How a Travel Center Operator Found What the System Said Was on the Shelf
Use case Inventory management & reconciliationTravel center operator, 22 locations
How a travel center operator found what the system said was on the shelf.
The system showed stock that wasn’t there, deliveries that never fully arrived and losses nobody could place. Cortex reconciled every item, every delivery and every count, every morning.
01 · The challenge
The system said in stock. The cooler said otherwise.
Weekend travelers were walking out empty-handed while the back office showed full shelves. Reorders never fired, because on paper nothing was low.
Store managers spent mornings walking aisles with clipboards, and direct-store deliveries were signed for without anyone checking the count against the invoice.
Before Cortex
What the team saw first
- Weekend salesSlipping on key items
- System on-handHealthy everywhere
- ReceivingSigned, rarely counted
- InstinctBlame weak demand
The difference
Same question. Different results.
The question“What are we out of right now, even if the system says we aren’t?”
Claude / GPT / Public LLMs
Uploaded file
Answer
All 22 stores show healthy stock. Nothing is below its reorder point.
Just an answer.
Cortex AI
Connected intelligence
Cortex AI- POS sales by hour
- Receiving scans
- Vendor invoices
- Cycle counts
- Delivery schedules
- Footfall
- Weather
- Local events
- Shrink history
- Shift data
- 410 daily sellers not selling
- 14% of drops invoiced short
- Shrink: 3 stores, overnight
Answer
410 items are in stock only on paper. Most sit behind fixed delivery days, and 14% of direct-store deliveries were short. Every manager has a pull list before opening.
A decision.
02 · A day with Cortex
Three different leaks, all hiding behind one number.
Cortex compared POS sales, receiving scans, vendor invoices and cycle counts for every item at every store. An item with stock on hand and no sales for days is almost always an empty slot.
Fixed delivery days left fast sellers empty by Saturday. Unchecked direct-store deliveries cost about $18,000 a month. Shrink sat in four categories at three stores, almost all on the overnight shift.
- Overnight reconciliation410 daily sellers flagged: stock on paper, none sold in 3+ days
- A pull list for every managerCheck the back room first, then reorder
- A beverage delivery is scanned inTwo cases short of the invoice. Credit requested automatically
- Delivery days rebalancedFast sellers moved to Thursday and Saturday drops
- Shrink by category, store and shiftOvernight counts added at three stores
03 · The evidence
Where the missing value went.
Every number traced back to its source record, so each fix had an owner and proof.
What Cortex found
Where the missing value went
Lost sales and losses Cortex traced in the first 30 days, across 22 stores.
Fix the slot, the delivery and the shift.
- A daily “stock on hand, zero sales” list for every store, before opening.
- Every direct-store delivery checked against its invoice at the door, with credits requested automatically.
- Fast sellers moved to twice-weekly deliveries, and overnight counts added at three stores.
04 · The impact
What changed.
On-hand is a guess until something checks it.
We thought we had a demand problem. We had a shelf problem, a delivery problem and a night-shift problem.
Inventory goes wrong in small ways, every day, in every store. The only fix that scales is checking it every day.
