While most companies can track their shipments just fine, many still struggle to connect the dots between purchase orders, landed costs, and actual inventory profitability in real time. This disconnect creates what’s known as “shadow inventory” — stock that appears strategic on paper but turns into a cash trap when conditions change rapidly. 

As we near the end of summer, most supply chain teams are deep in execution mode for holiday inventory. August back-to-school is here, and September holiday stock needs to be locked in. This timing makes shadow inventory particularly dangerous, with little room for error, when working capital costs are climbing, and every aspect of planning and execution must go smoothly.

What Shadow Inventory Actually Looks Like

Shadow inventory isn't your typical overstock sitting in a finished-goods DC. It's product where the financial reality doesn't match the planning assumptions — inventory that seemed profitable when ordered, but becomes a liability when real costs surface. 

Consider these possible scenarios: 

The tariff shuffle. A major retailer places orders for Mexican-made goods in May, budgeting for 15% tariffs. But classification changes in June bumped the rate to 25%. The shipment tracking shows everything moving smoothly, but the teams aren’t aware of the true landed cost until the goods clear customs, potentially weeks after they hit the balance sheet. 

Port congestion math. An electronics brand's Vietnam shipment sits at Long Beach for two weeks, racking up $180,000 in detention and demurrage charges. Their system shows the inventory as "in transit" at the original budgeted cost, but the actual landed cost jumps 8%. The team might not discover this until their freight forwarder bills them next month. 

Accessorial charge blindness. A retailer's new Mexican supplier consistently ships on schedule, but their freight forwarder is adding $2,500 per shipment in accessorial charges for oversized pallets that don't fit standard containers. The procurement system shows goods arriving at budgeted freight costs, but the actual logistics spend is 22% higher. These charges only surface when the freight forwarder sends monthly invoices, weeks after the inventory is already on the books and committed to customers.

The Integration Problem

Most companies have decent tracking for individual pieces. They know where shipments are, what they paid suppliers, and what's in their warehouses. But they can't connect these data points in real-time to understand the true financial picture. 

Your procurement team knows what they contracted to pay. Your logistics team knows what shipping actually costs. Your finance team knows what hit the P&L. But when circumstances change rapidly — tariffs shift, currencies move, ports clog up — there's no integrated system giving you the complete picture. 

Nearshoring doesn't create this issue, but it makes it worse. When companies ran everything through established Asian suppliers, they had predictable cost structures and well-understood processes. Moving to Mexico or Vietnam means new suppliers, new logistics lanes, new regulatory requirements — and new ways for costs to spiral without immediate detection. 

A retailer shifting from Shenzhen to Tijuana isn't just changing geography; they're dealing with different customs procedures, new freight forwarders, unfamiliar detention rules and currency exposures they haven't managed before. Each of these introduces cost variables that might not surface in their systems until weeks after the inventory is already committed. 

The key is building connections between data systems that currently operate in isolation. 

Freight invoice reconciliation. Compare actual shipping charges against budgeted costs within 48 hours of shipment departure. When your Mexico-to-Dallas lane consistently runs 15% over budget, you're accumulating inventory at the wrong cost basis without realizing it. 

Accessorial charge patterns. Track when freight forwarders add unexpected fees for oversized shipments, detention or special handling. These charges often don't surface until monthly invoices arrive, weeks after inventory is already committed. 

Customs clearance timing. Monitor how long shipments spend in customs versus historical averages on each lane. Extended clearance times usually mean classification issues that will show up as unexpected duty charges. 

Supplier communication analysis. Parse emails and documents from suppliers for requests to change payment terms, delivery schedules or specifications. These requests often signal cost pressures that will eventually impact your inventory economics. 

Rather than build new systems, identify opportunities to capture and analyze data with AI. Most companies already have the information they need — purchase orders and shipping documents in email attachments, cost updates in supplier spreadsheets — they just can't synthesize it fast enough to make decisions. 

Start by identifying where your cost data lives in different systems. Cross-reference your purchase order details with real-time freight conditions. Look for patterns where actual costs consistently exceed budgets on specific lanes or suppliers. 

Consider also using AI-powered natural language analytics tools that can analyze data across multiple formats. The goal is to create a single view of inventory profitability that updates as conditions change, not just when invoices arrive. 

This isn't about abandoning nearshoring plans. Most companies are already too far down that path to turn back now, and it might still prove to be the right long-term strategy. The issue is that procurement teams are making purchase decisions based on cost assumptions that logistics and finance teams can't validate until weeks later. 

Right now, with holiday inventory commitments and execution locked in, there's no time to build elaborate new systems. But connecting the data you already have can prevent nasty surprises when Q4 financials roll in. The freight forwarder emails, supplier invoices and customs documents contain the real cost information. The question is whether you can read and react to them fast enough to make decisions that matter. 

Stephen Dyke is principal solutions consultant manager at FourKites.