Digital Product Passports for Furniture: What Companies Need to Know

Executive Summary

Furniture is one of the few sectors named in the text of the Ecodesign for Sustainable Products Regulation itself. Article 18(5) of Regulation (EU) 2024/1781 obliges the European Commission to give priority, in its first working plan, to a specific list of product groups, and that list includes furniture, including mattresses. The Commission did exactly that: the Ecodesign for Sustainable Products and Energy Labelling Working Plan 2025-2030, COM(2025) 187, adopted 19 April 2025, ranks furniture and mattresses among its priority final product groups.

That is the strongest statement that can honestly be made today. It is not a deadline. As at the date of this article, no delegated act imposing Digital Product Passport requirements on furniture or mattresses has been adopted. There is no published furniture passport data model, no furniture carrier specification, no furniture identity rule and no furniture application date. Any vendor, consultant or conference slide presenting one is presenting an estimate.

What furniture companies do have today is a set of adopted obligations that are not passports but that generate almost all of the data a future passport would draw on: deforestation due diligence for wood and wood furniture under Regulation (EU) 2023/1115, general product safety traceability and recall duties under Regulation (EU) 2023/988, chemical restrictions under REACH including the formaldehyde restriction affecting wood-based panels and furniture, packaging obligations under Regulation (EU) 2025/40, and, in several Member States, national extended producer responsibility schemes for furniture that already demand product-level material declarations.

The strategic point of this guide is that furniture’s preparation problem is not legal, it is structural. Furniture is a variant-heavy, made-to-order, low-margin sector in which a single “product” is often a configuration space of thousands of finish, fabric, size and component permutations, built by contract manufacturers the brand does not own, from materials whose origin the brand’s own systems do not record. A passport requirement will not create those problems. It will expose them.

Table of Contents

• What Is a Furniture Digital Product Passport?
• Does Furniture Already Require a Digital Product Passport?
• Which Furniture Products Could Be Affected?
• Why Furniture Is Difficult
• What Information May Matter?
• Where Does Furniture Passport Data Come From?
• Wood, Materials and Chain of Custody
• Chemicals, Emissions and Substance Data
• Supplier and Contract Manufacturer Data
• Evidence and Regulatory Claims
• Data Quality and Validation
• Product Identity, Variants and Data Carriers
• GS1, Digital Link, EPCIS and Other Standards
• Durability, Repair and Spare Parts Information
• Enterprise Architecture
• Legal Responsibility
• Conformity and Existing Product Regulation
• Preparing Before Final Requirements Exist
• Pilot Strategy
• Worked Example
• Common Mistakes
• Preparation Checklist
• Frequently Asked Questions
• References

What Is a Furniture Digital Product Passport?

A Digital Product Passport is a structured, machine-readable record about a specific product, reachable from a data carrier physically attached to that product, containing information that a legal act requires to be there, accessible to the parties that the same legal act says may see it.

Three articles of the framework matter more than any commentary about them.

Article 9 establishes the passport obligation itself and makes it conditional: products may be placed on the market or put into service only if a passport is available in accordance with the applicable delegated act. No applicable delegated act, no obligation. This is why “furniture needs a passport by 2027” is not a statement of law.

Article 10 sets the essential requirements a passport must satisfy once a delegated act exists: connection to a unique product identifier, reachability through a data carrier, compliance with open and interoperable standards, availability to the actors specified in the act, and protection of personal and commercially confidential data.

Article 11 governs technical design and operation, including the requirement that data be held in a way that does not depend on a single provider remaining in business, and that access continues for a period that the delegated act sets.

Everything a furniture company actually wants to know, which fields, which language, which identity level, whether a mattress carries a passport per model or per unit, whether a made-to-order sofa carries one at all, who can read the confidential parts, is reserved to the delegated act. That is not evasion. It is the design of the framework, and it is the reason preparation must be about capability rather than field lists.

Does Furniture Already Require a Digital Product Passport?

No. It is worth separating five statements that are routinely blended together in furniture industry commentary, because only two of them are law.

Statement one: the framework is in force. True. Regulation (EU) 2024/1781 entered into force on 18 July 2024. Its framework provisions, including the empowerment to adopt product-specific delegated acts and several horizontal obligations, apply now.

Statement two: furniture is a named priority. True, and stronger for furniture than for most sectors. Article 18(5) of the Regulation itself directs the Commission to prioritise a specific list of product groups in its first working plan, and furniture including mattresses is on that statutory list. The Commission then ranked furniture and mattresses among the priority final product groups in COM(2025) 187, adopted 19 April 2025.

