Every supply chain and retail technology conference has the same crowded aisle these days: AI inventory forecasting and automated replenishment. Vendors promise demand forecasting accurate down to the SKU level, auto-replenishment that replaces human guesswork, and inventory turnover improvements within a single quarter.
It sounds like a solved problem. But the reality is that 67.4% of supply chain managers still rely on Excel as their primary planning tool, a number that has barely shifted in a decade. Among late adopters, it climbs to 75%.
Between the AI hype wave and the spreadsheet's stubborn grip, there is a gap most vendors would rather not discuss.
Excel Is Not a Sign of Technological Backwardness, It Is the Planner's Safety Net
When a $5M wholesale distributor spends $30,000 on an ERP inventory module, what is the first thing the planner does? Export to Excel and rebuild their model from scratch.
Gartner's 2024 Supply Chain Planning Summit polling produced an uncomfortable number: only about 32% of planners actually move onto the planning tool their company purchased. The other 68% went back to the place where they have total control, a spreadsheet with instant feedback, unlimited layout freedom, and formula ownership.
Not because they do not understand technology. Planners understand Excel's limitations better than anyone: no real-time sync, no access control, no audit trail, one deleted formula that can destroy a month's replenishment plan.
But they choose it anyway, for three reasons that no planning software has genuinely replicated:
Sub-second feedback. When a planner adjusts a safety stock parameter, they want to see the impact on total inventory value and stockout rates immediately, not wait 15 minutes for the system to run its calculation cycle.
Total layout freedom. They can put supplier lead times, seasonal coefficients, promotion schedules, and customer forecasts all on one sheet, color-code priorities, use conditional formatting to flag anomalies, and add notes in the margins. No enterprise planning portal offers this degree of freedom.
Formula ownership. Every formula in their spreadsheet was written by them. They know where each number comes from and where it goes. When management questions a replenishment recommendation, they can walk through the logic line by line.
These are not UX polish issues. They are about the planner's control over their own work.
AI Assumes You Have Clean Data. Most SMB Wholesalers Do Not.
The core assumption behind AI inventory tools is simple: you have complete sales history, accurate stock records, and standardized supplier master data.
In practice, the typical SMB wholesaler's data is fragmented across systems:
The POS system has one set of sales numbers. The e-commerce backend has another. The accounting software has a third. Nobody syncs them in real time. A part-time warehouse clerk reconciles everything manually at the end of each day. The reconciliation spreadsheet, not any of the three systems, is what actually drives replenishment decisions.
When AI generates replenishment recommendations based on that reconciliation spreadsheet, the planner's first instinct is not trust. It is verification, line by line, checking each SKU's sales trend, in-transit stock, and supplier lead time. And after that verification, they often find the AI's numbers are close to what they would have calculated by hand.
So where is AI's value?
McKinsey and Netstock benchmark data tells the story: the average mid-market distributor runs roughly 38% excess inventory, meaning nearly two-fifths of their capital is sitting in stock that will not sell. Leading distributors hold it to 26%. The difference comes down to real-time visibility: when inventory across all locations is unified, live, and accurate, excess stock becomes visible, and decision-makers can act, transfer, liquidate, or return to supplier.
AI does not replace the planner's intuition. It helps the planner spot anomalies faster, predict trends more accurately, and allocate limited working capital more intelligently, but only after the data foundation is solid.
Deploying AI before unifying your data is like installing a smart water meter on a leaking pipe, the dashboard looks impressive, but the water is still going somewhere you cannot see.
Fix the Pipe Before You Install the Smart Meter
For SMB wholesalers looking to escape Excel's gravitational pull, the answer is not an AI inventory revolution. It is a three-step data foundation:
Step one: get all locations into one system. Not one system per location with a monthly consolidation at headquarters, one database, one source of truth. When one warehouse receives stock, every other location's available quantity updates in real time. When a customer places an order, every salesperson sees the same number.
Step two: standardize at receiving. Every shipment must be scanned, matched against the purchase order, and only then become active inventory. Most inventory data errors do not originate in replenishment, they originate at the receiving dock. One unscanned box means 50 units of "phantom stock" in the system.
Step three: involve the planner in software selection. Do not just ask IT and the vendor. Ask the person who builds the Excel model every week: will this system give you sub-second feedback? Can you customize the view? Can you see the logic behind every recommendation? If the answer is no, adoption will likely not exceed 32%.
Only after these three steps does AI become genuinely useful, because at that point, AI is looking at complete, real-time, accurate data instead of a stitched-together monthly summary.
The Confidence Paradox: AI Enthusiasm Meets Spreadsheet Reality
Here is where the data gets interesting. RELEX Solutions surveyed 514 retail, manufacturing, wholesale, and supply chain leaders in January 2026 and found that 67% reported increased confidence in using AI for supply chain decisions year-over-year. Nearly half, 47%, said they are already using or planning to use AI-driven inventory optimization.
Yet in the same time period, the Excel usage rate for supply chain planning has not moved. It is not that operators are anti-AI. They are pro-control. And AI tools that do not give them the same level of control they have in Excel will never reach full adoption.
This is not unique to inventory management. Across supply chain technology, the pattern repeats: companies invest in sophisticated planning platforms, and the actual planning still happens in spreadsheets. The software becomes a reporting layer. The spreadsheet remains the decision engine.
The vendors who understand this do not fight it, they build planning tools that look and feel like spreadsheets, with AI as the suggestion engine underneath. The vendors who fight it insist on replacing the spreadsheet entirely, and then wonder why adoption stalls at 32%.
When to Stay With Excel, and When to Leave
Excel is not always the enemy. For single-location merchants under $1M in revenue with fewer than 200 SKUs, Excel with good naming conventions and weekly cycle counts works fine. Excel becomes a bottleneck when you cross a certain scale threshold.
That threshold is usually: more than two physical locations, over 500 active SKUs, or monthly order volume exceeding 200. At this point, Excel's "sub-second feedback" advantage is neutralized by its data fragmentation disadvantage, the planner can see the numbers in their spreadsheet instantly, but those numbers no longer reflect what is actually on the warehouse floor.
At that point, switching systems is not about chasing AI trends. It is about giving the planner back visibility into what matters: a single, real-time, credible picture of inventory.
