What the system shows
The ERP, the buyers' screens, and the online store
- Inventory on record
- $21.4M
- Active SKUs
- 38,000
- Minimums raised by buyers
- 2,900 SKUs
Case study · Sample site assessment
The warehouse that had everything in stock except what customers ordered
A founder-owned industrial distributor was weighing a $1.1M warehouse system. One order in nine shipped short or late, often for stock the system said was on the shelf. Our two-day site assessment traced the gap to moves made without a scan and showed how to fix the process first.
Sample deliverable. This case study and report are a composite. Company names, people, locations, and figures have been changed to protect client confidentiality.
Fast-pickOverflow
Show data
| Category | Locations |
|---|---|
| Matched the system | 98 |
| System shows more than the shelf | 13 |
| More on the shelf than the system | 4 |
| A different item in the location | 5 |
- Inventory record accuracy
- 84%to97%
- Today, and the six-month target
- Annual cost of inaccurate records
- $1.8Mto$2.2M
- Write-offs, emergency buys, search time, credits, and carrying cost
- Excess stock released
- $3.0M
- Over 12 months, once accuracy holds
The situation
A distributor whose customers can now see its stock
Brackwell sells next-day availability to plant maintenance teams. Its online store shows live stock from the ERP, so every unit the system counts and the shelf lacks is a promise a customer sees and the warehouse cannot keep.
Where the two drift apart
- Put-awayScanned 72% of the time; drivers use the nearest open slot
- ReplenishmentScanned 61% of the time
- Transfers between DCsHalf received by keying the paper manifest
- ReturnsFewer than half scanned back into stock
What is on the shelf
Two distribution centers, a hose shop, and two will-call counters
- Counts matching the system
- 84%
- On record, location unknown
- $1.3M
- Orders short or late for stock
- 1 in 9
The brief
Why do customers get short shipments when our system says the stock is here?
Asked by the president, with two follow-on questions:
- What are inaccurate records costing the business?
- Do we need a new warehouse system to fix it?
Company profile
- Distribution centers
- Dorrance and Millgate
- Floor space
- 250,000 sq ft
- People
- 146
- Sales
- About $68M a year
- Our time on site
- 2 days, one per DC
What we found
Where the stock goes missing
Four views of the same warehouse, from the short orders back to the moves behind them.
1 of 4
Short orders, traced back
Of 41,200 orders in 13 weeks, 4,614 shipped short or late because of a stock problem. The largest cause was stock the system showed on the shelf that pickers could not find. Three more causes were moves the system never saw.
Four in ten short or late orders were for stock the system said was on the shelf
Order and pick exception records. Values adjusted for confidentiality.
2 of 4
Accuracy by area
Counts matched the system 84% of the time. Accuracy follows how stock moves: highest in fast-pick, where every pick is scanned, and lowest in overflow and temporary locations, where stock is parked without a location scan.
Record accuracy runs from 92% in fast-pick to 62% in overflow
Cycle count records, May to July 2026. Values adjusted for confidentiality.
3 of 4
Scans skipped in the middle
Receipts and pick confirmations are scanned almost every time, because the system will not move forward without them. Put-aways, replenishments, transfers, and returns depend on habit, and under pressure the scan is the step that gets skipped.
Scans happen at the ends of the process and get skipped in the middle
Handheld scan logs and ERP transactions. Values adjusted for confidentiality.
4 of 4
Buyers buy around it
Buyers raised minimums on about 2,900 SKUs because they do not trust the counts, which carries $3.0M of extra stock. Another $1.9M has not moved in two years, and $1.3M is on record in locations nobody can find.
About $6.2M of the $21.4M on hand does no work for the business
ERP valuation and min/max settings. Values adjusted for confidentiality.
Four in ten short or late orders were for stock the system said was on the shelf
4,614 of 41,200 orders shipped short or late because of stock problems, May to July 2026
Order and pick exception records. Values adjusted for confidentiality.
Try the recovery
Close the gaps between scans and watch the short orders fall
A simplified version of the order-failure model in the report. Start from today's operation, then try the 90-day plan.
Today drivers often use the nearest open slot and plan to scan it later.
The rest are keyed from the paper manifest at the end of the shift.
Returned items wait on a cage until someone has time.
Stock on the dock is on hand in the system and in no pick location.
Top items counted monthly; every miss traced to the move behind it.
The assembly shop records the fittings and length it actually used.
Inventory record accuracy
Cycle counts that match the system within tolerance.
84%against a target of 97%
Off target. One order in 9 ships short or late for a stock reason.
Orders shipped short or late for stock, 13 weeks
- Missing from the shelf
- 1,892
- Not yet put away
- 877
- Transfer not received
- 646
- Wrong bin
- 554
- Hose parts
- 415
- Other
- 230
- Short or late orders
- 1 in 9
- Excess stock to release over 12 months
- Hold until 95%
Based on 4,614 stock-driven short or late orders out of 41,200 over 13 weeks, traced to their causes, and 6,240 cycle counts. Release excess stock only after accuracy holds above 95% for eight weeks. Your model is built from your own order and count records.
The plan
Fix the process, then buy the system
Each phase has a gate the warehouse must pass before the next one starts.
- A location scan on every put-away and replenishment, with no move to an unconfirmed slot
- Transfer trucks received on arrival, by scan only
- Returns scanned back within one shift and the returns cage cleared daily
- Every overflow and temporary location labeled and scanned
- Cycle counts by value and velocity, with the cause of every miss recorded
- A daily dock clear and a one-page weekly supervisor briefing
Gate to Phase 2
95% record accuracy and 95% scan compliance, held for eight weeks.
The deliverable
What the 10-page report covers
Our short format: one question, two days on site, and a focused report with the evidence, the cost, a 90-day plan with owners, and the requirements for any new system.
- Executive summaryThe finding, the cost, what to do first, and the warehouse system decision.
- Context and methodThe business, the two sites, and how we traced every short order.
- FindingsShort orders by cause, accuracy by area, scan discipline, receiving, transfers, and stock levels.
- 90-day action planEvery action with an owner, a timeline, and an expected outcome.
- After 90 daysA three-phase roadmap and the requirements a new system should meet.
- AppendicesData sources, metric specifications, and financial impact.
How a site assessment runs
About three weeks, start to finish
These are the Study and Outline stages of our S.O.L.V.E. framework.
- Step 1 · 1 week
Before the visit
You send order exceptions, cycle counts, scan logs, and inventory valuation. We trace every short or late order back to the location and the move behind it.
- Step 2 · 2 days
On site
One day at each site. We watch receiving, put-away, picking, and transfers on both shifts, check 120 locations blind, and interview the floor, purchasing, and customer service.
- Step 3 · 2 weeks
After the visit
We quantify the cost, build the inventory model, and deliver the report with a 90-day plan, owners, and system requirements.
Sample deliverable. A composite, with names and figures changed to protect client confidentiality.