Statement three: furniture therefore has a passport deadline. False. A working plan is a planning communication. It states what the Commission intends to work on and in what order. It creates no obligation on any manufacturer, importer or retailer, and it does not fix an application date.

Statement four: the furniture passport fields are known. False. No preparatory study output, no stakeholder consultation draft and no vendor data model is a legal requirement. Draft material is useful for anticipating themes. It is not a specification.

Statement five: furniture is otherwise unregulated until then. False, and this is the misconception that costs the most money. Several adopted regimes already impose documentation, traceability and due diligence obligations on furniture businesses, and the data they require overlaps heavily with what any future passport would carry.

Which Furniture Products Could Be Affected?

The sector label “furniture” covers product families with radically different manufacturing models, and the working plan’s phrase “furniture, including mattresses” deliberately captures a wide field. The table below is an analytical map for planning, not a legal scope statement, because no legal scope statement exists.

Two boundary cases deserve care. First, furniture with an electrical function, powered recliners, height-adjustable desks, lit display cabinets, mattresses with integrated electronics, may sit inside more than one regime at once, and the electrical parts already attract obligations of their own. Second, furniture containing batteries falls within Regulation (EU) 2023/1542, which introduces the battery passport for certain battery categories on its own timetable, independent of any furniture measure.

Why Furniture Is Difficult

Furniture companies frequently assume their passport problem will be smaller than an electronics or battery manufacturer’s, because their products are mechanically simple. The opposite is closer to the truth. Furniture’s difficulty is not technical complexity per unit, it is combinatorial complexity per range, combined with a sector cost structure that leaves little margin for data work.

Variant explosion. A single upholstered sofa model may be orderable in four sizes, three arm styles, two leg finishes and two hundred fabrics. That is a configuration space in the thousands. If a passport attaches to a product as placed on the market, the question of what “the product” is has thousands of possible answers within one catalogue entry, and material composition genuinely differs across them.

Made-to-order and project business. Kitchens, fitted wardrobes and contract fit-outs are manufactured against an order, not held as stock. There is no pre-existing master record for the thing that ships. The data record has to be produced by the order, which means the order system, not the product catalogue, becomes the source of passport data.

Material substitution as normal practice. Furniture manufacturing routinely substitutes a board supplier, a foam grade, a hinge or a fabric mill without changing the article number, because the substitute is functionally identical. It is not compositionally identical, and a substance or origin claim made at model level silently becomes wrong for part of the production.

Wood and fibre origin is deliberately obscured upstream. Particleboard, MDF and plywood are made from blended fibre and veneer from many sources. Chain of custody at the level of “which forest” is exactly what the deforestation regime is forcing into existence, and it is exactly what furniture brands have historically not held.

Contract manufacture is the norm. Many recognised furniture brands manufacture little or nothing. The legal responsibility for placing the product on the EU market sits with the brand or importer, and the data sits with a factory in another jurisdiction that has no contractual obligation to supply it in a structured form.

Long product life, short data memory. Furniture stays in use for decades and is resold, reupholstered and repaired. A passport requirement implies a data record that outlives the product range, the ERP migration and often the supplier relationship.

Low margin, small companies. The European furniture sector is dominated by small and medium enterprises, many of them workshops with no product data system beyond a spreadsheet and an accounting package. Any preparation advice that assumes a PLM implementation is advice for a minority of the sector.

What Information May Matter?

No furniture field list exists. What follows is a planning inventory derived from three defensible sources: the categories of information Article 7 of Regulation (EU) 2024/1781 permits a delegated act to require, the data already demanded by adopted regimes that apply to furniture today, and the criteria already used by voluntary schemes such as the EU Ecolabel for furniture and mattresses. Treat every row as a capability to build, not a field to publish.

Two observations matter more than the list. First, most of this data already exists somewhere in a furniture business; it exists as documents, not as data, and it is bound to a purchase order or a supplier, not to a product. Second, the hardest rows are not the exotic ones. Material composition by mass, for a sofa with a beech frame, a mixed polyurethane and fibre fill, a woven upholstery fabric of undisclosed blend and forty metal fasteners, is a genuinely difficult data problem for a business that has never been asked to answer it.

Where Does Furniture Passport Data Come From?

Every field in a passport has an origin, and the origin determines how reliable it is, who can change it and what happens when it is wrong. In furniture, four origins dominate.

Internally generated data. Article numbers, dimensions, model descriptions, assembly instructions, declared warranty terms. The business controls it, and it is usually the cleanest data available.

