← Insights

Textiles and clothing

Textiles and clothing: where margin leaks and how to recover it

One in three companies does not cover its operating costs. Margin leaks through costings, samples, orders and data requests that are still handled by hand.

Published 17 November 2026Updated 17 November 202620 min read

Before you read

Skip to the report ↓

Do you know your sector?

Question 1 of 5

In which size of garment maker is Portugal furthest from the European average in value created per person?

Companies with negative EBITDA in 2025[1]
34.0%
Labour costs as % of clothing GVA, Portugal vs EU[2]
65.1% vs 51.1%
EBITDA margin gap between large firms and SMEs (2025)[1]
6.3 pp
Companies sending structured invoices, Portugal vs EU (textiles, clothing and footwear; 2023 survey)[10]
17.2% vs 48.0%

Executive summary

  1. Portugal holds 9.8% of EU textile and clothing employment, against 2.5% of EU manufacturing employment [2].
  2. 34.0% of companies in the sector had negative EBITDA in 2025, and the margin gap between large firms and SMEs widened from 4.1 pp to 6.3 pp since 2019 [1].
  3. In clothing, labour costs absorb 65.1% of GVA, against 51.1% in the EU; GVA per hour is 50% of the European level [2].
  4. In the textiles, clothing and footwear aggregate, only 17.2% of companies sent structured invoices (EU: 48.0%) and 4.1% used CRM (EU: 17.6%) [10],[8].
  5. The destruction ban already applies to large companies, the environmental-claims directive has applied since 27/09/2026 and the digital passport is planned for adoption in Q4 2027 [40],[41],[39].
  6. In Vektrum's base scenarios, a 50-person garment maker gains €29,340 a year (+1.06 pp of margin) from costing, samples and subcontractors; the assumptions are shown.

01

Conclusions

Executive summary

In 2024, the textile and clothing industry (CAE 13 and 14, footwear excluded) had 11,965 enterprises, 118,469 people and €8,211m of turnover [2]. This report analyses the sector with data from Banco de Portugal, Eurostat and INE (Statistics Portugal), and turns the figures into concrete decisions for a textile or clothing company with 30 to 60 people.

The starting point is clothing. Labour costs absorb 65.1% of gross value added, against 51.1% in the EU, and the gross operating surplus keeps 18.3%, against 34.6% [2]. Staff cost per hour is 63% of the European level, but value added per hour is only 50%, and the minimum wage is 12.2% above its 2024 level [21],[23]. Vektrum reading: cheap labour is no longer enough to defend margin; what remains is to recover the hours that add no value to the product, such as costings redone, samples that never convert, orders retyped, subcontractors chased by phone and data requests from brands.

Share of the sector's sales made abroad (2025); 24.0% at micro enterprises[1]
60.6%
People employed in clothing between 2019 and 2025 (82,677 → 68,732); textiles −5.1%[1]
−16.9%
Clothing employment in firms of 50 to 249 people (2024), against 24.1% in the EU[3]
41.8%
Clothing workers aged 50 or over in 2025, against 32.4% in 2019[6]
48.3%

Seven conclusions for management

  1. Portugal weighs four times more in European textiles than in European manufacturing. It has 9.8% of the EU's textile and clothing employment and 6.0% of its companies (2024) [2]. Exports totalled €5,499m in 2025 (−0.8%), about 6.9% of goods exports; Spain bought 24% and the top three markets (Spain, France and Germany) about 47.5% [14],[16].
  2. Large firms are pulling away. The EBITDA margin of large firms rose from 10.0% to 13.2% of income between 2019 and 2025; that of SMEs only from 6.0% to 7.0%. The gap widened from 4.1 pp to 6.3 pp (Vektrum calculation) [1].
  3. One in three companies does not cover its operating costs. In 2025, 34.0% of companies in the sector had negative EBITDA; in clothing, 36.2%; among micro enterprises, 38.6%; among large firms, 5.6% [1].
  4. Not even the largest garment makers escape the gap. A clothing company with 250 or more people generates €32.9k of GVA per person, 48% of the European level, a wider gap than among medium-sized firms (73%), in 2024 [3]. Size alone does not solve the problem.
  5. Capital is locked in customers, above all at small firms. The sector collects in 67 days; micro enterprises in 88 and large firms in 51. Financing costs rose 86% between 2022 and 2024, and financial pressure went from 8.4% to 14.2% of EBITDA between 2022 and 2025 [1].
  6. There is ERP, but invoicing and decisions run on paper. In the combined textiles, clothing and footwear aggregate, ERP use was above the EU in 2023 (45.6% against 38.0%; by 2025 the EU had reached 48.7% and the Portuguese figure is not published), but only 17.2% of companies sent structured invoices, against 48.0% in the EU, and 81.0% still sent paper [8],[10]. In a CITEVE survey of 60 companies, about 90% did not collect data automatically [44].
  7. Process gains are worth margin points, not promises. In Vektrum's base scenarios, a 50-person private-label garment maker gains €29,340 a year from costing, samples and subcontractor follow-up (+1.06 pp of EBITDA margin); a 30-person textile company gains €12,533 from orders, invoices and data requests (+0.45 pp). These are estimates, with the assumptions shown in chapter 6.

02

Structure

The sector: six business models

The two activity codes have different profiles, and Portuguese firms are larger than European ones: 12 people per firm in textiles (EU: 8) and 9 in clothing (EU: 5) [2]. In Banco de Portugal's universe of corporations, SMEs account for 99.4% of firms, 84.8% of people and 76.9% of sales, and 36 large corporations generate 23.1% of sales (2025) [1].

Talking about "textiles" as one block hides business models with very different operational problems. For this report we distinguish six.

Textiles and clothing in figures (2024)
IndicatorTextiles (C13)Clothing (C14)Total
Enterprises3,6758,29011,965
Persons employed45,83772,632118,469
Turnover (€m)4,212.73,998.28,210.9
GVA (€m)1,342.81,447.92,790.7
People per firm (EU)12 (8)9 (5)–
Share of manufacturing employment (EU)6.11% (1.71%)9.68% (2.31%)–
GVA per person (€k) and % of EU29.3 (61%)19.9 (59%)–

Eurostat SBS. Totals and ratios to the EU are Vektrum calculations. Eurostat's universe includes sole proprietors; Banco de Portugal's covers corporations only (5,707 in 2025). Sources: [2],[1]

