Batch-to-batch traceability in Portuguese industry: ERP or paper

A ruled notebook, a black cover, corners curled by the humidity of the dyeing plant. Inside, in pen, each dye bath noted down with the batch number, the colour recipe, the time in and the time out. This notebook is the only proof of traceability at a finishing unit in the Vale do Ave — and when the client-brand asks for the full history of an item with a colour fastness problem, someone spends an afternoon leafing through six months of notes. They find the batch in two hours. Or they don't find it.

The thesis of this article is simple and uncomfortable: the paper notebook doesn't fail because it's paper — it fails because it doesn't cross-reference with anything else. Batch-by-batch traceability is only worth something when it links raw material, production, stock and dispatch in the same system. Without that link, you have an archive. You don't have traceability.

Over 35 years implementing vertical software in factories in the North, we have seen the same scene repeat itself in different contexts: the garment maker near Famalicão that reprocesses 4,000 pieces because it doesn't delimit the yarn batch, the food warehouse in Paços de Ferreira that withdraws ten times more units from the market than the problem warranted, the sole factory in Felgueiras that doesn't know which other pairs used the defective component. In all of them, the common denominator is not a lack of rigour. It's a lack of connection.

1. The real operational problem

Traceability is not an abstract audit requirement. It is the answer to a concrete question that arrives by email, usually on a Friday late afternoon: "Which yarn batch was used in this order? Which supplier did it come from? Which machine processed it?" The brand needs the answer within hours. The factory that only has the notebook needs days.

The time asymmetry is the heart of the problem. The client-brand — Inditex, Decathlon, Tom Tailor, Lacoste — works in tight reaction cycles. A defect detected in a store in Berlin has to be traced back to source before the same reference reaches the remaining stores. The Portuguese supplier is at the end of a chain that does not forgive slowness. When the answer takes three days, the brand is no longer asking about the cause of the defect — it's asking whether it's worth renewing the subcontract the following year.

The textile case: when the colour doesn't hold

At a dyeing and finishing unit in the textile cluster, a colour fastness problem can originate in three places: the greige yarn batch, the bath recipe, or the dye batch. Without batch-by-batch traceability, the factory cannot isolate the cause — and ends up reprocessing or scrapping far larger quantities than the real problem warranted. The cost of the defect multiplies because it cannot be delimited.

Consider the concrete mechanics. A dyeing plant in Vizela processes, in a single shift, baths that consume yarn from several suppliers and dyes from different suppliers. If fastness fails in a batch of dispatched knit fabric, the immediate question is: was it the greige yarn, was it the recipe, was it the dye? Without systematic cross-referencing, the factory investigates by elimination — testing hypotheses, redoing control baths, consuming days of laboratory time. And, when in doubt, it scraps everything that might be affected, because it cannot draw the exact boundary of the compromised batch. What should have been 200 kilos of scrap becomes two tonnes of scrap.

When the brand invokes compliance with the European strategy for sustainable and circular textiles, the demand goes up another notch: they want to know the origin of the fibre, the footprint of the process, the complete journey. This cannot be answered with a notebook. ATP has repeatedly warned that compliance with these requirements is no longer a competitive differentiator and is becoming an entry ticket — without it, there is no access to certain value chains.

The apparel case: the garment maker that doesn't own the product

Subcontract garment making lives a particular variant of the problem: often it doesn't own the raw material. The parent house sends the fabric, the trims, sometimes even the labels, and the Portuguese factory provides the garment-making service. The traceability the brand demands includes knowing which roll of fabric — with which origin batch number — entered which manufacturing order and left in which dispatch box.

Here the challenge is aggravated by an ageing workforce and by turnover. The knowledge of "which roll went to which order" often lives in the head of the line supervisor, who knows it by heart but doesn't record it. When that person is absent, or retires, the traceability leaves with them through the door. Digitalisation, here, is not about technology — it's about transferring tacit knowledge to a system before it disappears.

The footwear case: 800 to 1,200 SKUs per collection

In Felgueiras, a footwear sample collection easily has between 800 and 1,200 distinct SKUs, crossing three axes — colour, size and last/fitting. Each model consumes components from various suppliers: leathers, soles, linings, eyelets, glues. When an international buyer detects a defect in a specific sole, the question is immediate: which other pairs used this same sole, from this same batch?