Supplier-provided data. Board specifications, foam grades, fabric composition, adhesive and coating declarations, fitting materials. This is the largest category by volume and the weakest by reliability, because it arrives as a PDF attached to an email from a sales representative rather than as a governed declaration.

Measured or tested data. Emission test results, structural and durability test reports, fire performance results where applicable. Reliable but expensive, valid only for the configuration tested, and frequently assumed to cover variants that it does not cover.

Derived or calculated data. Composition percentages computed from a bill of materials, recycled content aggregated across components, mass calculated rather than weighed. Derived data is legitimate, but it inherits every weakness of its inputs and it must be labelled as derived, not presented as measured.

The furniture-specific version of this principle is that a large proportion of the highest-risk fields originate outside the business. A brand can improve its internal data by internal effort. It can only improve supplier data by changing what it asks for, in writing, in a purchase agreement.

Wood, Materials and Chain of Custody

Wood is the defining material problem of the furniture passport, and it is the one area where adopted law is already forcing the change.

Regulation (EU) 2023/1115, the deforestation regulation, requires operators placing relevant commodities and products on the EU market, including wood and products made of wood such as furniture, to exercise due diligence, collect information about the product including species, country of production and the geolocation of the plots where the wood was harvested, assess and mitigate risk, and submit a due diligence statement. The regulation’s application dates have been amended twice; according to the European Commission’s guidance on the regulation, application is now set at 30 December 2026 for large and medium operators and 30 June 2027 for micro and small operators, with micro and small operators already covered by the former EU Timber Regulation applying from 30 December 2026. Because these dates have moved before, verify the current position on EUR-Lex and the Commission’s deforestation pages before building a plan around them.

The significance for passports is direct. Geolocated, species-level, origin-traced wood data is exactly the kind of upstream traceability that a furniture passport would plausibly reference, and it is being built for a different legal reason. A furniture business that treats deforestation due diligence as a one-off compliance filing will build the data in a form it cannot reuse. A business that treats it as the first tranche of product-level material provenance will have solved a meaningful part of its passport problem before any furniture act exists.

Three practical consequences follow.

Panel products break plot-level traceability by design. Particleboard and fibreboard are manufactured from blended feedstock. Traceability resolves to a mill and a declared sourcing area, not to a tree. That is a legitimate answer, but it must be recorded as what it is rather than dressed up as plot-level precision.

Certification is evidence, not composition. Forest certification scheme claims are useful evidence that a management standard was met. They are not a statement of species, mass or origin, and they should be stored as evidence bound to a claim rather than substituted for the claim.

Non-wood materials need the same discipline. Metals, foams, fibres, glass, stone and plastics in furniture all have origin, composition and recycled content questions attached. Furniture businesses routinely build wood traceability well and then find that the metal frame, which is thirty per cent of the product’s mass, has no data at all.

Chemicals, Emissions and Substance Data

Furniture is a chemically dense product category: adhesives, coatings, lacquers, flame retardants, foaming agents, dyes and finishes. Several adopted regimes already generate substance data, and any passport requirement would draw on the same sources.

REACH applies across the board. Articles supplied in the EU carry obligations to communicate information about substances of very high concern present above the relevant threshold, and the SCIP database already receives article-level notifications from many furniture suppliers. That is a substance-to-article mapping exercise that closely resembles part of a passport data model.

Formaldehyde emissions from wood-based panels and furniture articles are restricted through REACH Annex XVII. The restriction was introduced by Commission Regulation (EU) 2023/1464 with transitional periods running into 2026 and 2027 for different article categories. Because the transitional structure differs by article type, confirm the current text of the Annex XVII entry rather than relying on a single headline date.

Persistent organic pollutants and flame retardants are restricted through separate instruments and are particularly relevant to upholstery and mattress components, where legacy flame retardant chemistry both constrains recycling and complicates recycled content claims.

The practical passport lesson is that substance data has to be held at component level and rolled up, not asserted at product level. A statement that a sofa contains no substance of concern above the threshold is only defensible if the business can show which components were assessed, on what supplier declarations, of what date, covering which production period.

Supplier and Contract Manufacturer Data

For most furniture brands, the passport problem is a supplier data problem wearing a different hat.

Furniture supply chains are typically shallow in tiers but wide in count. A single upholstery range may draw on a frame workshop, a foam converter, four fabric mills, a fittings importer and a packaging supplier, and the brand may deal directly with only two of them. Contract manufacture concentrates the difficulty further: the factory holds the true bill of materials, the actual substitutions and the real production records, and the brand holds the legal responsibility.

Four moves reliably improve the position.

Ask for data in a fixed structure, not in free text. A supplier who is asked for “material information” sends a brochure. A supplier who is sent a fixed template with defined fields, units and allowed values sends data. The template matters more than the technology.