  • Spinning and weaving. Sells yarn and fabric to garment makers and brands. Longer runs, capital-intensive and long collection periods (textile firms collect in 75 days) [1].
  • Knitwear. Knitting and knitted garments, the sector's largest export category. Short cycles, many samples and orders with colour and size grids.
  • Finishing, dyeing and printing. An industrial service for third parties, intensive in energy and water. Each batch is priced by process and recipe, and one reprocessing run can consume the batch margin.
  • Technical textiles. Products with technical specifications, testing and certification, and growing: Garantia Mútua reports an expansion of the segment in Covilhã [18].
  • Home textiles. Collections and catalogues for retailers and brands, often private label, with orders from large customers.
  • Garment making: private label vs own brand. A maker working for international brands receives tech packs, prices them, makes samples and coordinates subcontractors for the sewing. An own-brand maker also carries stock, returns, environmental claims and extended producer responsibility.
Subsegments: who they work for and where the bottleneck is
SubsegmentTypical customersTypical bottleneckExports Jan–Sep 2026
Spinning and weavingGarment makers and brandsPlanning, energy, collectionsSynthetic fibres: −25.1%
KnitwearInternational brandsSamples, colour × size gridKnitted clothing: €1.62bn (+1.1%)
Finishing, dyeing and printingSpinners, weavers and garment makersProcess pricing, energy, reprocessingNo separate series
Technical textilesCustomers with technical specificationsTesting, certification, documentationCoated and technical fabrics: €294.4m (+14.1%)
Home textilesRetailers and brandsCatalogue, collection samples€578.1m (+2.3%)
Private-label garment makingInternational brandsCosting, samples, subcontractorsWoven clothing: €783.7m (+9.5%)
Own-brand garment makingRetail and end consumersStock, environmental claims, EPRNo separate series

Vektrum synthesis. Exports are the categories published by ATP on INE data and do not map exactly to the subsegments; woven and knitted clothing also include own-brand production. Sources: [15],[18]

Exports: value recovers, volume does not

Exports, Jan–Sep 2026: year-on-year change by product (%)
Coated fabrics and technical textiles14.1%Woven clothing9.5%Total textiles and textile articles2.5%Home textiles2.3%Knitted clothing1.1%
Exports, Jan–Sep 2026: year-on-year change by product (%)
Coated fabrics and technical textiles14.1%
Woven clothing9.5%
Total textiles and textile articles2.5%
Home textiles2.3%
Knitted clothing1.1%

Portugal Têxtil on INE data; total rate: Vektrum calculation (€103.2m on €4,106.8m). Synthetic fibres (−25.1%) are not shown. Sources: [15]

Between January and September 2026, exports of textiles and textile articles totalled €4.21bn, €103.2m more than in the same period of 2025: Spain €991.2m (+1.1%), France €618.4m (+4.6%), Germany €338.7m (−2.8%), the US €328.1m (+3.5%) and the Netherlands €252.3m (+9.6%) [15]. Output volume keeps falling: in textiles, −7.6% year on year in September; in clothing, −9.3% in August [5]. The textile production index stood at 84.5 in 2025 (2021 = 100) [4]. Vektrum reading: value rising while volume falls is consistent with higher prices or a richer mix and fewer pieces, but it has not been measured.

In 2025, Spain bought €1,317.8m (about 24.0%), France €834m and Germany €461m; the US €420.4m (−3.3%, about 7.6%), Italy fell 13% and Morocco rose 22.1% [14],[16]. Foreign markets account for 60.6% of the sector's sales: 66.4% in clothing and 73.6% at large firms, but only 24.0% at micro enterprises [1]. For the US, the August 2025 commitment is to apply the higher of the most-favoured-nation (MFN) rate and 15% to European products [26]; in clothing, where MFN rates reach 32% according to ATP [27], the 15% floor may add nothing. Among fashion executives surveyed by McKinsey, 76% believe tariffs and trade disruption will shape 2026, and McKinsey attributes a 35% short-term rise in clothing sourcing costs to the 2025 tariffs [28].

Proximity: the advantage, and the dependency

Inditex's supply chain had 6,684 factories in 49 markets in 2025, and the company produces a significant share in proximity markets: Spain, Portugal, Morocco and Turkey [25]. ATP's president sums up the risk: "There are groups built on Inditex. That is a Portuguese problem today." [20]. Proximity has value because the fashion cycle is capped at about six weeks to market, against six months in the past (McKinsey, 2018) [29], and because 71% of clothing chief procurement officers surveyed in 2021 planned to increase nearshoring, with Turkey entering the top three sourcing origins for Europe [30].

American nearshoring, however, is happening in Central America: in 2026, 76% of the US fashion companies surveyed bought from CAFTA-DR countries, against 64% in 2025 (n=30) [31]. Portugal's proximity advantage is a European one.

The structure also differs from Europe's. In clothing, 41.8% of employment is in firms with 50 to 249 people, against 24.1% in the EU; in textiles, 37.4% against 32.2% [3]. Portuguese garment making is an industry of medium-sized factories, concentrated in the North: Braga and Porto held 85.4% of clothing employment in 2020 (49.1% and 36.3%), the latest year published by district [17].

03

Economics

Margins, costs and capital

The EBITDA margin Banco de Portugal publishes for the sector is a percentage of income, not of sales [1]. When we divide EBITDA by sales, we say so. The aggregates are weighted by size and are meant to read trends, not to benchmark a specific company.

EBITDA margin as % of income, 2019–2025
Aggregate2019202020212022202320242025
Textiles and clothing (total)6.88.08.48.99.18.78.5
Textiles (CAE 13)8.110.59.69.89.710.010.0
Clothing (CAE 14)5.65.57.18.08.47.56.9
Micro enterprises5.65.97.96.33.75.46.0
Small enterprises1.44.43.47.25.96.05.9
Medium-sized enterprises8.29.010.19.59.48.17.7
Large enterprises10.011.210.510.212.813.313.2
SMEs (total)6.07.27.98.67.97.27.0

Banco de Portugal, Central Balance Sheet Database, textiles and clothing aggregate (CAE 13 and 14, non-financial corporations, footwear excluded). Sources: [1]

As EBITDA over sales, textiles reached 10.6% and clothing 7.1% in 2025; the sector as a whole 8.8%, below the 10.3% of manufacturing (Vektrum calculations) [1],[52]. Between 2022 and 2025, sales fell 7.6% and EBITDA 12.6% [1]. Net margin was 2.8% of income (textiles 3.2%, clothing 2.4%), and return on equity fell from 9.54% in 2022 to 5.30% in 2025.

