With an overall commercial margin of 4.6% in wholesale trade in Portugal (INE, 2024), there is no room for stock location errors. An unnecessary transfer between warehouses, a stockout at a dispatch point when there is stock available 40 km away — these events are not operational inconveniences. They are the difference between a month in the black and a month in the red. The thesis of this guide is simple and uncomfortable: most Portuguese distribution companies do not have a stock problem. They have a stock visibility problem — and they solve it by buying more, when they should configure better what they already have.
At the end you will find a 12-point checklist to audit your multi-warehouse management in under two hours.
What you need before you start
Before configuring replenishment rules or creating picking zones, confirm that you have these foundations in place. Without them, any advanced configuration will amplify existing problems rather than solve them.
You need an up-to-date physical map of each warehouse — zones, aisles, coded locations. You need a clear definition of which warehouse is for receiving, which is for dispatch and which are for transit or consignment. The safety stock policy must be per warehouse, not global: a single reorder point for three distinct physical locations is a recipe for unnecessary transfers and duplicate purchases.
Also define the stock ownership rules — who can consume from warehouse B when A is out of stock — and confirm that the integration between the WMS and the ERP generates a document and moves stock in real time on every internal transfer. Operational barcode or RFID at all entry and exit points. And a named person in charge per warehouse with individual write access to the system — not a shared login that no one audits.
Step 1 — Map the real flows, not the org-chart flows
The most frequent mistake we see in INFOS projects: the company designs the stock flow as it should be, not as it is. The raw materials warehouse feeds production — in theory. In practice, the finished-goods warehouse has boxes of RM that "ended up there" three weeks ago because production needed the space. No one created a document. The system does not know.
Do this exercise before any configuration. Physically follow a purchase order from receiving through to consumption or dispatch. Record every point where stock changes location without generating a document in the system. Identify the "ghost warehouses" — physical zones with real stock but no code in the ERP. Document who authorises transfers between warehouses and by what means: paper, WhatsApp, verbally. In distribution companies of the Lousada–Paços de Ferreira corridor, the usual answer is "Zé knows where it is" — and Zé is on holiday in August.
Any multi-warehouse configuration built on incorrect flows will amplify errors, not correct them.
- ☐ Real flow documented and validated by the warehouse manager
- ☐ Ghost warehouses identified and coded or eliminated
- ☐ Transfer authorisations formalised
Step 2 — Define the warehouse hierarchy in the ERP
A warehouse management system that does not know the hierarchy between locations will suggest absurd transfers: looking for stock in a warehouse 40 km away when there is stock available on the shelf next door, simply because they are in different "entities" in the system. We have seen this happen with generalist ERPs that treat all warehouses as equal — no priority, no type, no consumption logic.
The correct hierarchy is: Company → Establishment → Warehouse → Zone → Aisle → Location. Each warehouse is given an explicit type — receiving, production, dispatch, returns, consignment, quarantine — and consumption priority rules: the system must know that warehouse A fulfils sales orders before touching warehouse B. Configure automatic transfer alerts when stock drops below the reorder point, but only after testing the priorities with real orders.
The warehouse hierarchy is not a configuration detail — it is the backbone of all replenishment logic. If it is wrong, the MRP will calculate incorrect requirements and purchase orders will fire for items that are in stock, but in the wrong warehouse.
- ☐ Hierarchy configured in the ERP with all levels
- ☐ Warehouse types assigned
- ☐ Consumption priority rules defined and tested
Step 3 — Standardise internal transfer processes
A transfer between warehouses without a document is stock that disappears from the system. It is that simple. And in distribution companies with three to five warehouses in the Lousada–Paços de Ferreira corridor, this happens dozens of times a day — with a radio in hand and no terminal in sight. The warehouse manager knows what he did. The system does not.
