In wholesale trade in Portugal, the overall commercial margin was 4.6% in 2024 — according to INE, in a sector that employs 860,000 people and moves 201.8 billion euros. With margins like these, digitalising the warehouse is not a question of modernity: it is a question of arithmetic survival. Every mislocated pallet, every receiving note filled in by hand and transcribed two hours later, every inventory done with pen and A4 paper is margin that disappears before any negotiation with the customer.
The thesis of this article is uncomfortable: most Portuguese industrial warehouses that "have already tried mobility" did not fail because of the software. They failed because they kept paper as a backup — and the backup became the process. What follows is an honest operational comparison between the available options, with the trade-offs that sales presentations leave out.
The problem that paper hides so well
Paper works. That is the trap.
At a distribution platform in the Lousada-Paços de Ferreira corridor, with 8,000 m² and 40 operators, the paper receiving process appears to flow: the driver hands over the CMR, the checker signs, the note goes to the office, someone posts it into the ERP at the end of the day. Stock updates at 5.30 p.m. Until then, no one knows what is really available.
The problem is not paper itself. It is the time lag between the physical event and the digital record — and the decisions made in that interval with the wrong information. An order confirmed on the basis of stock that has already been allocated to another picking task. A batch of raw material received without validation with the supplier. A dispatch with the wrong quantity because the picking was checked visually, not by code scanning.
Paper does not lie — it simply arrives too late to prevent the error that has already happened.
The invisible cost of the lag
Quantify the lag in your warehouse before evaluating any solution. The exercise is simple: measure the average time between the physical arrival of a pallet and the moment the ERP reflects the correct stock. In paper-based operations, that interval is rarely under two hours. During receiving peaks — Monday morning, return from holidays, end of campaign — it can reach a full day.
With a margin of 4.6% in wholesale (INE, 2024), two days of phantom stock per week are not an operational problem. They are a profitability problem. Logistics and stock efficiency is what decides whether that margin stays in the company or evaporates in errors that no one accounts for.
What has changed to make this urgent now
Regulatory traceability and compliance
DL 28/2019 and the requirement for monthly SAF-T communication (Portaria 195/2020) created a new level of demand around the recording of stock movements. The AT can cross-reference transport documents, invoices and inventory movements. A warehouse that posts receipts two days late has, by definition, inconsistencies in the SAF-T that are difficult to justify in an inspection — and the risk is not theoretical: inspections of mid-sized distribution operators have increased in recent years precisely because file cross-referencing is now automatic.
The NIS2 Directive, transposed into Portugal by DL 65/2025, adds another layer: operators in sectors deemed essential or important — including large-scale distribution — have data traceability and integrity obligations that a paper-based process simply cannot guarantee in an auditable way. Having the software is not enough: there must be proof that the digital process is the real process.
Pressure from industrial customers
The international brands that subcontract production in Portugal — in clothing, footwear and textiles — require batch traceability, real-time dispatch confirmation and EDI integration. A supplier that still sends dispatch confirmations by email with a manually generated PDF is creating friction that the international buyer will eliminate — by switching supplier. This is not a hypothetical scenario: it is what happens when the brand's auditor visits the warehouse and finds paper.
Labour cost and turnover
The shortage of qualified warehouse operators in the North of Portugal is real and structural. Training a new checker in a paper-based process takes weeks — and they take the knowledge with them when they leave. A process supported by industrial mobility guides the operator step by step: the learning curve compresses from weeks to days, and the knowledge stays in the system, not in the head of someone who might leave on Friday.
Real technical options: what exists on the market
There are four configurations that we come across repeatedly in Portuguese industrial warehouses. Each has a different cost and risk profile — and none is universally wrong.
Pure paper means receiving, picking, inventory and dispatch on physical documents, with manual posting into the ERP at the end of the shift or the next day. Hardware cost: zero. Licence cost: zero. Real cost: the lag, the transcription errors and the impossibility of real-time traceability. For warehouses with fewer than 200 stable references and a single shift, it may still be defensible — but not for long.
Excel + manual ERP is the most dangerous variant. The operator records on paper, the warehouse manager consolidates in Excel, someone posts it into the ERP. Three points of failure, three moments of information loss. The illusion of control is more dangerous than pure paper — because it looks like there is a process. It is the most common configuration in Portuguese industrial SMEs with 10 to 50 operators, and it is the one that most often precedes an inventory crisis discovered at the worst possible moment.
Generic apps disconnected from the ERP capture the movement but do not close the loop: stock in the ERP continues to be updated manually or in batch. Useful for very small warehouses with a simple ERP. Insufficient for operations with multiple references, batches, or wave picking. The specific risk here is CSV file integration — which fails silently when the format changes, and no one notices until the next inventory.