The sector's cadence aggravates the pressure. International buyers visit twice a year — men's footwear in August, women's footwear in February. In that window, they decide entire collections. A quality complaint poorly resolved in one season contaminates trust in the next. And APICCAPS never tires of stressing that the competitiveness of Portuguese footwear rests on quality and rapid response, not on price — which makes traceability a pillar of the cluster's own value proposition, not an administrative cost.

A generalist ERP doesn't model these three axes well simultaneously. It treats colour-size-last as loose variants, and traceability breaks at the point where it matters most. That's why the footwear sector needs a vertical product structure, not a flat catalogue forcibly adapted.

A defect in a sole is not a defect in one pair. It's a defect in every pair that used that sole batch — and the only question that matters is how many they are and where they are.

The metal/plastic case: the genealogy of the critical component

In the Aveiro-Marinha Grande corridor, the mould and plastic injection industry faces traceability in its most demanding form. An injected component for the automotive industry has to be traceable back to the raw material batch, the injection machine, the process parameters, the mould used and the production shift. AIMMAP and the very demands of the car manufacturers impose traceability records that, at the limit, reach the individual serial number.

Here the cost of a traceability failure is not reputational — it's existential. A poorly delimited automotive recall can mean contractual penalties that exceed the annual turnover of a mould SME. The margin of error is zero, and that's why this sector has, historically, digitalised traceability earlier than textiles and footwear.

The food distribution case: the recall that has to be surgical

At a food distribution warehouse in the Lousada/Paços de Ferreira corridor, traceability is not optional — it's law. Regulation (EC) No 178/2002 mandates traceability "one step back, one step forward" throughout the food chain. When there is a safety alert about a supplier batch, the company has to know, within minutes, which customers it dispatched that batch to and in what quantities.

The difference between a recall that withdraws 300 units and a recall that withdraws 30,000 is, almost always, the quality of the traceability. Paper doesn't delimit. The system delimits.

The warehouse manager knows this all too well. They also know that any new system that pulls them off the floor for more than two hours a day will meet resistance. This is the central tension of any traceability project: the rigour that management wants and the friction that the shop floor feels. A warehouse management system that captures the batch at picking without adding ten seconds to each movement is the difference between adoption and silent sabotage.

Shelf life adds a dimension that textiles don't have: the food batch has a date. FEFO — first expired, first out — obliges the warehouse to dispatch by expiry order, not by entry order. Without date-based batch traceability, FEFO is a good intention that converts, in practice, into losses through out-of-date product. Traceability, here, protects margin even before it protects public health.

2. What exactly is batch-by-batch traceability in industry in Portugal

Traceability is the ability to reconstitute the complete journey of a material or product — backwards (where it came from) and forwards (where it went to) — at the level of the batch, not just the generic item.

Batch, serial number and the difference that changes everything

There are three tracking granularities, and confusing them is the most common mistake in specifications:

• Item-level traceability — you know you used "combed 30/1 yarn", but you can't distinguish batches. Practically useless for investigating defects.
• Batch-level traceability — you know you used batch 20240517-A of the yarn, with this recipe, on this machine, on this date. Sufficient for 90% of industrial cases.
• Serial-number traceability — identifies each individual unit. Necessary for critical components (automotive, aeronautics, medical devices), excessive in most textiles and footwear.

The granularity decision is not technical — it's economic. Tracking by serial number in a sock factory is burning money. Tracking only by item in a plastic injection for an automotive component is taking on a recall risk the company won't survive.

The following table summarises where each granularity makes sense in the Portuguese industrial fabric:

Upstream, downstream and internal traceability

The ISO 9001 standard and quality management systems distinguish three directions:

• Upstream — from the finished product back to the raw material and the supplier. Answers "where did the problem come from?".
• Downstream — from the raw material or production batch to the customers who received it. Answers "who was affected?".
• Internal — within the unit itself: which machine, which operator, which shift, which recipe. This is where most Portuguese factories have the biggest hole.

Complete traceability requires all three. The dyeing plant's notebook covers — poorly — the internal one. It doesn't cross-reference with purchasing or with dispatch. And it's that cross-referencing that transforms loose data into real traceability, as we detail in batch-by-batch traceability in production.

It's worth stressing why internal traceability is the weakest link. Purchasing is already, almost always, in a system — the supplier issues an invoice, there is a recorded receipt, the origin batch enters a file. Dispatch too — the delivery note is mandatory, the invoice is reported to the AT. The hole is in the middle: what happens to the batch between receipt and dispatch. It's on the shop floor, precisely where there is no keyboard, that traceability is lost. Closing that hole requires capturing the data at the moment and place of transformation — not at night, from memory, in the notebook.