Companies with negative EBITDA (2025; 29.9% in 2022, peak of 37.5% in 2023)[1]
34.0%
Companies with negative equity (2025); 26.2% in clothing[1]
23.9%
Financing costs as % of EBITDA (2025; 8.4% in 2022; 17.2% at micro firms)[1]
14.2%
Equity-to-assets ratio of the sector (2025); 26.5% at micro enterprises[1]
49.6%

Where each €100 of garment-making turnover goes

Clothing (C14), 2024: from €100 of turnover to gross operating surplus
Turnover100 €Purchases of goods and services (incl.subcontracting)−65.4 €Staff costs−29.6 €Change in stocks and other items(residual)+1.6 €Gross operating surplus6.6 €
Clothing (C14), 2024: from €100 of turnover to gross operating surplus
Turnover100 €
Purchases of goods and services (incl. subcontracting)−65.4 €
Staff costs−29.6 €
Change in stocks and other items (residual)+1.6 €
Gross operating surplus6.6 €

Vektrum calculation on Eurostat SBS: purchases and staff costs ÷ turnover; the end value is the published gross operating rate (gross operating surplus ÷ turnover). The residual groups changes in stocks, capitalised production, other income and taxes and subsidies. Sources: [2]

Productivity: the problem is not only being small

Clothing: Portuguese GVA per person as % of the EU, by size class (2024)
0–9 people65%10–19 people66%20–49 people70%50–249 people73%250+ people48%
Clothing: Portuguese GVA per person as % of the EU, by size class (2024)
0–9 people65%
10–19 people66%
20–49 people70%
50–249 people73%
250+ people48%

Vektrum calculation on Eurostat sbs_sc_ovw (GVA per person employed, Portugal ÷ EU27). Sources: [3]

Unlike textiles, where firms with 250 or more people are the closest to the EU (61%), the largest Portuguese garment makers are the furthest away: €32.9k of GVA per person, against €68.6k in the EU (48%) [3]. Within Portugal, the same large firm generates 2.8 times more than a micro one. Wage-adjusted productivity (GVA ÷ staff costs) fell between 2021 and 2024 from 152.3% to 134.2% in textiles and from 120.1% to 113.9% in clothing, against 132.7% and 131.1% in the EU [2]. In textiles, each euro of wages still generates as much value as in the EU; in clothing, it now generates less.

Labour: wages rising, workforce ageing

Labour cost, productivity and age: Portugal vs EU
IndicatorTextiles PTTextiles EUClothing PTClothing EU
Staff cost per hour worked (€, 2024)12.4322.2610.0516.06
Labour cost per FTE (€k, 2024)21.9139.2317.5428.99
GVA per hour worked (€, 2024)17.4432.6312.2924.56
Labour costs as % of GVA (2024)55.853.465.151.1
Workers aged 50 or over (%, 2025)41.043.748.346.2

Eurostat SBS and labour force survey (lfsa_egan22d). Age shares are Vektrum calculations; Portuguese samples by activity are small, so treat them as an order of magnitude. FTE = full-time equivalent. Sources: [2],[6]

The minimum wage went from €820 in 2024 to €870 in 2025 and €920 in 2026, up 12.2% in two years [21],[22],[23]; the tripartite agreement points to €1,020 in 2028 [24]. In clothing, average base pay was about 1.2 times the minimum in 2020 (€764.29 with the minimum at €635) [17], so each rise in the minimum reaches much of the pay scale. ATP's president speaks of increases of "an average of 7% a year over the last five years" [20]; in official data, the hourly cost in clothing rose about 21% between 2021 and 2024, about 6.5% a year (Vektrum calculation) [2].

The age pyramid flipped in six years. In clothing, workers aged 50 or over went from 32.4% (2019) to 48.3% (2025), and those under 40 from 30.4% to 22.4% [6]. Vektrum reading: know-how on costing, pattern making and subcontractor relationships is concentrated in experienced people, often with no written record. Capturing it is a continuity issue before it is an efficiency one.

Working capital, interest and energy

Average collection and payment periods (days, 2025)
AggregateCollectionPayment
Textiles and clothing (70 days to collect in 2022)6773
Textiles (CAE 13)7577
Clothing (CAE 14)5869
Micro enterprises8899
Small enterprises7081
Medium-sized enterprises6973
Large enterprises5159

Banco de Portugal. The sector series extracted do not publish inventories by aggregate, so there are no stock-days figures for CAE 13 and 14. Sources: [1]

The sector finances its customers more than it is financed by its suppliers: net trade-credit financing is −4.6% of sales, and −8.1% at micro enterprises [1]. Inditex reports paying 72% of supplier payment volume within 30 days in 2025, against contractual terms of 60 [25]. Vektrum reading: the delays are likely to sit mainly in other links of the chain, such as subcontracting between Portuguese firms, which has not been measured. Financing costs went from €69.3m in 2022 to €129.2m in 2024 and €103.2m in 2025 [1].

Energy weighs mostly on finishing, dyeing and spinning. Textile energy purchases rose 57% from 2021 to 2022 (€188.3m → €295.5m) and reached 8.8% of purchases; in 2024 they were back to €165.7m (5.6%) [2]. Industrial gas prices multiplied by 4.1 between 2021 and 2022, but in 2026-S1 they were about 17% below the EU average, and non-household electricity cost about 71% of the European average [13],[12]. In Europe, 21% of textile companies surveyed by EURATEX say energy exceeds 20% of production costs, and more at small firms [19]. At a dye house, it is a cost that must enter the price of each batch, not an annual average.

04

Diagnosis

Where time and margin are lost

There are no official Portuguese data on costing time, sample rounds and their conversion into orders, subcontractor planning or defect rates in the sector. Vektrum reading: these costs are spread across the daily work of several people, and so are rarely measured. What follows combines indirect official indicators, studies with their scope stated and Vektrum estimates identified as such.

Textiles, clothing and footwear: Portugal vs EU (companies with 10+ people, %)
ERP, Portugal (2023)45.6%ERP, EU (2023)38%Sends structured invoices, Portugal17.2%Sends structured invoices, EU48%Electronic supply-chain sharing,Portugal11.4%Electronic supply-chain sharing, EU14.6%CRM, Portugal (2025)4.1%CRM, EU (2025)17.6%
Textiles, clothing and footwear: Portugal vs EU (companies with 10+ people, %)
ERP, Portugal (2023)45.6%
ERP, EU (2023)38%
Sends structured invoices, Portugal17.2%
Sends structured invoices, EU48%
Electronic supply-chain sharing, Portugal11.4%
Electronic supply-chain sharing, EU14.6%
CRM, Portugal (2025)4.1%
CRM, EU (2025)17.6%

Eurostat isoc_eb_iipn2 and isoc_eb_icsn2, C13–C15 aggregate, which includes footwear. Invoicing and sharing: 2023 survey, covering 2022. In 2025, EU ERP use reached 48.7%; the Portuguese figure is not published. Sources: [8],[10]

The pattern matches manufacturing in general: the core system exists, the link with customers and suppliers does not. In the textiles, clothing and footwear aggregate, 81.0% of companies with 10 or more people still sent paper invoices, against 52.8% in the EU [10]. In a CITEVE survey of 60 companies, more than 70% had no formal digital strategy, about 90% did not collect data automatically and 88% did not base decisions on data [44]. The sample is small and not representative, but it is consistent with the official data.

