← Back to list

Summary card

EN 2026-06-15 23:00
LEANInventory ManagementBusiness Efficiency

Globex's request to unify EC inventory and orders. How LEAN exposed the waste hidden in dual management, and an efficiency design that smooths the flow and unifies inventory.

ROI Case File No.536: Two Warehouses Were Counting the Same Goods Separately

EN 2026-06-15 23:00

ICATCH

Two Warehouses Were Counting the Same Goods Separately


Chapter 1: We're Counting the Same Inventory Twice

"Our own EC site and Rakuten manage inventory separately. We want to resolve the dual management."

Seiichi Ebihara, CEO of Globex, spoke while showing his screen. "Rakuten accounts for about 90% of sales. Our own EC only recently went into full operation and has few SKUs. But managing the two separately is causing problems."

"What concrete damage does dual management cause?" Claude asked.

"Inventory drifts during peak season," Ebihara answered. "Something sells on one side, but the inventory count on the other isn't updated. As a result, we sell an out-of-stock item and apologize to the customer. This happens constantly. When we list goods, too, there's the double work of registering separately on the two sites."

"What about customer information management?" I confirmed.

"That's split as well," Ebihara answered. "Rakuten's repeat data is only viewable for two years. Anything before that is managed by hand. The same customer is treated as a different person when they cross sites. In short, the work of managing the same thing twice is everywhere."

"Doing the same job twice is textbook waste," I responded. "Let's smooth the flow with LEAN."

Chapter 2: LEAN Asks Where the Waste Is Hiding

"This case needs LEAN."

Claude wrote "LEAN" on the whiteboard.

"LEAN is the idea of thoroughly removing waste that creates no value and smoothing the flow of work," I explained. "Its source is the Toyota Production System. You identify and cut waste—dual inventory management, waiting, rework, double handling. What matters is not speeding up individual tasks but smoothing the whole flow. It's a tool for erasing the waste of counting the same goods twice through flow design."

"Let's measure the current cost first," Gemini said, opening ROI Polygraph and entering the data Ebihara had provided.

"The monthly dual-management cost is in," Gemini read out. "Manual dual inventory management averages 190 hours a month; at ¥3,400 an hour, that's ¥646,000 a month. Stockout and apology handling from inventory drift averages ¥700,000 a month. The double work of listing goods averages ¥450,000 a month. Manual management of customer repeat information averages ¥300,000 a month. Lost sales opportunity from stockouts averages ¥550,000 a month. The total is ¥2,646,000 a month—roughly ¥31.75 million a year."

Ebihara stared at the figures. "I thought it was only the labor of inventory management. Add stockout apology handling and lost sales opportunity, and dual management was cutting this much."

"Then let's design it with LEAN," I continued.


[Defining Value—What Is Value to the Customer]

"First, we define value," Claude said. "Value to the customer is that the ordered item arrives for certain. Apologizing for a stockout because inventory drifted is the prime example of waste that damages value. Defining what is value and what is waste is the origin of improvement."


[Identifying Waste—Where in the Flow Does Double Handling Occur]

"Next, we identify the waste," Gemini continued. "Visualize the flow of inventory data and you see double handling linked together—dual management, inventory drift, double listing, split customer information. Not points, but where in the flow value is being damaged—that's what we identify."


[Designing the Flow—Erase Waste Through Unification]

"We erase the identified waste through flow design," I continued. "We unify inventory management and order management, linking the two sites' inventory in real time. Product information is centrally managed and auto-updated. We make a structure where the very act of counting twice never arises—through flow design."


[Continuous Improvement—Smooth While You Loop]

"Finally, we keep improving," Claude continued. "Not done at deployment; we monitor regularly, find new waste, and cut it. We continue employee training, too. LEAN isn't a one-time grand reform but a movement of finding and erasing waste continuously."


[Estimating the Payback]

"Let's run the numbers with ROI Proposal Generator," Gemini proposed.