Origin batch and internal batch: the transformation that creates genealogy

There is a concept that separates serious traceability from façade traceability: batch genealogy. When the greige yarn batch enters dyeing and leaves as a dyed knit batch, a new batch is born — but that new batch has to maintain the link to the origin batch that composed it. A pair of shoes has a genealogy of dozens of components, each with its own batch. Losing any link in the genealogy breaks the entire chain.

This is why it's not enough to "have batches". The system has to maintain the parent-child relationships between batches throughout each transformation. An Excel doesn't do this reliably — the relationships live in fragile formulas or in columns that no one guarantees. A vertical ERP treats genealogy as a native structure: each transformation automatically records which input batches it consumed to produce which output batch.

A brief history: from HACCP to the Digital Product Passport

Traceability was formalised first in the food chain, with the HACCP principles in the 90s and then with European food legislation. From food it passed to pharmaceuticals, to automotive, to aeronautics. Textiles and footwear arrived late — and are arriving now with double pressure: from client-brands and from the European regulator, which is moving towards the Digital Product Passport under the European ecodesign regulation for sustainable products. When that passport becomes mandatory for textiles, the factory that doesn't have digital traceability will not be able to comply — not out of ill will, but out of physical impossibility of reconstituting the history from notebooks.

The historical lesson is consistent: each sector that adopted traceability did so first out of regulatory obligation and only then discovered the operational benefit. Textiles and footwear are exactly at that inflection point — the obligation is coming, and the factories that anticipate it will discover, as food did before them, that traceability pays for itself in avoided scrap and delimited recalls long before the regulator knocks on the door.

3. The picture in Portugal today

Portugal has an industrial base concentrated and ageing in processes, but with growing digitalisation pressure coming from the international value chains. Some figures anchor the context.

The scale of the logistics problem

Commerce in Portugal moved 201.8 billion euros in 2024, distributed across some 218,800 companies and 860,600 workers, according to INE. Within this universe, the overall trade margin in wholesale trade was just 4.6% in 2024 (INE). On such a thin margin, every stock error, every poorly delimited scrap, every overly broad recall eats directly into profitability. Traceability is not a compliance cost — it's a margin protection mechanism.

With wholesale margins of around 4.6%, the difference between a surgical recall and a panic recall can be the difference between the year closing in profit or in loss.

The industrial fabric of the North

Textiles and apparel represent a relevant slice of Portuguese goods exports, with the Vale do Ave cluster concentrating the highest density of spinning, knits and finishing in the country. Footwear, led by the Felgueiras and Guimarães region, exports the overwhelming majority of its production. These two realities share one trait: they depend on international buyers who increasingly impose traceability and compliance requirements as a condition of supply — not as a friendly request.

Geography also matters. The concentration of textiles in Famalicão, Guimarães, Barcelos and Vizela, and of footwear in Felgueiras, S. João da Madeira and Oliveira de Azeméis, creates dense and short subcontracting chains. A parent house subcontracts the garment making, which subcontracts the embroidery, which subcontracts the printing. Traceability has to cross these boundaries between companies — and that's where it breaks most frequently, because each link has its own system, or none.

The predominance of the micro-enterprise and its effect on traceability

The Portuguese business fabric is dominated by micro and small enterprises. According to INE and IAPMEI data, the overwhelming majority of companies in Portugal have fewer than ten workers. This structure has a direct consequence for traceability: in a subcontracting chain, the weakest link defines the rigour of the whole journey. It only takes one micro garment maker without a system to break the traceability of an order that passed through five digitalised companies.

That's why client-brands are beginning to demand digital traceability not only from direct suppliers, but from the whole chain. And that's why the digitalisation pressure descends from the large company to the micro — not out of the micro's conviction, but out of imposition from whoever gives it work.

The real digital gap

On the exact penetration of digital batch-by-batch traceability in Portuguese industrial SMEs there is not yet solid and public sector data. What is observed on the ground is clear: many units have an ERP for the financial and invoicing part — required by the AT — but keep production and traceability in Excel, notebooks and the tacit knowledge of veteran operators. The ERP exists. The shop-floor traceability, that one, still lives on paper. It's this disconnection that PT2030 and PRR funds aim to close.