Put the obligation in the contract. Data provision, update on change, notification of substitution and the right to audit should be terms of purchase, not favours. Renewals are the practical moment to introduce them.

Segment suppliers by risk, not by spend. The highest-risk supplier is often a low-spend one: the fittings importer who changes source without notice, or the fabric agent who cannot name the mill.

Make substitution notification the single non-negotiable term. Almost every material data failure in furniture traces back to an undeclared substitution. A supplier that reliably tells you when something changed is more valuable than one that supplies perfect data once.

Small furniture businesses often object that they have no leverage over suppliers. Frequently true, and the honest answer is that the smaller company’s strategy is different: record what you can obtain, record explicitly what you could not obtain and from whom, and keep the request trail. A documented, dated request that a supplier declined to answer is a materially stronger position than silence.

Evidence and Regulatory Claims

A passport claim is only as good as the artefact behind it, and furniture has a specific weakness here: its test evidence is usually configuration-specific while its claims are made range-wide.

Structural and durability testing is performed on a representative sample. Emission testing is performed on a specific board or a specific finished article. Fire performance testing, where applicable, is performed on a specific fabric and filling combination. Each report is valid for what was tested. The commercial temptation is to publish the resulting claim across a range that includes untested combinations, and in a passport that temptation becomes a documented misstatement rather than a marketing generalisation.

Evidence management for furniture should therefore capture, for every report:

• what exactly was tested, described in terms that map to the product configuration model
• which product variants the report is asserted to cover, and on what technical justification
• who performed the test and under what standard
• the date and the validity period, where one applies
• what change would invalidate it, a different foam grade, a different fabric weight, a different frame supplier

Data Quality and Validation

Furniture data fails in predictable ways, and each failure mode has a specific check.

Completeness. Every field a claim depends on has a value, and every missing value is explicitly recorded as missing rather than silently absent. In furniture, the most common gap is the fabric or fitting whose composition nobody ever obtained.

Consistency. Composition percentages sum correctly, declared mass matches the sum of component masses within tolerance, units are used consistently, and the same material is named the same way across suppliers. Furniture bills of materials are notorious for mixing metric and imperial fastener specifications and for describing the same foam grade three different ways.

Plausibility. A wardrobe that reports a total mass of nine kilograms, a recycled content figure of a hundred and four per cent, or a beech frame with a declared density outside the physical range for beech should be rejected by a rule, not by a reviewer’s intuition.

Evidence sufficiency. Every claim requiring evidence has evidence attached, in date, and covering the configuration to which the claim is applied.

Currency. Data has a review date. Supplier declarations expire. A record that has not been touched in four years across three supplier changes should be flagged as stale even if nothing is provably wrong.

Product Identity, Variants and Data Carriers

Identity is the decision that determines everything else, and it is the decision furniture businesses are least prepared for.

Article 10 requires a passport to be connected to a unique product identifier. The delegated act decides what that identifier identifies: a model, a batch, or an individual item. Each answer implies a different data architecture, a different production process and a different cost base.

The furniture-specific complication is that the same business commonly needs all three at once. A flat-pack bookcase produced in long runs is naturally batch-level. A configured kitchen is inherently item-level. A mattress with a hygiene and warranty history has strong reasons to be item-level regardless of what any act requires.

Variants are the hidden cost. A range with two hundred fabric options is either two hundred data records or one record with a documented rule for what varies. Both are workable. What is not workable is a single record that quietly asserts one fabric’s composition for all two hundred, which is the current state of many furniture catalogues.

Made-to-order needs order-driven records. For configured products, the passport record cannot be authored in the catalogue because the configuration does not exist until the order does. The realistic architecture generates the record from the order configuration plus the component data library, at the moment of manufacture.

Carriers must survive furniture reality. A furniture data carrier has to survive warehousing, flat packing, home assembly, upholstery cleaning and twenty years of use. A paper label under a sofa cushion does not. Options in practical use include printed labels under seat frames or inside drawer boxes, engraved or branded marks on the frame, carriers on the assembly instructions, and carriers on the packaging, the last of which is the least durable, since the packaging is discarded within an hour of delivery. The delegated act will decide what is acceptable. The engineering work of finding a placement that survives the product’s life can start now.

GS1, Digital Link, EPCIS and Other Standards

Furniture has weaker identification hygiene than food, pharmaceuticals or electronics. Many ranges are sold on internal article numbers, and global trade item numbers are applied inconsistently or only for the retail channel.

Three standards families are worth understanding, without over-committing to any of them before the delegated act specifies interoperability requirements.