  • Initial cost: building an inventory/order unified-management system, linking the own-EC and Rakuten, centralized product-information management, data migration, and training—¥6.4 million total
  • Monthly cost: system fees and ongoing operation combined—¥220,000 a month
  • Monthly savings: dual-management hours cut = ¥520,000 a month (assuming 80% reduction); stockout and apology handling cut = ¥500,000 a month; double-listing cut = ¥360,000 a month; lost sales opportunity recovered = ¥420,000 a month—¥1,800,000 a month total
  • Net monthly savings: ¥1,800,000 − ¥220,000 = ¥1,580,000 a month
  • Payback period: ¥6.4 million ÷ ¥1,580,000 = about 4.0 months

"Payback in four months," Gemini summarized. "What makes it work is not speeding up the task but erasing the twice-done task itself. Unify inventory, and the task of dual management no longer arises. Not reducing waste but cutting the source of waste through flow design—this is how LEAN works."

Ebihara said as he checked the figures, "I'd only thought of making inventory management more efficient. Smooth the flow and eliminate counting twice. The notion was different."

"LEAN is a tool for cutting the source of waste through flow design," I responded.

Chapter 3: A Rollout Plan That Smooths the Flow

"Let me lay out the approach," I said, standing at the whiteboard.

"Month one—interview the current inventory/order management and visualize the data flow. Month two—identify waste and select the unified-management system. Months three and four—build the inventory/order unified-management system and link the own-EC and Rakuten. Month five—implement centralized product-information management and migrate data. Month six—go live and train employees. Month seven onward—monitor and improve, integrate customer information, and remove new waste."

"Rakuten is 90% of sales. Won't the integration cause trouble?" Ebihara confirmed.

"That's why we visualize the flow before designing," Claude responded. "Rather than connecting straight away, we pinpoint where in the data flow the drift occurs, then unify. The link with Rakuten, the sales mainstay, is designed most carefully. Generating new trouble in order to erase waste defeats the purpose. We look at the flow, then cut."

Ebihara said as he took notes, "Counting twice had become a matter of course. It was something we could eliminate."

Chapter 4: The Day Inventory Became a Single Number

Nine months later, a report arrived from Ebihara.

Three months after the unified-management system went live, manual dual inventory management was down 80% versus before. "The two sites' inventory was consolidated into one number. Sell on one side and it reflects instantly on the other. The task of counting twice disappeared," Ebihara wrote.

Stockouts from inventory drift dropped sharply too. Peak-season stockout apologies plummeted. "Because inventory links in real time, we no longer sell what isn't there. The time spent writing apology emails vanished," the report said.

The biggest change showed up on the peak-season floor. The period that had been most chaotic benefited most. "Peak season was exactly when inventory drifted and we chased apologies. Now inventory matches even in peak season. The hardest time became the easiest," Ebihara wrote.

The double work of listing goods was resolved too. Centralized management and auto-update meant registration on the two sites was done once. "The work of registering the same item twice is gone. Launching new products got faster," the report said.

As a side effect, customer information integration began advancing. A base for grasping repeat behavior across sites was built. "We're starting to see the same customer as one person across sites. The next move is coming into view," Ebihara wrote.

At the end of Ebihara's report, he had written this: "I thought of it as making inventory management more efficient. But the essence was erasing the waste of counting twice from the flow. The moment we visualized the data flow with LEAN, where value was being damaged became visible. Waste can be cut not by reducing it but by a design that prevents it from arising."

The day a company where two warehouses counted the same goods separately became one that manages inventory as a single number, inventory management had turned from double-handling labor into a smoothly flowing mechanism, the report read.

"Requests for business efficiency usually come in as 'we want to make the task faster.' But before speeding up, you must ask: is that task even necessary? What LEAN asks is where in the flow the waste is hiding. Count the same inventory twice, register the same item twice, apologize at every stockout—identify the value-less tasks and cut them at the source through flow design. The day a company where two warehouses counted the same goods separately could make inventory a single number, what changed was not the inventory system but the very perspective that cuts waste rather than reducing it."


lean

Tools Used

  • ROI Polygraph — Visualizing dual-management hours, stockout apology handling, and lost sales opportunity
  • ROI Proposal Generator — Payback simulation for inventory unification through flow design

Describe Your Case