WMS integrated with industrial mobility — where KORA Inventory Suite fits — uses industrial terminals or ruggedised smartphones with barcode or QR scanning, integrated in real time with the MULTI ERP, covering receiving, checking, put-away, picking, packing list and dispatch. The physical movement and the digital record are simultaneous — not sequential. The implementation cost is the highest of the four options. The operational risk is the lowest — provided the implementation is done without shortcuts.
| Option | Stock lag | Batch traceability | Implementation cost | Operational risk |
|---|---|---|---|---|
| Pure paper | 2–24 hours | Manual / non-existent | Zero | High (errors, loss, SAF-T) |
| Excel + manual ERP | 1–8 hours | Partial / not auditable | Low | High (three points of failure) |
| Generic disconnected app | Batch (hours) | Limited | Low–Medium | Medium (fragile integration) |
| Integrated WMS (KORA Inventory) | <30 seconds | Complete, auditable | Medium–High | Low (guided process) |
Trade-offs by scale: what no one tells you in the sales presentation
Real cost vs. visible cost
The licence and hardware cost of an industrial mobility solution is visible in the proposal. The cost of paper is on no line of the budget — it is dispersed across hours of re-posting, returns due to picking errors, annual inventories that immobilise the warehouse for two days, penalties from customers for wrong dispatches. Calculate the ROI with this logic: how many hours per week are spent on manual re-posting? What is the cost of a return due to a picking error — transport, reprocessing, credit note? How many physical inventories do you do per year and how many people do you immobilise for how long? These numbers exist within the company. They just are not brought together in the same document.
Implementation time
A well-executed industrial mobility implementation — with process mapping, terminal configuration, ERP integration and operator training — typically takes between 8 and 16 weeks, depending on the complexity of the warehouse. Warehouses with multiple zones, cross-docking and wave picking are at the upper end. Warehouses with a simple receiving-put-away-dispatch flow can be operational in 8 weeks.
The most common mistake: compressing this timeframe to 4 weeks under calendar pressure and going live without adequate training. The result is the worst of both worlds — paper comes back as a "backup" and never leaves again. We return to this point later, because it is the most frequent and least documented failure mechanism.
Infrastructure complexity
Industrial mobility assumes reliable Wi-Fi coverage throughout the warehouse. In a warehouse with tall metal structures and cold-storage areas, that is not trivial. Assess the network infrastructure before assessing the software — a poorly positioned access point in an aisle of 12-metre-high racks creates blind spots that the operator will work around by going back to paper. The rule of thumb: do the Wi-Fi site survey with the warehouse in normal operation, not empty. Interference from forklifts and metal structures only appears with the warehouse running. Define a minimum coverage of -65 dBm at all operating points before awarding the hardware.
Operational resistance
The warehouse supervisor who has worked with paper for 15 years knows every colour reference by heart. To him, the terminal is an obstacle, not a help — at least in the first two months. This resistance is legitimate and predictable. Ignoring it is the number-one cause of failed rollouts — not the software, not the integration, not the Wi-Fi.
The terminal does not replace the warehouse supervisor's knowledge. It documents it — and makes it transferable to whoever comes next.
Decision matrix: when it makes sense to proceed
| Criterion | Paper still acceptable | Mobility urgent |
|---|---|---|
| Volume of references (SKUs) | <200 stable references | >500 references or high seasonal variation |
| Warehouse operators | <5 operators, 1 shift | >10 operators or 2+ shifts |
| Batch traceability | Not required by customer or regulator | Required (food, pharmacy, sustainable textiles) |
| Picking/dispatch errors | <0.5% of lines | >1% of lines (or any customer penalty) |
| Physical inventory | 1x/year, <1 day of downtime | Rolling inventory needed or >2 days of downtime |
| Integration with customers | No EDI, no ASN required | EDI, ASN or real-time dispatch confirmation |
| SAF-T compliance | Simple operation, low risk of inconsistency | History of inconsistencies or recent AT inspection |
What works in practice
Pattern 1: receiving as a single point of entry
The most common mistake in industrial mobility implementations is trying to digitalise everything at once. The pattern that works: start with receiving. It is the moment when stock enters the system — if it is captured correctly here, all subsequent movements are more reliable. Implement receiving with a terminal and code scanning during the first 4 weeks. Only then move on to picking and dispatch.
At a construction materials distributor with 15,000 references and three receiving docks, digitalising receiving first eliminates the manual posting into the ERP at the end of the day and reduces the inventory discrepancies that were previously only discovered at the annual inventory — by which point there is no longer any way to work out where and when the error happened. Digital receiving creates the proof that stock came in, with which batch, in what quantity and at what time. That record did not exist on paper.
