Most planning leaders in the industry can describe them precisely. The challenge is the transition: moving from an environment built around fragmented tools and manual processes to one built around an integrated, AI-powered planning platform, all without disrupting the operational continuity that their customers depend on.”

Santiago Poveda

VP of Retail, Distribution & Apparel, o9 Solutions

Suppose demand sensing identifies an unexpected increase in a product category. That signal needs to change inventory targets. Those targets need to translate into replenishment requirements. Procurement needs to understand supplier capacity and landed cost. Logistics needs to know whether additional volume can be moved economically. Commercial teams need to understand whether the opportunity is actually profitable.

And finance needs to see what all of that does to the P&L.

The same logic applies to commercial growth. A new category, supplier, channel, or service proposition may increase revenue while also adding inventory, capacity requirements, transportation cost, or cannibalization elsewhere. Connected planning helps distributors test those trade-offs before scaling a growth decision, so commercial opportunity is evaluated against its true operational and financial impact.

That chain should not require six systems, four spreadsheets, and a series of meetings to reconcile.

In an integrated planning environment, demand can drive inventory, inventory can drive replenishment, and replenishment can flow into supplier and logistics decisions using the same underlying model. Suppliers can work against shared forecasts and constraints rather than periodic emails and disconnected purchase orders.

This changes the nature of planning, so that the goal moves from producing a forecast to orchestrating a response.

AI changes the equation (if the foundation is there)

The next stage is automation.

Machine learning can already improve forecasting, segment SKUs intelligently, surface exceptions, and automate routine planning work. Agentic AI extends that idea further, allowing systems to interpret context, evaluate options, and recommend or execute decisions within defined guardrails.

But adding an AI agent to a fragmented planning landscape does not remove the fragmentation underneath it.

If demand, inventory, supplier constraints, financial targets, and commercial policies are disconnected, an AI system has no reliable common plan against which to reason. Which is why the data and planning foundation matters.

“The distributors who build integrated, AI-powered planning capabilities now are positioning themselves to outperform in this environment. They are not simply automating what they already do, but fundamentally changing how decisions are made, how teams collaborate, and how quickly the organization can respond to a market that rarely stands still.”

Brent Hasenkamp

Vice President, Industry Solutions, o9 Solutions

Agentic AI becomes materially more useful when it sits on top of connected data and structured planning logic, rather than operating as an isolated intelligence layer.

The destination is not AI for its own sake, but instead a distribution operation where routine decisions increasingly happen automatically, planners focus their attention on genuine exceptions, and every major decision can be evaluated against service, cost, margin, and working capital at the same time.

Small improvements become very large numbers

The business case becomes clearer when the levers are viewed together.

Across o9 wholesale distribution deployments, outcomes include 10–30% reductions in network inventory, 20–50% reductions in stockouts, 10–30% reductions in expediting and flex freight costs, and 40–60% reductions in manual data preparation time.

Those percentages matter because distribution economics magnify them.

For a $1 billion distributor operating on thin margins, modest improvements across inventory, freight, service, spoilage, and productivity can translate into tens of millions of dollars in profitability and cash-flow impact.

That brings us back to the game of inches.

Wholesale distribution may never become a high-margin, low-complexity business. That is not the opportunity.

The opportunity is to become much better at deciding where every inch is won or lost, because when forecasting, inventory, suppliers, logistics, commercial decisions, and finance are planned as one connected system, small improvements stop being isolated efficiencies and start compounding into real advantage.