GS1 identification keys provide globally unique product identification. For furniture, the practical gain is not compliance, it is the elimination of the identity ambiguity that makes retailer, marketplace and logistics data reconciliation expensive today.

GS1 Digital Link expresses an identifier as a resolvable web address, so a single carrier can serve consumers, retailers and authorities with different content according to who is asking. This is the mechanism most obviously suited to the passport’s differentiated access requirement in Article 10.

EPCIS records events, produced, shipped, received, repaired, resold, rather than static attributes. For a product category with a twenty-year life, a second-hand market and a repair economy, event data is where the sector’s long-term circularity story lives, even though no current obligation requires it.

The standards landscape for passports as a whole, including the European standardisation work underway for Digital Product Passport interoperability, is covered separately in the standards article referenced at the end of this guide. The disciplined position for furniture is to adopt globally unique identification now, because it pays for itself in commercial data quality, and to defer carrier and transport format commitments until a requirement exists.

Durability, Repair and Spare Parts Information

Durability and repairability are recurring ecodesign parameters across every sector the Commission has addressed, and furniture is unusually exposed because its repair economy is informal and its spare parts practice is weak.

The realistic preparation questions are structural rather than regulatory.

• Can the business identify, for a product sold six years ago, which fittings, mechanisms, foams and fabrics it contained?
• Does a spare parts list exist for each range, with part identifiers a customer or repairer could quote?
• Are assembly instructions retained in a form that maps to the specific configuration shipped, including superseded versions?
• Is there a documented disassembly sequence, and does it identify which materials separate cleanly for recycling?
• Are declared warranty and expected service life statements supported by test evidence, or are they commercial statements?

Furniture businesses often find that answering the first question is impossible beyond about two years, because the bill of materials is overwritten in place whenever a component changes. That single practice of overwriting rather than versioning is the most damaging data habit in the sector, and correcting it requires no regulation and no software purchase, only a decision.

Enterprise Architecture

There are two credible architectures for furniture, and the sector’s SME weighting means the smaller one matters more than the vendor literature suggests.

The larger manufacturer or brand. Product data sits across an ERP article master, a PLM or specification system, a supplier portal or quality system, and a document store. The passport-relevant work is to establish a governed product data layer that resolves identity, holds component and material data with origin and evidence, and can express variants and effective dates. The passport itself becomes a projection of that layer, not a separate system.

The small manufacturer or workshop. There is no PLM, and there will not be one. The realistic architecture is a disciplined component library, a single governed list of the boards, foams, fabrics, finishes and fittings the business actually uses, each with composition, origin and supplier evidence attached, plus a product record that references those components rather than restating them. This is achievable in modest tooling, and it converts an unbounded per-product data burden into a bounded per-component one.

The second pattern is worth stating plainly because it is the one that will determine whether the European furniture sector copes. A workshop making forty products from sixty components does not have a forty-product data problem. It has a sixty-component data problem, and components are reused.

Legal Responsibility

Responsibility for product data follows the economic operator model used throughout EU product law, and the party that holds it is frequently not the party that made the product.

The manufacturer, the person who manufactures the product, or has it designed or manufactured and markets it under their own name or trade mark, carries the primary obligation. A furniture brand that designs a range and has it built by a contract factory is the manufacturer in law, regardless of who operated the machinery.

The importer placing furniture from outside the EU on the Union market takes on verification duties and cannot discharge them by pointing at the exporter. In practice this is the most exposed position in the furniture sector, because importer-brands frequently have the least visibility of the underlying material data.

The distributor must act with due care, must not supply products they know or should presume to be non-compliant, and has cooperation duties towards market surveillance authorities.

Fulfilment service providers and online marketplaces have their own duties under the market surveillance and general product safety regimes, and the growth of direct-to-consumer furniture selling has made those duties commercially significant.

Two furniture-specific traps recur. First, private label: a retailer that puts its own brand on a factory’s catalogue product becomes the manufacturer, with all the data obligations that implies, often without realising it. Second, modification: substantially modifying a product before placing it on the market, recovering a frame in a different fabric, for example, can make the modifier the manufacturer for legal purposes.

Conformity and Existing Product Regulation

Furniture is not, as a general category, subject to CE marking. There is no furniture directive analogous to the machinery or toys regimes, and this leads many furniture businesses to assume that product regulation is light. The assumption is wrong in a specific way: furniture is regulated primarily through the general regimes rather than a sectoral one, and general regimes are broader, not narrower.