Pattern 2: rolling inventory as continuous validation
With mobility, the two-day annual inventory ceases to make sense as the only way of validating stock. The mature pattern is rolling inventory: every day, an operator counts the references of a specific zone with the terminal, and the system compares them with the theoretical stock. Discrepancies are handled on the same day, not accumulated over a year. The practical effect is that the annual inventory, when it is still done, confirms what is already known — instead of revealing surprises that paralyse the warehouse for two days.
This pattern is particularly relevant for distribution warehouses with high stock rotation and for textile operations with colour and composition batches that require traceability to meet the requirements of the EU Strategy for Sustainable and Circular Textiles.
Pattern 3: digital packing list as proof of dispatch
Dispatch confirmation with a packing list automatically generated by the terminal — with references, quantities, batches and number of units — eliminates the dispute with the customer over what was sent. It is also the basis for the automatic issuing of the transport document and the invoice, closing the documentary loop without manual intervention. For operations that export — and where Incoterms define transport and insurance responsibilities — the timestamped digital packing list is a piece of evidence in the event of a dispute. Paper with the checker's handwriting does not carry the same weight in an international commercial arbitration tribunal.
How to implement: operational sequence
- Audit the current process before touching the software. Map every physical stock movement — receiving, checking, put-away, picking, dispatch, returns — and identify where paper is generated, who reads it, who posts it and with what delay. This mapping takes two to three days and is the most valuable document of the entire implementation. Without it, you configure the system for the process you imagine, not for the process that exists.
- Validate the network infrastructure with the warehouse in operation. Do the Wi-Fi site survey with forklifts circulating and structures loaded. Define a minimum coverage of -65 dBm at all operating points. One extra access point costs 300 euros. An implementation that fails due to insufficient coverage costs months.
- Configure the terminals for the process, not the process for the terminals. The operator's screen should reflect exactly what they do — not a simplified version of what the software is able to display. If the receiving process has batch checking and expiry date, the terminal must request those fields in that order, without additional navigation. Every extra click is an opportunity for the operator to go back to paper.
- Go live with a pilot group of 3 to 5 operators for 4 weeks. Choose the most sceptical operators — not the most enthusiastic ones. If the process works with them, it works with everyone. Use this period to adjust flows, not to validate the purchasing decision.
- Measure before and after with the same indicators. Picking error rate (lines with errors over total lines), average receiving time per pallet, stock lag at any moment of the day, number of discrepancies discovered per week. Without a baseline, there is no argument for the next phase of investment — neither for the board nor for PT2030 funding.
The insight the manuals do not mention: the real failure mechanism
There is a pattern we see repeatedly in Portuguese warehouses that attempted the transition without completing it. Paper comes back as a backup in exception situations — unstable Wi-Fi, a terminal with a low battery, a new operator who does not know how to use the system. It seems reasonable: it is just to keep the operation running. Six months later, half the movements are done on paper and posted manually, and the company has the worst of both worlds: the cost of the mobility system and the operational cost of paper.
The detail the manuals leave out: the paper backup is not adopted by the most resistant operators. It is adopted by the most experienced ones — who know that the process must continue even when the system fails, and who have the informal authority to decide that. They are precisely the people whose example the others follow. When the shift supervisor picks up paper "just this once", the whole team realises the system is not mandatory.
The solution is not technical. It is a matter of process: explicitly define that there is no paper backup. If the terminal fails, the movement waits until the terminal is available. This rule will generate resistance — but it is the only way to guarantee that the digital process is the real process, not a parallel layer that coexists with paper indefinitely.
Industrial mobility is not an IT project. It is a process decision that IT supports. Those who decide with the IT director alone fail — those who decide with the warehouse supervisor in the room have a much higher probability of success.
Post-implementation success metrics
What to measure in the first 12 weeks
Five indicators define whether the implementation is working or merely appears to be working. The picking error rate — lines with errors over total lines dispatched — should be below 0.3% in warehouses with well-implemented mobility; above 1% is a sign that the guided process is not being followed. The receiving time per pallet, including checking, code scanning and confirmation in the system, should be compared with the previous average time — paper plus manual posting — and not just with the code-scanning time. The stock lag between physical and ERP at any moment of the day should be under 30 minutes under normal conditions. The rolling inventory coverage — percentage of references counted at least once a month — should reach 100% of A and B references and 80% of C references. And the number of movements recorded on paper should be zero: if this indicator is not explicitly measured, the paper backup installs itself without anyone declaring it.