The European digitalisation indicators — namely the DESI index, which the European Commission monitors — place Portugal in an intermediate position in business digital adoption, with a strong contrast between large companies and SMEs. Financial digitalisation is resolved by legal obligation; operational digitalisation is not. It's exactly the space between these two that traceability occupies.

4. The traceability implementation models

There are four practical approaches. Each has a cost, a level of rigour and a company profile it serves. There is no best in absolute terms — there is the right one for the company's moment and size.

Comparison of the four approaches

Paper and Excel: why they're not enough

Paper and Excel share the same structural flaw: they are isolated records, dependent on human discipline, with no automatic link to reality. Someone has to note it down. Someone has to note it down correctly. And no one can cross-reference the dyeing plant's notebook with the purchasing sheet and the dispatch note without hours of manual work.

Excel is worse than it seems, precisely because it seems good. It gives the illusion of a system — it has columns, filters, formulas. But each operator has their own version, the formula breaks when someone inserts a row, and the history of who changed what does not exist. In a serious audit, an Excel file is not proof of traceability — it's proof of good intention.

There is also the problem of access concurrency. Two people don't edit the same file simultaneously without conflict, and the version that remains is always the last one saved — not the most correct one. In a factory with three shifts, the traceability file becomes a Tower of Babel of versions: "production_final_v3_JMS_corrected.xlsx". Anyone who has ever managed a quality audit on a file like that knows that the honest answer to "is this traceable?" is: not with confidence.

Excel doesn't fail because it's bad. It fails because it seems sufficient — and that illusion costs more than the absence of any system, because it delays the decision to digitalise seriously.

Generalist ERP vs vertical ERP

A traceability module in a generalist ERP solves the simple case: one item, one batch, one supplier. Where it breaks is in vertical complexity — the three axes of footwear, the bath recipes of the dyeing plant, the component genealogy of a mould. There, the generalist ERP forces expensive customisations that become technical debt with every update.

The technical debt deserves explanation, because it's the hidden cost that only appears years later. When you customise a generalist ERP to model the three axes of footwear, you create bespoke code that is not part of the standard product. With every supplier update, that customisation has to be rewritten, tested, revalidated. After a few years, the company is stuck: it can't update without breaking the customisations, and it can't stop updating without losing support and security. It's the classic trap of forcibly fitting a vertical reality into a horizontal product.

The vertical industrial ERP, like the MULTI ERP, already brings the product structure of the origin sector. It doesn't need to contort the data model to fit reality — the model was already designed for that reality. And when you connect the real-time production capture of KORA Productivity, internal traceability stops depending on the notebook: the record is made at the terminal, at the moment, by the operator who is right there.

The traceability that depends on someone remembering to write in the notebook at the end of the shift is not traceability — it's an optimistic reconstitution.

The role of MES and automatic capture

Where volume and complexity justify it, traceability capture can be automated through an MES and barcode or RFID identification. The operator reads the input batch, the system records it, and the genealogy is built without manual typing. This reduces human error — the operator doesn't need to write the right number, only to read the label — and reduces capture time to seconds.

It's not for everyone. A garment maker of 20 people doesn't need RFID. But a footwear factory with 150 employees and 1,000 SKUs per collection, where each pair goes through dozens of operations, gains a lot from automatic capture at the critical points. The rule is simple: automate where the transaction volume makes manual capture a bottleneck, keep it manual where the volume doesn't justify the investment in terminals and readers.

5. How to assess whether your company needs it

Not every company needs the same level of traceability. An automotive component manufacturer and a small subcontract garment maker have radically different risk exposures. The diagnosis begins by measuring that exposure.

Signs that paper is no longer enough

• You receive traceability requests from client-brands and the answer takes more than half a day.
• You have already had a recall or complaint in which you could not delimit the affected batch and had to scrap in excess.
• Compliance with the brands' sustainability requirements is becoming a condition of supply.
• The knowledge of "where what is" lives in the heads of two or three veteran operators near retirement.
• Cross-referencing purchasing, production and dispatch to investigate a problem consumes days of manual work.

If you recognise three or more of these signs, the cost of not digitalising has almost certainly already exceeded the cost of digitalising. The fifth sign — the knowledge in the heads of veterans near retirement — is the most underestimated and the most dangerous. It's a demographic time bomb that most factories in the North have armed and ignore.