Eight loss points

  1. Development and samples. A 3D sampling vendor estimates that 40% of samples never move into production [47]; it is a vendor figure without methodology, useful only as an order of magnitude. Without conversion records by customer and style, a garment maker keeps developing samples for customers who never order; estimated volumes and costs are in chapter 6.
  2. Costing and quoting. Purchases of goods and services, including subcontracting, account for 65.4% of clothing turnover and 70.3% of textiles [2]. A 1% error in the cost of materials and make equals about 9% of a garment maker's EBITDA and 7% of a textile company's (Vektrum calculation on Eurostat and BdP) [1]. A full costing combines tech pack, consumption, sewing minutes, make, finishing and margin.
  1. Order intake. Only 11.4% of companies in the aggregate share supply-chain information electronically with customers or suppliers (EU: 14.6%) [10]. Orders arrive by email, PDF or brand portal and are retyped into the ERP. Each order has 20 to 200 lines (style × colour × size) and takes 15 to 40 minutes to enter and check (Vektrum estimates). An error in the grid produces garments nobody ordered.
Vektrum suggestionThe brand's order reaches the ERP without being retyped

Starting point

Orders arrive by email, as PDF or through a portal, each brand with its own format. Someone copies them line by line, style × colour × size, into the ERP the company already has.

What gets built

  • Reading of the order in the format the brand already sends, without asking the brand for anything
  • A draft order in the ERP, with the colour and size grid filled in
  • An alert when the grid does not match the style or previous orders

What stays with the team

Checking the draft and confirming it. Nothing goes into production without that validation.

How to know if it is worth it

How many orders arrive each month and how long each one takes to enter. The chapter 6 scenario works with 15 to 40 minutes per order and excludes the grid errors avoided. See the calculated impact →

Impact 3/3 · Effort 2/3 · 30 to 90 days

First step: gather ten recent orders from three different brands.

Or we can run that review with you →
  1. Subcontractor planning. Clothing spent only €0.22m on agency work in 2024 [2]: capacity flexibility comes from subcontracting the sewing, not from agencies. A garment maker coordinating 5 to 30 subcontractors spends 10 to 25 hours a week planning and chasing work in progress by phone, WhatsApp, visits and Excel (Vektrum estimates). Vektrum reading: subcontractors tend to be micro and small firms, segments that collect in 88 and 70 days [1].
  2. Quality and claims. At Impetus, in Barcelos, AI inspection on 5 of 20 looms cut final defects by 8% in that part of production [48]; these are vendor figures reported by the press, and they concern on-loom inspection, not claims analysis. We estimate that claims and returns from business customers represent 1% to 3% of orders, at a cost of 2% to 5% of the affected order's value (Vektrum estimates, to be validated against last year's credit notes). They are almost never classified by cause, line or subcontractor.
  3. Stock. Overproduction in clothing worldwide is estimated at 10% to 40% a year [32]; Inditex's inventory surplus was 0.57% of items sold in 2024 [25]. For own-brand companies, the ban on destroying unsold clothing already covers large companies and will cover medium-sized ones (dates in chapter 8) [40]. The Banco de Portugal series extracted do not publish inventory days for CAE 13 and 14; each company has to compute this figure from its own balance sheet.
  4. Data requests and brand compliance. In an international survey of 525 facilities, the typical factory had 3 to 5 social audits a year and 45% paid USD 5,000 to 20,000 a year for audits, almost always borne by the factory [33]; the sample is mostly large factories outside Europe. Inditex carried out 6,039 social audits in 2025 [25]. Each customer questionnaire means gathering certificates, raw-material invoices and supplier data.
  1. Invoicing. With 81.0% of companies in the aggregate sending paper invoices [10], most supplier invoices, including those for subcontracted sewing, can be expected to arrive as PDF or paper. The law in force accepts PDF as an electronic invoice only until 31/12/2026 [49]; the 2027 State Budget bill extends that deadline to 31/12/2027, but it is not yet law: the final overall vote is scheduled for 24/11/2026 [51],[50].
Vektrum suggestionSewing invoices read, checked and ready to pay

Starting point

Subcontractors' invoices arrive as PDF or on paper. Someone checks them against the sewing order, posts them and prepares payment. The law in force accepts PDF as an electronic invoice only until 31/12/2026, and the format of some of them will change.

What gets built

  • Reading of each invoice from the AT (Tax Authority) QR code first and OCR only as fallback, without sending invoices to external AI services
  • Automatic matching against the sewing order: quantities, price per piece and discrepancies
  • A queue of only the invoices with doubts, and payment batches from the approved ones

What stays with the team

Approving payments and resolving discrepancies. Every change is logged.

How to know if it is worth it

Count the supplier invoices that come in each month and time ten of them. In the textile company's base scenario, this point is worth 7.2 times more a year than the evidence repository. It is also the capability closest to what Vektrum has already built (chapter 7). See the calculated impact →

Impact 2/3 · Effort 2/3 · 30 to 90 days

First step: set aside one month of sewing invoices and see how many carry the QR code.

Or we can run that review with you →

05

Levers

Opportunities by subsegment

In the textiles, clothing and footwear aggregate, 9.5% of companies with 10 or more people used at least one AI technology in 2025, against 12.0% in the EU and 4.2% a year earlier [7]. Adoption more than doubled, but it is concentrated in text analysis (5.6%); language-generation AI stands at 1.2% (EU: 4.9%) and workflow automation at 0.2% (EU: 1.8%). Business intelligence reaches 5.7% of companies (EU: 9.7%) [8], and 46.6% have very low digital intensity, against 38.8% in the EU (2024) [9]. In Europe, only 20% of companies in the textile ecosystem have high digital intensity [19].

The sector has its own digitalisation agenda under the Recovery and Resilience Plan: Texp@ct, led by Impetus, with 39 partners and €29.2m of incentives, which ran until 30/06/2026 [46]. In a related survey of 72 companies, 95% rate the impact of digitalisation positively, and it is driven mainly by external pressure [45]. We organise the opportunity into four levers: revenue (answer brands faster), margin (cost with real data), cost (fewer hours of repetitive work) and capital (less stock and faster collection).

Opportunities by subsegment
SubsegmentRevenueMarginCostCapital
Spinning and weavingSame-day answers to quote requestsPrice per article with current energy and fibre costsOrders and invoices read automaticallyCollections tracked per customer
Knitwear and private-label garment makingCosting from the tech pack in hoursSamples only for customers who convertWork-in-progress board for subcontractorsInvoice on the day of shipment
Finishing, dyeing and printingPricing by process and recipeReprocessing cost visible per customerJob orders without retypingEnergy allocated to each batch
Technical and home textilesTechnical file ready for each customerPrice by specification, not by averageCertificates and tests in a single repositoryCollection stock per item
Own-brand garment makingProduct data ready for the digital passportEnvironmental claims with evidenceClaims classified by causeLess overproduction and dead stock

Vektrum synthesis. Six of these opportunities are quantified in chapter 6.