Regulation (EU) 2023/988 on general product safety has applied since 13 December 2024. It requires that consumer products be safe, that the responsible economic operator be identifiable, that products carry information allowing them to be traced, that technical documentation exist, that accidents be notified, and that corrective action including recall be conducted through defined channels. For furniture, a category with genuine stability, entrapment and flammability risks, these are live operational duties, and they demand exactly the traceability foundation that a passport would later reuse.

Regulation (EU) 2019/1020 on market surveillance governs how authorities check products and requires a responsible person established in the Union for certain products offered online.

Harmonised and national standards for furniture strength, stability and durability supply the technical benchmarks against which safety is usually demonstrated. They are not passport requirements, but the test reports they generate are the evidence base a passport would draw on.

National measures matter more in furniture than in most sectors. Extended producer responsibility for furniture is not harmonised at EU level; where schemes exist, they are national, and several already require product-level material and mass declarations for registration and fee calculation. Businesses selling across multiple Member States are therefore already reporting fragments of passport-shaped data into several incompatible national systems.

Preparing Before Final Requirements Exist

The preparation principle for furniture is the same as for every sector where the delegated act is pending: build the capability that is required no matter what the act says, and defer the choices the act will make for you.

Phase 1: Establish identity. Decide how the business names products and variants, apply globally unique identification, and write down the rule for what constitutes a distinct product versus a variant. Nothing else works until this is settled.

Phase 2: Build the component library. One governed record per board, foam, fabric, finish, fitting and fastener actually used, with composition, supplier, origin and evidence attached. Reuse across products is what makes the effort bounded.

Phase 3: Version the bill of materials. Stop overwriting. Record effective dates so the business can state what a product contained at the time it was made.

Phase 4: Fix supplier data collection. Fixed templates, contractual data terms, mandatory substitution notification, and a dated log of requests and refusals.

Phase 5: Bind evidence. Attach every test report and declaration to the specific components and configurations it covers, with validity dates and an explicit statement of coverage.

Phase 6: Reuse the deforestation work. Whatever origin, species and geolocation data the business is already assembling for wood due diligence should be stored as product data, not as a filing.

Phase 7: Validate at entry. Completeness, consistency, plausibility and evidence rules applied when data is captured.

Phase 8: Decide the carrier engineering, not the carrier standard. Establish where on each product family a durable marking can physically live. Leave the encoding and content decisions to the act.

Phase 9: Assign monitoring. One named person tracking EUR-Lex, the working plan and the furniture preparatory work, reporting quarterly. This is a small commitment that prevents both panic and complacency.

Pilot Strategy

A pilot exists to discover where a business’s data actually breaks, which is never where the strategy deck predicted. For furniture, the pilot selection rule is counter-intuitive: choose a product that is representative of the difficulty, not one that is easy to complete.

Choose one product family with real variants. A single upholstered range with multiple fabrics and sizes will surface the identity problem within a week. A single-finish flat-pack table will not.

Include at least one contract-manufactured item. The supplier data gap is the finding that changes budgets, and it only appears when the data has to come from outside.

Include a mattress or a foam-containing product if the business makes one. Foam chemistry, recycled content and flame retardant history concentrate the substance data problem.

Measure the right things. Not “did we produce a passport”. Instead: what proportion of required fields could be populated from existing systems without a human asking someone; how many supplier requests were needed; how long each took; how many claims had evidence that genuinely covered the configuration; and how many variants the data model could express before it broke.

Timebox to eight to twelve weeks. Longer pilots turn into implementations before the findings are understood.

Write down the failures. The pilot’s deliverable is a defect list and a cost estimate, not a demonstration. A pilot that succeeds cleanly was scoped too narrowly to be informative.

Worked Example

Company. Ashcombe Furniture, a fictional mid-sized manufacturer with 180 employees, making domestic case goods in-house and importing an upholstery range from two contract factories outside the EU. It sells in five Member States, directly and through retail partners.

Starting position. Article numbers are internal and inconsistent between the ERP and the website. Bills of materials exist for case goods and are overwritten on change. The upholstery range has no bill of materials at all, only a factory specification sheet in PDF. Wood origin data is being assembled for deforestation due diligence by a compliance consultant, in a spreadsheet, outside the ERP. Test reports sit in a shared drive organised by year.

Trigger. A large retail customer asks Ashcombe to supply material composition and wood origin data per article, in a structured file, as a condition of range renewal. There is no passport obligation involved. The commercial deadline is four months.

What the pilot finds.