Step-by-step diagnosis

1. Map the batch entry points. List where raw material with a supplier batch number enters: receipt of yarn, dyes, components, packaging. These are the upstream traceability points.
2. Identify the transformation points. Where the input batch transforms into something else and gains a new internal batch: dyeing, cutting, sewing, injection, assembly. Each transformation is a potential break in traceability.
3. Map the exit points. Dispatch notes, invoices, packing lists. Link each dispatch to the internal batches that make it up. This is downstream traceability.
4. Measure the real reconstitution time. Choose an item dispatched three months ago and time how long it takes to reconstitute the complete journey with the current system. That number is your honest starting point.
5. Quantify the risk exposure. Estimate the cost of a poorly delimited recall in your sector — quantity to scrap, reprocessing cost, contractual penalty from the brand, reputational damage. Compare it with the cost of digitalising.

This exercise, done seriously, tends to reveal the true cost of paper — which is not the paper, it's the lost time and the undelimited risk. It's the same mechanism we describe in when the ERP reveals what management didn't see.

The decision-making trio's conversation

In Portuguese family businesses, the decision to digitalise traceability almost always goes through the same trio: the CEO — often the founder or second generation —, the CFO who guards the treasury, and the IT manager. The latter is frequently a self-taught person with 15 years of business knowledge and no formal degree, who knows where all the skeletons of the current system are and whose opinion carries more weight than that of any external consultant.

Ignoring any vertex of this trio kills the project. The CEO needs to see the risk exposure in business terms — lost customers, contracts at stake. The CFO needs to see the full TCO and the return in avoided scrap. The IT manager needs to be involved as a partner, not as the executor of a decision already made — because they are the one who will live with the system every day, and their silent resistance is fatal.

6. What to choose and why, by company size

The recommendation changes with size and product complexity. There is no single answer — there is the right answer for your profile.

Decision matrix

Micro and small enterprise: when Excel still serves

Let's be honest: a micro garment maker of eight people, low risk, simple product, doesn't need a vertical ERP tomorrow morning. Structured Excel, with discipline and a single owner, can serve as long as the risk is low. Selling it a complete industrial system would be commercial bad faith.

But there are two conditions. First: that the risk really is low — no rapid-recall demands, no client-brands imposing digital traceability. Second: that there is a migration plan for when the risk rises. And the risk always rises, because the pressure descends from the value chain. The micro that today lives on Excel should know, from now, what the next step is — so as not to be caught by surprise when the client-brand demands, with a three-month deadline, that which takes a year to implement well.

Industrial SME: the most frequent decision point

Most Portuguese textile and footwear factories fall in the 50-200 employee band with complex variants. For this profile, the cost of not digitalising has already exceeded the cost of digitalising — but the TCO has to count the implementation, the training and the shop-floor resistance, not just the software licence. A traceability project that ignores the human cost of change fails, regardless of the technical quality of the tool.

In this segment, funding frequently makes the difference between moving forward and postponing. The instruments of PT2030, PRR, COMPETE 2030 and Norte 2030 support industrial digitalisation — but experience teaches that approval depends less on the idea and more on the quality of the technical report. Good applications get stuck in weak descriptive memos. A partner who has already been through dozens of these processes saves months of back-and-forth with the intermediary body.

The classic mistake: buying by module, not by journey

The mistake we have seen repeat itself for 36 years: the company buys "the traceability module" as if it were an isolated box. But traceability is not a module — it's a property that runs across purchasing, production, stock and dispatch. If the module doesn't talk to the rest of the ERP, you've bought a more expensive digital notebook. The right question is not "does it have a traceability module?" — it's "can I, starting from a dispatch note, reach the supplier batch and the machine that processed it in three clicks?".

Traceability is not a module you buy. It's a property that emerges when purchasing, production, stock and dispatch share the same batch in the same system.

Medium and large size: complexity and branches

For companies with multiple units and high process complexity, QAD Adaptive ERP allows sophisticated traceability processes to be modelled without hard-coding, and Multi Connect ensures that traceability is not lost at the boundaries between branches and partners — the point where most systems break.

The multiplicity of branches introduces a specific problem: the same batch can cross two or three units of the same company before leaving for the customer. If each unit has its own isolated system, the genealogy breaks at the internal boundary. Integration between branches is not a large-company luxury — it's the condition for traceability to survive the corporate structure. A company with spinning in one unit and dyeing in another has to ensure that the yarn batch produced in the first arrives identified at the second, without manual re-labelling that introduces error.