Private-label garment making: development before production. Brands ask for low minimum order quantities and last-minute changes [28]. Vektrum reading: in a garment maker, sewing is measured to the minute, but what happens before it (costings, samples, fit comments) is rarely measured. The lever is knowing, by customer and style, how many samples were made, how many converted and what they cost.

Finishing and dyeing: energy and reprocessing in the price. In an industrial service where energy reached 8.8% of textile purchases [2], process-based pricing and allocating energy to each batch decide whether the batch makes money. The information exists in job orders and energy bills; it is often not connected.

Own brand: product data becomes an obligation. The digital product passport will require fibre composition, origin, economic operators, repair and end of life [39]. CITEVE points to the multiplicity of certification schemes and the difficulty of accessing supplier data as obstacles [36]. Portugal is the 7th country by GOTS-certified facilities, with 4.8% of 17,800 [34], and the 105 companies in the sector's sustainability report hold 2,526 environmental certifications [35]. Vektrum reading: the certificates exist; what is missing is linking them to each item and batch.

Vektrum suggestionKnow which customers convert before cutting the next sample

Starting point

The sample record usually lives in a spreadsheet, in emails and in the memory of whoever runs development. The information exists, but it is not linked to the order that did, or did not, follow.

What gets built

  • A simple sample register, linked to customer, style and cost, in the ERP or in a shared sheet the team already uses
  • Automatic linking of each sample to the matching order, when it arrives
  • A monthly summary of conversion by customer and garment type

What stays with the team

Deciding whether the next sample for a customer who does not convert is made, charged for or declined. The system shows the history; it does not decide.

How to know if it is worth it

Count the samples from the last twelve months that led to no order. The weight of this point on the margin is in the sample-tracking scenario in chapter 6. See the calculated impact →

Impact 3/3 · Effort 1/3 · 30 to 90 days

First step: export the 2026 sample list and cross it with orders.

Or we can run that review with you →

06

Quantification

Impact scenarios

The following scenarios apply the official data to two typical profiles and separate data (with a source) from Vektrum assumptions. Each has three variants, which combine the low, central and high values of each assumption, in the order shown. The aim is to show the order of magnitude of the annual gain and of the margin effect, not to promise a return. The value of hours assumes the time freed is redeployed to useful work; if it is not, the cash effect is smaller. Gains are stated before any running costs. We do not present implementation costs, because the investment depends on each project's scope. The last row of each scenario is not a price: it shows the largest investment the annual gain recovers in 24 months, as a planning ceiling. Replace the assumptions with your company's figures.

On the cost of an hour: we use €15/h (€13 to €18/h) for technical-administrative work (costing staff, pattern makers, sales, planners, accounting). We start from manufacturing's €16.5/h [11] with a discount of about 10% for the sector. The average hourly costs in textiles (€12.43) and clothing (€10.05) [2] mostly reflect machine operators and understate these profiles.

Company profiles used in the scenarios
IndicatorPrivate-label garment makerTextile company (weaving or finishing)
People5030
Estimated turnover€2,760,000€2,760,000
Reference EBITDA margin (EBITDA ÷ sales)7.1%10.6%
Reference EBITDA (rounded)€196,500€291,900

Vektrum calculation. Turnover = turnover per person of the activity code (Eurostat, 2024: €55.1k in clothing and €91.9k in textiles) × people, rounded to €10,000. Margin = EBITDA ÷ sales of each activity code (BdP, 2025), an average dominated by larger firms; many SMEs sit below it. Sources: [2],[1]

Profile A: private-label garment maker, 50 people

Garment maker, 50 people: costing from tech packs and historical costs

Assumptions

  • Costing requests per month30 / 60 / 120Vektrum estimate
  • Hours per full costing1.5 / 2.0 / 3.0Vektrum estimate
  • Time reduction with pre-filling from the tech pack and history40% / 50% / 60%Vektrum estimate
  • Loaded cost of one technical-administrative hour[11],[2]13 €/h / 15 €/h / 18 €/hVektrum estimate
  • Reference EBITDA margin (clothing, EBITDA ÷ sales)[1]7.1%Vektrum calculation
ResultCautiousBaseFavourable
Hours freed per year2167202,592
Annual gain€2,808€10,800€46,656
Effect on EBITDA margin+0.10 pp+0.39 pp+1.69 pp
Investment recovered in 24 months (ceiling)€5,600€21,600€93,300

The assistant prepares the costing and a person validates it and sets the price. It excludes the effect of faster answers on the conversion rate of quotes into orders (20% to 40%, Vektrum estimate), which is the main value multiplier and should be measured in the company's own history. The 24-month ceiling is not a price: the investment depends on each project's scope (method in the chapter introduction).

Garment maker, 50 people: sample tracking and conversion analysis

Assumptions

  • Samples and prototypes per month40 / 80 / 120Vektrum estimate
  • Direct cost per sample (material, pattern, cutting, sewing, shipping)40 € / 80 € / 120 €Vektrum estimate
  • Unproductive samples avoided[47]10% / 15% / 20%Vektrum estimate
  • Administrative minutes per sample20 / 30 / 45Vektrum estimate
  • Reduction of follow-up time30% / 40% / 50%Vektrum estimate
  • Loaded cost of one technical-administrative hour[11],[2]13 €/h / 15 €/h / 18 €/hVektrum estimate
  • Reference EBITDA margin (clothing, EBITDA ÷ sales)[1]7.1%Vektrum calculation
ResultCautiousBaseFavourable
Hours freed per year48192540
Annual gain€2,544€14,400€44,280
Effect on EBITDA margin+0.09 pp+0.52 pp+1.60 pp
Investment recovered in 24 months (ceiling)€5,100€28,800€88,600

Vektrum estimate. 80% of the base gain is the direct cost of samples no longer made (€1,920 / €11,520 / €34,560); the rest is follow-up hours. It is the most uncertain assumption in this chapter: there are no public data on sample conversion in Portugal, and the only benchmark (40% of samples never moving forward) comes from a vendor [47]. It only counts samples for customers and styles that historically do not convert; if the brand pays for part of the samples, or if declining some costs orders, the saving is smaller. Validate it against the last twelve months of sample records. The 24-month ceiling is not a price: the investment depends on each project's scope (method in the chapter introduction).

Garment maker, 50 people: work-in-progress board for subcontractors

Assumptions

  • Weekly hours planning and chasing subcontractors10 / 15 / 25Vektrum estimate
  • Working weeks per year46Vektrum estimate
  • Hours reduction with a shared board30% / 40% / 50%Vektrum estimate
  • Loaded cost of one technical-administrative hour[11],[2]13 €/h / 15 €/h / 18 €/hVektrum estimate
  • Reference EBITDA margin (clothing, EBITDA ÷ sales)[1]7.1%Vektrum calculation
ResultCautiousBaseFavourable
Hours freed per year138276575
Annual gain€1,794€4,140€10,350
Effect on EBITDA margin+0.06 pp+0.15 pp+0.37 pp
Investment recovered in 24 months (ceiling)€3,600€8,300€20,700

The main value usually lies in fewer delivery delays and penalties, which we do not estimate without the company's data. The 24-month ceiling is not a price: the investment depends on each project's scope (method in the chapter introduction).