The rule is no exceptions: every internal transfer generates an internal transport note, even between warehouses in the same building. Stock only changes location in the system when the recipient confirms receipt — not when the sender dispatches. Urgent transfers have a simplified approval flow, but not one that is eliminated. And there is a daily cut-off for reconciling pending transfers — the end of the afternoon shift is the right moment, before the night shift starts working on out-of-date data.
The detail that the manuals do not mention: the biggest generator of undocumented transfers is not urgency — it is the lack of a terminal at the exit point. If the employee has to walk 80 metres to record a transfer, they will not record it. Place mobile terminals or scanning points in the exit aisles of each warehouse before implementing any process rule.
- ☐ Internal transfer note mandatory on all movements
- ☐ Receipt confirmation before stock update
- ☐ Daily reconciliation cut-off implemented
- ☐ Recording terminals available at exit points
Step 4 — Configure inter-warehouse replenishment
Automatic inter-warehouse replenishment is where most companies lose control. They set a global reorder point and then realise that the system is replenishing the dispatch warehouse from the production warehouse — which is halfway through a batch and cannot be interrupted. Result: the batch stops, dispatch is delayed, and the production manager calls the warehouse manager at 5.45 pm on a Friday.
Follow this configuration sequence:
- Set minimum and maximum stock per warehouse, per reference — not per product family.
- Configure the preferred source warehouse for each type of replenishment.
- Establish time windows: do not allow automatic transfers during dispatch peaks.
- Test with 20 high-turnover references before activating for the full catalogue.
- Review the replenishment suggestions weekly during the first month.
- ☐ Minimum/maximum stock defined per warehouse and per reference
- ☐ Source warehouse configured per movement type
- ☐ Replenishment time windows defined
- ☐ Test with sample references completed
Step 5 — Implement real-time visibility
Having stock in three warehouses without consolidated real-time visibility is worse than having a single warehouse. The false sense of abundance leads to unnecessary purchases and simultaneous stockouts in different locations. The buyer sees a total stock of 400 units, not knowing that 380 are in the wrong warehouse and the 20 correct ones have already been reserved for another order.
Configure operational dashboards that show, on the same screen: available stock per warehouse and per location in real time; transfers in transit — stock that has left A but has not yet arrived at B; picking orders in progress per warehouse; imminent-stockout alerts with the available alternative warehouse. With Qlik Sense it is possible to cross-reference this data with sales history and seasonality, anticipating stockouts before they happen — not after the customer calls.
- ☐ Consolidated stock dashboard per operational warehouse
- ☐ In-transit transfers visible in real time
- ☐ Stockout alerts configured with warehouse alternative
Audit checklist — 12 points in 2 hours
| # | Audit point | How to check | Estimated time |
|---|---|---|---|
| 1 | All physical warehouses have a code in the ERP | Compare ERP list with physical map | 10 min |
| 2 | Theoretical stock vs. physical stock in 3 high-turnover references | Manual count in each warehouse | 20 min |
| 3 | Undocumented transfers in the last 30 days | Cross-reference movement log with issued notes | 15 min |
| 4 | Quarantine stock identified and physically separated | Physical visit to the quarantine zone | 5 min |
| 5 | Consumption priority rules tested | Simulate a sales order with stock in two warehouses | 10 min |
| 6 | Reorder point defined per warehouse (not global) | Check the parameterisation in the ERP | 5 min |
| 7 | Named person in charge per warehouse with individual access | Check user profiles in the system | 5 min |
| 8 | Average receipt confirmation time under 2 hours | Analyse the transfer log for the last month | 10 min |
| 9 | Stock dashboards visible at the workstations of warehouse managers | Check access and real-time updates | 5 min |
| 10 | Rolling inventory in progress (not just annual inventory) | Check the schedule and count records | 5 min |
| 11 | WMS-ERP integration with no synchronisation errors in the last 7 days | Check the integration error log | 5 min |
| 12 | Stock discrepancy report generated and reviewed weekly | Check the date of the last report and who received it | 5 min |
Common mistakes and how to avoid them
Mistake 1 — A single warehouse code for physically separate locations. This happens when the company grows and rents the building next door without creating a new warehouse in the ERP. The stock gets mixed up and picking becomes a guessing game. Create a separate warehouse in the system, even if it is in the same postcode, and transfer the stock formally with a document.