7. Regulatory framework and applicable compliance

Traceability crosses several legal regimes in Portugal. Ignoring any one of them creates real exposure.

Invoicing, SAF-T and document integrity

DL 28/2019 and Ordinance 195/2020 mandate electronic invoicing with ATCUD, AT-certified software and monthly reporting of the SAF-T. This is not product traceability, but it crosses it: the dispatch note and the invoice are the link point between the produced batch and the customer who received it. A system that treats invoicing and production as separate islands breaks downstream traceability exactly where it matters most legally.

The link is deeper than it seems. When the transport note is reported to the AT with the dispatched items, and those items are linked to the internal batches in the same system, downstream traceability is automatically sealed in a document with legal value. Separating production and invoicing into different systems requires manually reconciling two worlds — and it's in that manual reconciliation that error enters and traceability degrades.

Food and product safety

For food distribution and retail, Regulation (EC) No 178/2002 imposes traceability "one step back, one step forward" throughout the chain. The capacity for a rapid recall is not best practice — it's a legal obligation with ASAE enforcement. For food retail with integrated POS and back office, this means the batch has to survive to the point of sale, linking the consumer's purchase to the supplier batch when the recall demands it.

GDPR, NIS2 and the security of traceability data

Traceability data often includes operator data — who processed it, on which shift. GDPR and Law 58/2019 apply. And with the transposition of the NIS2 Directive (Directive (EU) 2022/2555, transposed by DL 65/2025) into the national legal order, companies in sectors considered critical will have reinforced cybersecurity obligations — including the protection of the integrity of the systems where traceability lives. A recall investigated on tampered data is worth nothing. Data integrity is as important as its existence.

ENISA and the European cybersecurity framework stress a point that industry tends to underestimate: digital traceability only protects if the data is trustworthy. A system where any user can retroactively edit a batch record offers no audit guarantee. The audit trail — who changed, when, what — is part of traceability, not an extra. Certifications such as ISO 27001 formalise these integrity controls.

Sustainability and the Digital Product Passport

The European ecodesign regulation for sustainable products (ESPR) paves the way for the Digital Product Passport, which will require verifiable information on origin, composition and environmental footprint — starting with textiles. The factory without digital traceability will not be able to generate this passport. Future compliance is built today, in the data model.

The following table summarises the regimes and what each one requires of the traceability system:

8. How INFOS approaches this

INFOS has been building vertical software for industry for more than three decades, and traceability is not, for us, a stand-alone feature — it's a natural consequence of having the data model designed for textiles, footwear, metal and plastic from the origin. The MULTI ERP treats the batch as a first-class citizen: it enters at purchase receipt, crosses production with the variant structure specific to each sector, and leaves at dispatch linked to the note and the invoice.

On the shop floor, KORA Productivity captures production at the moment and at the terminal, which closes the internal traceability hole without depending on anyone remembering to write in the notebook. And when management wants to see the complete picture — which suppliers generate the most defects, which batches have the most reprocessing — Qlik Sense transforms traceability data into actionable KPIs instead of dead archive.

In the warehouse, the KORA Inventory Suite keeps the batch alive at picking and dispatch, ensuring that downstream traceability is not lost in the last metre — the metre where the product goes out the door and stops being traceable if no one captured the batch on the note. This is where many systems fail: they have traceability in production, but lose it at dispatch, precisely where the client-brand is going to ask.

We don't sell traceability as an isolated project. We sell the link — because it's the link that was missing, and the link is where real traceability lives. The hard part is not technical; it's the change management on the shop floor, and that's where vertical experience decides success.

In 35 years, we have never seen a traceability project fail for technical reasons. We have seen it fail when the shop floor was not heard and the veteran operator sabotaged, in silence, what no one had explained to them.

What we learned from projects that went wrong

We would be dishonest if we painted a picture of uninterrupted success. We learned, in projects that stumbled, that traceability capture imposed on the operator without showing them the benefit generates poor-quality data — they read the wrong label on purpose, or skip the step when the supervisor isn't looking. We learned that starting with the whole factory instead of a pilot flow multiplies the failure points and kills trust in the very first week. And we learned that underestimating the training time is the most common budget mistake — the licence is the cheapest thing; the change of habit is the most expensive.

These lessons are today embedded in the way we invoice and plan a traceability project. It's not generosity — it's the scar of those who have already paid the price of ignoring them.

9. 30/60/90-day roadmap

Digitalising traceability is not a big bang. It's a sequence of verifiable milestones. This is the journey that works in a Portuguese industrial SME.