Profile B: textile company, 30 people

Textile company, 30 people: orders from email, PDF and brand portals posted to the ERP

Assumptions

  • Orders and purchase orders per month (incl. replenishment)40 / 100 / 250Vektrum estimate
  • Manual minutes per order (20 to 200 lines)15 / 25 / 40Vektrum estimate
  • Reduction with automatic reading and human validation[10]60% / 65% / 70%Vektrum estimate
  • Loaded cost of one technical-administrative hour[11],[2]13 €/h / 15 €/h / 18 €/hVektrum estimate
  • Reference EBITDA margin (textiles, EBITDA ÷ sales)[1]10.6%Vektrum calculation
ResultCautiousBaseFavourable
Hours freed per year723251,400
Annual gain€936€4,875€25,200
Effect on EBITDA margin+0.03 pp+0.18 pp+0.91 pp
Investment recovered in 24 months (ceiling)€1,900€9,800€50,400

The reduction is below the one used for invoices because brand portal and PDF formats are heterogeneous. It excludes the value of fewer errors in the colour and size grid. The 24-month ceiling is not a price: the investment depends on each project's scope (method in the chapter introduction).

Textile company, 30 people: supplier-invoice reading

Assumptions

  • Supplier invoices per month (raw materials, services and subcontracting)200 / 400 / 600Vektrum estimate
  • Manual minutes per invoice6 / 8 / 10Vektrum estimate
  • Reduction of manual time[10]60% / 70% / 80%Vektrum estimate
  • Loaded cost of one technical-administrative hour[11],[2]13 €/h / 15 €/h / 18 €/hVektrum estimate
  • Reference EBITDA margin (textiles, EBITDA ÷ sales)[1]10.6%Vektrum calculation
ResultCautiousBaseFavourable
Hours freed per year144448960
Annual gain€1,872€6,720€17,280
Effect on EBITDA margin+0.07 pp+0.24 pp+0.63 pp
Investment recovered in 24 months (ceiling)€3,700€13,400€34,600

Validate the invoice volume against the company's purchase SAF-T. It excludes fewer exceptions, early-payment discounts and readiness for the end of PDF as an electronic invoice. The 24-month ceiling is not a price: the investment depends on each project's scope (method in the chapter introduction).

Textile company, 30 people: evidence repository for data requests and traceability

Assumptions

  • Customer questionnaires and data requests per year[33]10 / 25 / 40Vektrum estimate
  • Hours per request3 / 5 / 8Vektrum estimate
  • Reduction with a single repository by item and batch40% / 50% / 60%Vektrum estimate
  • Loaded cost of one technical-administrative hour[11],[2]13 €/h / 15 €/h / 18 €/hVektrum estimate
  • Reference EBITDA margin (textiles, EBITDA ÷ sales)[1]10.6%Vektrum calculation
ResultCautiousBaseFavourable
Hours freed per year1263192
Annual gain€156€938€3,456
Effect on EBITDA margin+0.01 pp+0.03 pp+0.13 pp
Investment recovered in 24 months (ceiling)€300€1,900€6,900

The relevant value is not losing customers for lack of data and being ready for the digital passport, which we do not quantify. Hours alone do not justify this project. The 24-month ceiling is not a price: the investment depends on each project's scope (method in the chapter introduction).

Summary of the base scenarios
ProfileHours freed per yearAnnual gainEffect on EBITDA marginInvestment recovered in 24 months (ceiling)
Garment maker, 50 people1,188€29,340+1.06 pp€58,700
Textile company, 30 people836€12,533+0.45 pp€25,100

Vektrum estimate: sum of each profile's three base scenarios. The ceiling is not a price; the investment depends on each project's scope.

07

Transferable capabilities, not textile case studies

What Vektrum has built

The systems below solve problems close to those described in chapter 4: supplier documents in different formats, price lists a person has to review, meetings that generate written work and data that sits in the system but never reaches management. The concrete actions for the sector are in the matrix in chapter 8.

Vektrum case

Transferable capability

Supplier catalogues and price lists loaded into an online store

Built for a distributor that receives its supplier's catalogue and prices as Excel or CSV and has to reflect them in its online store.

What was built

  • Reading of the Excel or CSV file in the format the supplier already sends
  • Matching of supplier references to the company's catalogue
  • Sale-price suggestion, always subject to human approval
  • Store changes made by a separate workflow, once approved

What we learned

  • In textiles, the same pattern would apply to the yarn, fabric and trim price lists that feed costing
  • Separating data reading from writing to the target system makes each step verifiable

Transferable capability, built in another sector. It is not a textile case and no outcome metrics are presented.

Vektrum case

Transferable capability

Supplier-invoice reading and accounts payable

Built for a property-management company, to read supplier invoices and prepare payment without taking control away from the finance team.

What was built

  • Reads the AT (Tax Authority) QR code first, with local OCR as fallback; no invoice is sent to external AI services
  • Cost-centre suggestion for each invoice
  • Validation queues for invoices with doubts or discrepancies
  • Payment batches prepared from approved invoices
  • Audit log of every change

What we learned

  • At a garment maker, subcontractors' sewing invoices would be the first candidates
  • The same reading engine could serve for brands' PDF orders

Transferable capability, built in another sector. It is not a textile case and no outcome metrics are presented.

Vektrum case

Transferable capability

Meeting recording, transcription and a CRM with automatic briefings

Built for Vektrum's own commercial operation, so that no meeting depends on manual notes and every meeting starts with the client's context.

What was built

  • Recording and transcription of meetings, with summary, decisions and tasks
  • Sent to the CRM only after human review
  • Automatic briefing before each meeting, from the contact's history

What we learned

  • At a garment maker, fit and development meetings with brands generate comments that must reach pattern making intact

Transferable capability, built for Vektrum's internal use. It is not a textile case and no outcome metrics are presented.

Vektrum case

Transferable capability

Dashboards and natural-language queries over the company's own database

Connection to the structured data the company already holds (ERP, invoicing, orders) to build management dashboards and allow plain-language questions.

What was built

  • Management dashboards fed directly from the company's database
  • Natural-language queries, with each question translated into a verifiable query
  • Read-only access to source data

What we learned

  • In textiles, the first questions to ask would be margin per customer, sample conversion and claims by cause

Transferable capability, built in another sector. It is not a textile case and no outcome metrics are presented.

08

Priorities

Action plan

The matrix ranks eleven actions by impact, effort and horizon: start with what can be measured in weeks with the systems the company already has, treat regulation as a calendar with dates and leave for last the projects that depend on an organised history.