• Identity is the first wall. The same wardrobe exists as three article numbers across channels, and the oak and walnut finishes share one number in the ERP but differ in mass and coating chemistry.
• Case goods composition can be assembled to roughly eighty per cent from existing bills of materials, with fittings and adhesives missing entirely.
• The imported upholstery range cannot be answered at all internally. Both factories respond, one within a week with a usable structured reply, one after six weeks with a marketing brochure.
• Two fabric suppliers decline to disclose blend ratios. One later agrees under a confidentiality addendum; the other does not.
• The wood origin spreadsheet is usable but keyed to purchase orders, not to products, so it cannot answer “which forest region is in this wardrobe” without manual reconciliation.
• A durability report cited for the whole seating range covers one of four frame constructions.

What Ashcombe changes.

• Adopts a single governed article identity, with an explicit variant model separating finish, size and fabric as attributes rather than hidden differences.
• Builds a component library covering 240 components, which turns out to cover the entire case goods range and most of the upholstery range.
• Stops overwriting bills of materials and adds effective dates.
• Adds data provision, substitution notification and audit rights to both contract factory agreements at renewal, and moves twenty per cent of volume away from the unresponsive factory.
• Moves the wood origin data into the component library, keyed to boards and timber items rather than to purchase orders.
• Re-binds every test report to the specific configurations it covers, and commissions two additional tests to close the seating gap.

Outcome. Ashcombe meets the retailer’s deadline for case goods and misses it by six weeks for one upholstery range. More importantly, it discovers that its passport readiness cost is dominated by two things that no software vendor sells: supplier contract terms and the discipline of not overwriting data. When a furniture delegated act eventually arrives, Ashcombe will be configuring, not rebuilding.

Common Mistakes

Preparation Checklist

Identity and structure

• A single governed article identity exists, consistent across ERP, website and retail channels.
• The rule distinguishing a distinct product from a variant is written down and applied.
• Globally unique identification is applied to products sold through trade channels.
• The business can state which identity level each product family would realistically need.

Component and material data

• A governed component library exists, covering boards, foams, fabrics, finishes, fittings and fasteners.
• Each component records composition, supplier, origin where relevant and attached evidence.
• Material naming is standardised, and units are consistent.
• Bills of materials are versioned with effective dates and are never overwritten.

Supplier governance

• Structured data templates are issued rather than free-text requests.
• Purchase agreements contain data provision, update-on-change and substitution notification terms.
• A dated log records every request made and every refusal received.
• Suppliers are segmented by data risk, not only by spend.

Wood and provenance

• Deforestation due diligence data is stored as product data keyed to components, not only as filings.
• Species, country of production and, where obtainable, plot-level origin are recorded.
• Panel products record mill and declared sourcing area honestly, without implying plot precision.

Chemicals and evidence

• Substance data is held at component level and rolled up rather than asserted at product level.
• Every test report states what was tested, which configurations it covers and what invalidates it.
• Evidence has validity dates and is bound to claims rather than filed by year.

Validation and operations

• Completeness, consistency, plausibility and evidence rules are applied at data entry.
• Stale records are flagged by a review date.
• A named person monitors EUR-Lex and the working plan and reports quarterly.
• General product safety traceability and recall capability are tested, not assumed.

Frequently Asked Questions

Does furniture need a Digital Product Passport today? No. Regulation (EU) 2024/1781 creates passport obligations only through product-specific delegated acts, and no delegated act covering furniture or mattresses has been adopted.

Furniture is named in the working plan. Does that not make it certain? It makes it a stated priority, and for furniture the priority is unusually strong because Article 18(5) of the Regulation itself lists furniture including mattresses. It still does not create an obligation, fix a date or define content.

When will a furniture delegated act arrive? No adopted instrument states a date for furniture. Treat any specific year attributed to a furniture passport as an estimate and verify it against the working plan text and EUR-Lex before relying on it.

Are mattresses treated separately from furniture? The Regulation and the working plan refer to furniture including mattresses, and mattresses are named explicitly. Whether a future act treats them as one product group or two is unknown.

Does the ban on destroying unsold goods apply to furniture? The specific destruction prohibition introduced by the Regulation targets unsold apparel and footwear. The separate transparency duty on the discarding of unsold consumer products is broader and applies to larger companies. Check the current obligation text against your company size before concluding either way.

We import furniture from outside the EU. Who is responsible? The importer placing the product on the Union market carries verification duties, and a business selling under its own brand is the manufacturer in law regardless of who built it. Both positions carry data obligations that cannot be delegated to the factory.

Will a passport be per model, per batch or per item? Unknown, and it may differ by product family. Made-to-order and configured furniture points strongly towards item-level records; long-run flat-pack production points towards batch level.