Mistake 2 — Automatic replenishment without time windows. The system launches transfers during the 2 pm–4 pm dispatch peak, blocking the aisles with pallets in transit. Configure automatic replenishment windows for low-activity periods — early morning or start of the day — and lock the system during dispatch windows.
Mistake 3 — Consignment stock in the same warehouse as own stock. This mistake has direct fiscal consequences: consignment stock is not the company's property and should not appear on the balance sheet. With the monthly SAF-T reporting required by Portaria 195/2020, discrepancies between accounting stock and physical stock become visible to the AT. Create a separate consignment warehouse with a distinct ownership type — it is not a best practice, it is an obligation.
Mistake 4 — Annual inventory as the only control mechanism. An annual inventory detects the accumulated errors of 12 months — when there is nothing left to be done. Implement a rolling inventory: divide the catalogue into groups and count one section per week. High-turnover references should be counted monthly. Discrepancies detected in good time still allow you to correct the process that generated them.
Multi-warehouse management does not fail for lack of technology. It fails because the real process was never documented, the hierarchy in the system was configured in a rush at go-live and no one reviewed it again. The checklist above does not replace a technical assessment — but in two hours it tells you whether you have a configuration problem or a process problem. They are different solutions and they cost differently.
Frequently asked questions
How can I tell whether my problem is stock or visibility?
Carry out a full physical inventory and compare it with the system. If the difference is greater than 5%, the problem is visibility — unrecorded transfers, ghost warehouses or a lack of barcodes. If the inventory matches but you still have stockouts, the problem is replenishment configuration or an incorrect warehouse hierarchy in the ERP.
What is the difference between a ghost warehouse and a coded warehouse?
A ghost warehouse is a physical zone with real stock but no code in the ERP — the system does not know about it. A coded warehouse is registered in the system's hierarchy with a type, location and priority rules. Without coding, the stock is invisible to the MRP and purchase orders fire unnecessarily.
Why do transfers between warehouses need an internal transport note?
Because without a document, the stock disappears from the system. The note guarantees traceability and enables daily reconciliation. Stock only changes location in the ERP when the recipient confirms receipt, not when the sender dispatches. This avoids duplications and false stockouts.
What is the warehouse hierarchy and why is it important?
It is the Company → Establishment → Warehouse → Zone → Aisle → Location structure. It defines consumption priorities — which warehouse fulfils sales orders first. Without a correct hierarchy, the system suggests absurd transfers between locations and the MRP calculates incorrect requirements.
How do I avoid unnecessary transfers between warehouses?
Configure clear consumption priority rules in the ERP — the system must know that warehouse A fulfils orders before warehouse B. Set reorder points per warehouse, not global ones. And place mobile terminals at the exit points to ensure that transfers are recorded in real time.
How long does it take to audit multi-warehouse management?
The article includes a 12-point checklist that allows you to audit in under two hours. It starts by mapping real flows, then validates the hierarchy, transfer processes and replenishment configuration. The time invested saves weeks of subsequent implementation.
Do I need RFID or barcodes in all warehouses?
Yes, at least operational barcodes at all entry and exit points. RFID is optional, but barcodes are mandatory for traceability and reconciliation. Without this, transfers are not recorded and the system becomes out of date.
Sources
- Instituto Nacional de Estatística (INE) — Wholesale Trade Statistics in Portugal, 2024
- ISO/IEC 27001:2022 Standard — Information Security Management Systems (reference for auditing integrated systems)
- Associação Portuguesa de Logística (APL) — Best Practice Guides in Warehouse Management and Distribution
- IAPMEI — Logistics Modernisation Programme for Portuguese SMEs