What not to measure
Do not use "number of active terminals per shift" as a success metric. A warehouse with mature mobility has operators who use the terminal in every movement — but the relevant indicator is the outcome of the process, not the use of the hardware. A terminal used to bypass the process does not count. And a warehouse with 100% terminal usage and a 2% picking error rate is in a worse position than one with 80% usage and a 0.2% error rate.
For a broader perspective on how shop-floor digitalisation interacts with production and OEE, the article KORA Productivity: real-time shop-floor control develops the logic of real-time capture applied to production — which shares the same terminal infrastructure and the same philosophy of eliminating the lag. And for those assessing the IT infrastructure that supports these implementations, the industrial IT infrastructure guide for operations directors covers the network, hardware and security requirements that an industrial mobility operation demands.
KORA Inventory Suite integrates natively with the MULTI ERP and with QAD Adaptive ERP — which eliminates the integration layer that, in generic solutions, is frequently the most fragile point of failure. For operations that combine warehouse with a mobile sales force, KORA Sales Suite shares the same platform and the same real-time stock data.
The question is not whether paper will disappear from the Portuguese industrial warehouse. It is whether the transition will be made in such a way that the digital process is the real process — or just one more layer over the paper that never left. The difference between the two situations does not lie in the software. It lies in the decision not to have a backup.
Sources
- INE — Statistics Portugal, Trade Statistics 2024. Data on turnover (€201.8 bn), number of companies (218,800), workers (860,600) and overall commercial margin (4.6%) in wholesale trade in Portugal.
- Decree-Law no. 28/2019, of 15 February — Rules applicable to the processing of invoices and other tax-relevant documents. Diário da República, 1st series, no. 31.
- Ordinance no. 195/2020, of 13 August — Monthly communication of the SAF-T(PT) file to the Tax and Customs Authority. Diário da República, 1st series, no. 157.
- Directive (EU) 2022/2555 of the European Parliament and of the Council (NIS2), of 14 December 2022, on measures for a high common level of cybersecurity across the Union. Official Journal of the European Union, L 333.
- Decree-Law no. 65/2025 — Transposition of the NIS2 Directive into the Portuguese legal order. Diário da República.
Frequently asked questions
What is the commercial margin in Portuguese wholesale trade and why does it matter for digitalisation?
The overall commercial margin in wholesale trade in Portugal was 4.6% in 2024. With margins so slim, every operational inefficiency — a mislocated pallet, a hand-filled note, a paper inventory — represents a direct loss of profitability. Digitalising the warehouse ceases to be modernity and becomes an arithmetic necessity for survival.
What is the main problem with paper in an industrial warehouse?
Paper does not lie, but it arrives too late. The real problem is the time lag between the physical event (a pallet's arrival) and the digital record in the ERP. In that interval, decisions are made with the wrong information: orders confirmed with already-allocated stock, dispatches with the wrong quantity, or batches received without validation.
On average, how long does it take to update stock in the ERP with paper-based processes?
In paper-based operations, the interval between the physical arrival of a pallet and the correct stock update in the ERP is rarely under two hours. During receiving peaks — Mondays, return from holidays or end of campaign — it can reach a full day, creating phantom stock that affects business decisions.
What regulatory obligations have made digitalisation more urgent?
DL 28/2019 and Portaria 195/2020 require monthly SAF-T communication. The AT automatically cross-references transport documents, invoices and inventory movements. A warehouse with two-day delays in posting receipts generates SAF-T inconsistencies that are difficult to justify in an inspection. The NIS2 Directive (DL 65/2025) adds traceability and data integrity obligations for essential operators.
Why do international customers require digital traceability?
The international brands that subcontract production in Portugal — clothing, footwear, textiles — require batch traceability, real-time dispatch confirmation and EDI integration. A supplier that sends confirmation by email with a manual PDF creates friction that the buyer eliminates by switching supplier. This is not hypothetical: it happens when auditors visit warehouses with paper-based processes.
How does industrial mobility reduce the training time for new operators?
A paper-based process takes weeks to train a new checker, and the knowledge leaves with them when they give up the role. Industrial mobility guides the operator step by step, compressing the learning curve from weeks to days. The knowledge stays in the system, not in the head of someone who might leave on Friday.
Which is the most dangerous configuration among pure paper, Excel and manual ERP?
Excel + manual ERP is the most dangerous variant. The operator records on paper, the manager consolidates in Excel, someone posts it into the ERP — three points of failure and three moments of information loss. It is the most common configuration in Portuguese industrial SMEs with 10 to 50 operators and frequently precedes inventory crises discovered at the worst possible moment.