Do we need to serialise every piece of furniture now? No. Establish the identity model and confirm that a durable marking placement is physically feasible on each product family. Serialisation at scale is a decision to make when a requirement exists.

What should a small workshop do first? Build a component library. It is bounded, immediately useful for purchasing and quality, and it converts future requirements into a lookup rather than a research project.

Does forest certification satisfy the origin requirement? It is evidence, not a composition or origin statement. Record species, country of production and sourcing area as data, and keep the certificate as supporting evidence bound to the component.

How does the deforestation regulation relate to a furniture passport? It is separate law with its own timetable, but it produces species, origin and geolocation data at product level. That data is highly likely to be reusable for any future passport, if it is stored as product data rather than filed as a compliance return.

Is packaging part of a furniture passport? Packaging is governed by Regulation (EU) 2025/40. Keep packaging data separate from product data so that neither regime’s obligations contaminate the other’s record.

Should we buy passport software now? Buy nothing that requires you to commit to a field list, a carrier standard or a transport format that no adopted act has defined. Invest instead in identity, component data, evidence binding and supplier terms, which are portable across any eventual requirement.

What if our supplier refuses to disclose composition? Record the request, the date and the refusal. A documented, dated attempt is a materially better position than an unexplained gap, and it identifies which supplier relationships need commercial attention at renewal.

Related Articles

• Digital Product Passports for Textiles and Apparel: What Companies Need to Know
• Digital Product Passports for Electronics: What Companies Need to Know
• Digital Product Passports for Batteries: What Companies Need to Know
• Digital Product Passports for Steel and Aluminium: What Companies Need to Know
• How to Manage Evidence for Digital Product Passports
• How to Prepare Suppliers for Digital Product Passports
• Who Needs a Digital Product Passport?
• Digital Product Passports for Construction Products: What Companies Need to Know

References

• Regulation (EU) 2024/1781 establishing a framework for the setting of ecodesign requirements for sustainable products: https://eur-lex.europa.eu/eli/reg/2024/1781/oj
• European Commission, Ecodesign for Sustainable Products and Energy Labelling Working Plan 2025-2030, COM(2025) 187, adopted 19 April 2025: https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=CELEX:52025DC0187
• Regulation (EU) 2023/1115 on the making available on the Union market and the export from the Union of certain commodities and products associated with deforestation and forest degradation: https://eur-lex.europa.eu/eli/reg/2023/1115/oj
• European Commission, Regulation on deforestation-free products, including current application dates: https://environment.ec.europa.eu/topics/forests/deforestation/regulation-deforestation-free-products_en
• Regulation (EU) 2023/988 on general product safety: https://eur-lex.europa.eu/eli/reg/2023/988/oj
• Regulation (EU) 2019/1020 on market surveillance and compliance of products: https://eur-lex.europa.eu/eli/reg/2019/1020/oj
• Regulation (EC) No 1907/2006 (REACH): https://eur-lex.europa.eu/eli/reg/2006/1907/oj
• Commission Regulation (EU) 2023/1464 amending Annex XVII to REACH as regards formaldehyde and formaldehyde releasers: https://eur-lex.europa.eu/eli/reg/2023/1464/oj
• Regulation (EU) 2025/40 on packaging and packaging waste: https://eur-lex.europa.eu/eli/reg/2025/40/oj
• Regulation (EU) 2023/1542 concerning batteries and waste batteries: https://eur-lex.europa.eu/eli/reg/2023/1542/oj
• Commission Decision (EU) 2016/1332 establishing the ecological criteria for the award of the EU Ecolabel for furniture: https://eur-lex.europa.eu/eli/dec/2016/1332/oj
• EUR-Lex, register of delegated and implementing acts, for the current adoption status of measures under Regulation (EU) 2024/1781: https://eur-lex.europa.eu/homepage.html

About This Article

Related Docs

• Industry Guides
• Digital Product Passports for Batteries: What Companies Need to Know
• Digital Product Passports for Construction Products: What Companies Need to Know
• Digital Product Passports for Textiles and Apparel: What Companies Need to Know
• Digital Product Passports for Electronics: What Companies Need to Know
• What is the Ecodesign for Sustainable Products Regulation (ESPR)?
• What Data Goes in a Digital Product Passport, and Where Does It Come From?
• Who Is Legally Responsible for a Digital Product Passport?
• The Digital Product Passport Standards Landscape
• How to Prepare Suppliers for Digital Product Passports
• How to Validate Digital Product Passport Data
• How to Manage Evidence for Digital Product Passports
• Digital Product Passport
• ESPR
• Working Plan
• Delegated Act
• Product Traceability
• Supply Chain Tier