Days 1-30: map and measure

• Complete the diagnosis in section 5 — batch entry, transformation and exit points.
• Time the real reconstitution time of a historical case. Record it as a baseline.
• Identify the two or three points where traceability breaks today (almost always the internal transformation).
• Gather the decision-making trio — CEO, CFO and IT manager — and align the risk exposure in euros, not in the abstract.

Days 31-60: pilot in one flow

• Choose a single product flow — not the whole factory. One item, one complete journey, from raw material to dispatch.
• Configure the batch capture in that flow, with a terminal on the shop floor for the internal record.
• Involve the veteran operators as owners of the pilot, not as guinea pigs. Their tacit knowledge is the asset, not the obstacle.
• Measure the reconstitution time again in the pilot flow. Compare it with the baseline.

Days 61-90: consolidate and extend

• Fix what the pilot revealed — it always reveals something the paper map didn't show.
• Extend the capture to the higher-risk or higher-volume flows, in order of exposure.
• Link traceability to dispatch and invoicing, closing downstream traceability end-to-end.
• Build the first traceability dashboard — defects by supplier, reprocessing by batch — to transform compliance data into real management by objectives.

After the 90 days: from compliance to advantage

The mistake would be to stop at 90 days with traceability as a ticked compliance box. Traceability data, once collected, is worth much more than answering audits. It tells you which supplier systematically generates the most reprocessing — real negotiation information. It tells you which machine or which shift concentrates defects — maintenance and OEE information. It tells you which items have their margin eroded by hidden scrap.

This is the turn that separates those who digitalise out of obligation from those who digitalise out of advantage: the former see traceability as a cost; the latter discover that the same data that satisfies the client-brand also exposes the points where the factory loses money without knowing it.

At the end of these 90 days, the client-brand's question that arrived on a Friday late afternoon stops costing an afternoon. It costs three clicks. And the dyeing plant's black-covered notebook can go back to being what it always should have been: a sentimental keepsake, not the only proof that the factory knows what it produced.

Sources

• INE — Statistics Portugal, commerce statistics (2024 data): turnover, number of companies and workers, trade margin in wholesale trade.
• INE and IAPMEI — dimensional structure of the Portuguese business fabric (pred

Frequently asked questions

What distinguishes real traceability from a simple paper archive?

Real traceability requires a link between raw material, production, stock and dispatch in the same system. A paper notebook is merely an archive — it doesn't cross-reference information or answer concrete questions in useful time. True traceability allows you to identify, within hours, which raw material batch was used, which machine processed it and where it was dispatched.

Why does a colour fastness problem in a dyeing plant become so costly?

Without batch-by-batch traceability, the factory cannot isolate whether the defect came from the greige yarn, the bath recipe or the dye. When in doubt, it scraps everything that might be affected. A problem that would warrant 200 kilos of scrap becomes two tonnes, because the exact boundary of the compromised batch cannot be traced.

What is the impact of a slow response to a quality complaint?

International brands work in tight cycles. When the answer takes three days, the brand no longer investigates the cause — it questions whether it's worth renewing the subcontract. A poorly resolved complaint during a buyers' visit season contaminates trust in the next collection, affecting future commercial decisions.

How does traceability work in a subcontract garment maker?

The garment-making factory receives fabric and trims from the parent brand and needs to record which roll of fabric — with a batch number — entered which manufacturing order and left in which dispatch box. The challenge is aggravated when this knowledge lives in the head of the line supervisor and is not digitalised, creating risk when that person is absent or retires.

Why does a generalist ERP fail in footwear traceability?

A footwear collection has 800 to 1,200 distinct SKUs, crossing colour, size and last. A generalist ERP treats these variants as loose and traceability breaks at the point where it matters most. The sector needs a vertical product structure that simultaneously models the three axes.

What does compliance with the European strategy for sustainable textiles mean?

The brand demands to know the origin of the fibre, the footprint of the process and the complete journey of the product. This requirement has stopped being a competitive differentiator and has become an entry ticket — without compliance, there is no access to certain value chains. A paper notebook cannot answer these demands.

What is the risk of losing traceability when an employee retires?

In garment making, the knowledge of "which roll went to which order" often lives in the memory of the line supervisor. When that person retires or is absent, the traceability leaves through the door. Digitalisation is not about technology — it's about transferring tacit knowledge to a system before it disappears.