Priority matrix

Impact

EffortLow → High

Sort by

0 to 30 days

  1. 1

    Measure the baseline

    Count and time costings, samples, orders, subcontractor follow-up, data requests and invoices for two weeks.

    Impact 2/3 · Effort 1/3 · 0 to 30 days

  2. 2

    Prepare for e-invoicing

    Under the law in force, PDF is accepted until 31/12/2026; confirm with the software vendor and the accountant what changes, without relying on the 2027 budget bill before it becomes law.

    Impact 2/3 · Effort 1/3 · 0 to 30 days

  3. 3

    Regulatory map by customer and size

    Record which rules apply to the company and to each customer: destruction ban (large firms since 19/07/2026, medium-sized from 19/07/2030, micro and small exempt), environmental claims and EPR.

    Impact 2/3 · Effort 1/3 · 0 to 30 days

30 to 90 days

  1. 4

    Sample register with conversion

    Link each sample to customer, style, cost and outcome; review conversion per customer every month.

    Impact 3/3 · Effort 1/3 · 30 to 90 days

  2. 5

    Brand orders into the ERP

    Automatic draft order, with colour and size grid, from email, PDF or portal, for validation.

    Impact 3/3 · Effort 2/3 · 30 to 90 days

  3. 6

    Subcontractor work-in-progress board

    Status of each subcontracted order visible in one place, updated by the subcontractors.

    Impact 2/3 · Effort 2/3 · 30 to 90 days

  4. 7

    Supplier-invoice reading

    Read, match to the order and prepare posting and payment, with human validation; start with sewing invoices.

    Impact 2/3 · Effort 2/3 · 30 to 90 days

90 to 180 days

  1. 8

    Evidence repository by item and batch

    Certificates, composition and origin linked to each item; the basis for the digital passport, whose delegated act is planned for adoption in Q4 2027 and, except in justified cases, cannot apply earlier than 18 months after entering into force.

    Impact 2/3 · Effort 2/3 · 90 to 180 days

  2. 9

    Product data for EPR

    For own brands: quantities placed on the market and composition by item, in time for the national schemes required by 17/04/2028.

    Impact 2/3 · Effort 2/3 · 90 to 180 days

  3. 10

    Costing from the tech pack

    Depends on an organised history of costings and actual costs; start with one product family.

    Impact 3/3 · Effort 3/3 · 90 to 180 days

  4. 11

    Claims and defect analysis

    Classify credit notes and claims by cause, line and subcontractor, from the ERP.

    Impact 2/3 · Effort 2/3 · 90 to 180 days

Regulatory calendar reaching textile and clothing companies
DateRuleWho is coveredStatus
19/07/2026Ban on destroying unsold clothing and footwear (ESPR, art. 25)Large companies; medium-sized from 19/07/2030In force
27/09/2026Ban on generic environmental claims (Directive 2024/825)Anyone making claims to consumersApplicable; Portugal had notified no transposition on 27/10/2026
31/12/2026End of PDF acceptance as an electronic invoice (law in force)All companies2027 budget bill extends to 31/12/2027
19/03/2027Transposition of Omnibus I for the CSRDValue-chain companies up to 1,000 employees protectedIn force since 18/03/2026
17/06/2027Transposition of the revised Waste Framework DirectiveMember StatesDirective in force since 16/10/2025
Q4 2027ESPR delegated act for textiles, with digital passportTo be defined in the actPlanned adoption; not yet adopted
17/04/2028EPR schemes for textiles and footwearProducers; micro enterprises (fewer than 10 people, turnover and balance sheet up to €2m) from 17/04/2029Legal deadline

Dates confirmed in the legal texts (EUR-Lex) and the Official Journal. The digital passport delegated act does not yet exist; application dates will depend on it. Sources: [40],[41],[42],[49],[50],[43],[37],[38],[39]

09

Method

Assumptions and limitations

Every figure comes from a research dossier verified on 17/11/2026 against the primary sources; numbers that could not be confirmed were excluded. These are the limitations readers should keep in mind.

  • Scope. CAE 13 (textiles) and CAE 14 (clothing); footwear (CAE 15) is excluded. The sum of both codes' sales at Banco de Portugal (4,074.8 + 4,127.5 = €8,202.3m) confirms that the aggregate excludes footwear [1].
  • Digital data include footwear. Eurostat indicators for AI, ERP, CRM, business intelligence, e-invoicing and digital intensity exist only for the C13–C15 aggregate [7],[8],[10],[9]. The 2025 Portuguese ERP figure is not published.
  • Different universes. Banco de Portugal covers corporations (5,707 in 2025); Eurostat includes sole proprietors (11,965 in 2024); the labour force survey measures people and has small samples by activity. The figures do not add up across sources.
  • Margin and costs. Banco de Portugal's EBITDA margin is a % of income; EBITDA ÷ sales is a Vektrum calculation. The cost structure comes from Eurostat, whose purchases include raw materials and services, subcontracting included. There are no inventory days or separate subcontracting share for CAE 13 and 14.
  • Exports and regions. 2025 values are provisional; January–September 2026 values refer to "textiles and textile articles", which includes fibres [14],[15]. The breakdown by district dates from 2020 and covers clothing only [17].
  • International and vendor benchmarks. Sample data [47] come from a vendor; audit data [33] from large factories outside Europe; overproduction data [32] from global clothing; the Portuguese CITEVE and Texp@ct surveys [44],[45] cover 60 and 72 companies. None describes the average Portuguese SME.
  • Tariffs and regulation. For US tariffs, only the August 2025 commitment is confirmed in an official source [26]; the framework changed several times in 2026. The digital passport delegated act has not been adopted and e-invoicing depends on the 2027 budget vote [50]. This report will be revised when these dates change.
  • Assumptions without a public source. Volumes of costings, samples, orders, invoices and data requests, minutes and hours per task, time reductions and the €15/h hourly cost are Vektrum estimates. They were chosen prudently and should be replaced with your company's data. The 24-month investment ceiling is not a price.

Sources

Every figure in the report points to a numbered source. Vektrum calculations and estimates are labelled as such.

  1. [1]BPstat, Central Balance Sheet studies: textiles and clothing industry (CAE 13 and 14), annual indicatorsBanco de Portugal · 2026 · Official source
  2. [2]Structural business statistics, sbs_ovw_act (C13, C14 and C)Eurostat · 2026 · Official source
  3. [3]Structural business statistics by size class, sbs_sc_ovwEurostat · 2026 · Official source
  4. [4]Annual industrial production, sts_inpr_aEurostat · 2026 · Official source
  5. [5]Monthly industrial production, sts_inpr_mEurostat · 2026 · Official source
  6. [6]Labour force survey by activity and age, lfsa_egan22dEurostat · 2026 · Official source
  7. [7]Use of artificial intelligence in enterprises by activity, isoc_eb_ain2Eurostat · 2026 · Official source
  8. [8]Use of ERP, CRM and business intelligence by activity, isoc_eb_iipn2Eurostat · 2026 · Official source
  9. [9]Digital intensity index by activity, isoc_e_diin2Eurostat · 2026 · Official source
  10. [10]E-invoicing and supply-chain information sharing by activity, isoc_eb_icsn2Eurostat · 2024 · Official source
  11. [11]Hourly labour costs, lc_lci_levEurostat · 2026 · Official source
  12. [12]Electricity prices for non-household consumers, nrg_pc_205Eurostat · 2026 · Official source
  13. [13]Gas prices for non-household consumers, nrg_pc_203Eurostat · 2026 · Official source
  14. [14]Textile and clothing exports dip slightly in 2025, on INE data (10/02/2026)Portugal Têxtil (ATP) · 2026 · Study
  15. [15]Textile and clothing exports accelerate to September, on INE data (09/11/2026)Portugal Têxtil (ATP) · 2026 · Study
  16. [16]Textile and clothing exports fall 0.8% to €5,499m in 2025Forbes Portugal · 2026 · Press
  17. [17]Study of the Clothing Industry, based on Quadros de Pessoal employer records (2023)FESETE · 2023 · Study
  18. [18]Sector note: Textile and Clothing Industry (October 2025)Garantia Mútua · 2025 · Study
  19. [19]Facts & Key Figures 2026EURATEX · 2026 · Study
  20. [20]Interview with Ricardo Silva, president of ATP (27/11/2025)ECO · 2025 · Press
  21. [21]Decree-Law 107/2023 (guaranteed minimum monthly wage for 2024)Diário da República · 2023 · Official source
  22. [22]Decree-Law 112/2024 (guaranteed minimum monthly wage for 2025)Diário da República · 2024 · Official source
  23. [23]National minimum wage for 2026 set at €920 (Decree-Law 139/2025)APCMC · 2025 · Official source
  24. [24]Guaranteed minimum monthly wage 2025 and the 2025–2028 tripartite agreementCuatrecasas · 2024 · Study
  25. [25]Consolidated Statement of Non-Financial Information 2025 (company report)Inditex · 2026 · Study
  26. [26]U.S.-EU Tariffs and Trade Framework Agreement (IF13107)Congressional Research Service · 2025 · Official source
  27. [27]Textiles and clothing expect a "significant impact" from the new US tariffsRTP · 2025 · Press
  28. [28]4 fashion supply chain trends to watch in 2026, on The State of Fashion 2026Supply Chain Dive (McKinsey & BoF) · 2026 · Study
  29. [29]Apparel nearshoring cut lead times, on "Is apparel manufacturing coming home?" (16/10/2018)Supply Chain Dive (McKinsey) · 2018 · Study
  30. [30]71% of fashion companies turning to nearshoring, survey of 38 chief procurement officers (November 2021)Fibre2Fashion (McKinsey) · 2021 · Study
  31. [31]2026 Fashion Industry Benchmarking Study (23/07/2026)Sheng Lu / USFIA · 2026 · Study
  32. [32]Doing more with less (20/02/2023)OC&C Strategy Consultants · 2023 · Study
  33. [33]How Many Social Audits Are Enough? (16/10/2023; 525 facilities in 32 countries)WRAP · 2023 · Study
  34. [34]Annual Report 2025GOTS · 2026 · Study
  35. [35]Third Sustainability Report of the textile and clothing sector, be@t project (22/12/2025)Jornal Económico · 2025 · Press
  36. [36]Digital passport could add value to Portuguese origin (28/10/2026)Portugal Têxtil · 2026 · Press
  37. [37]Directive (EU) 2025/1892 (revision of the Waste Framework Directive)EUR-Lex · 2025 · Official source
  38. [38]Revised Waste Framework Directive enters into force (16/10/2025)Comissão Europeia · 2025 · Official source
  39. [39]Textile Apparel and the Digital Product PassportComissão Europeia · 2026 · Official source
  40. [40]Regulation (EU) 2024/1781 (ESPR), articles 4 and 25 and Annex VIIEUR-Lex · 2024 · Official source
  41. [41]Directive (EU) 2024/825 (empowering consumers for the green transition), article 4EUR-Lex · 2024 · Official source
  42. [42]The EU Directive on Empowering Consumers for the Green Transition two years on (27/10/2026)Mayer Brown · 2026 · Study
  43. [43]Directive (EU) 2026/470 (Omnibus I: CSRD and CSDDD)EUR-Lex · 2026 · Official source
  44. [44]Portuguese fashion is embracing AI: CITEVE survey of 60 companies (16/10/2025)Portugal Têxtil · 2025 · Press
  45. [45]Digitalisation and robotics take hold in textiles and clothing: Texp@ct survey of 72 companies (02/07/2026)Portugal Têxtil · 2026 · Press
  46. [46]Project sheet of the Texp@ct mobilising agenda (November 2025)F3M · 2025 · Study
  47. [47]Game-Changing Benefits Of 3D Digital Sampling, article by the CEO of a 3D sampling vendor (2024)Textile World (VARIANT3D) · 2024 · Vendor
  48. [48]Artificial intelligence is already "stripping" textile waste (05/04/2025)ECO · 2025 · Press
  49. [49]Law 73-A/2025 of 30 December (2026 State Budget), article 95Diário da República · 2025 · Official source
  50. [50]2027 State Budget: Bill 110/XVII/2, voting calendarAssembleia da República · 2026 · Official source
  51. [51]Government again postpones the obligation to submit the accounting SAF-T file (08/10/2026)ECO · 2026 · Press
  52. [52]Industry, production and logistics: where margin leaks and how to recover it (16/11/2026)Vektrum Insights · 2026 · Vektrum

Methodology

Sources: Banco de Portugal (Central Balance Sheet Database, via BPstat) and Eurostat for Portuguese and EU data; INE, via ATP, for exports; EUR-Lex and the Portuguese Official Journal for the legal framework [40],[49]; sector and international studies identified as such, with their scope stated. All sources were accessed on 17/11/2026.

Verification: every number in the underlying dossier was checked against its primary source; derived ratios were recalculated from the original series, and regulatory dates were confirmed in the legal texts. Values that could not be confirmed were excluded.

Vektrum calculations: ratios to turnover or sales, EBITDA ÷ sales, Portugal/EU comparisons, changes and shares are arithmetic on published series, with the formula stated.

Scenarios: hours freed = volume × time per item × reduction; annual gain = hours × hourly cost, plus the direct cost of samples avoided; margin effect = annual gain ÷ turnover; investment ceiling = annual gain × 24 months ÷ 12. They exclude implementation and running costs, which depend on each project's scope. All results are calculated in code from the published assumptions.

Other reports

We like meeting new companies, understanding how they work and finding where AI can make a difference.

Talk to us