# Where Does a Product Fact Become True?

Source: https://activatedigital.ai/knowledge/evidence/where-product-facts-become-true
Last verified: 10 September 2026
Summary: A practical method for working out where a product fact becomes true, who holds it, what evidence establishes it and whether a downstream business can reuse it.

## Direct answer

A product fact can start with a business decision, a production event, a measurement or a rule. Before you chase suppliers or choose a database, separate where the fact becomes true from who holds it, what evidence establishes it and whether the same fact can be established another way.

For each fact you need downstream, ask five questions in order:

- **Where does the fact become true?** Identify the decision, event, measurement or determination that makes the value knowable.
- **Who usually holds it now?** The current holder may be the originator, but it may equally be a supplier, laboratory, distributor or downstream business.
- **What evidence object establishes it?** Name the actual record: a controlled product master, production record, label, test report, declaration or another defined object.
- **What does that evidence cover?** Check the subject and scope, such as model, variant, batch, item, facility, date, method or legal entity.
- **Can the fact be established another way?** Some facts can be measured or determined again. Others depend on a production event or upstream record that cannot safely be reconstructed downstream.
Keep those questions separate from the internal system-of-record decision. A business can govern an approved value in a PIM, ERP or compliance system without that system being where the fact originated or the evidence that proves it.

**Important distinction:** on this page, **fact origin** means the point, process or determination at which a value becomes knowable. It does not mean customs country of origin.

The practical model is:

**Fact becomes true → evidence object → governed answer → downstream reuse**

At every handoff, the identity and scope have to stay attached to the value.

## Fact origin, holder and system of record are different jobs

A common failure is to ask, “Which supplier has this field?” and treat the answer as proof of where the fact came from. That shortcut confuses several different roles.

| Role | What it means | What it does not prove |
|---|---|---|
| **Originator** | The actor or process at which the fact is first assigned, measured, recorded or determined | That the originator still holds the evidence |
| **Current holder** | The organisation that currently has the value or supporting record | That it created the fact or is the best authority for every use |
| **Internal system of record** | The governed place where a business maintains its approved value | That the system itself establishes the underlying real-world claim |
| **Downstream publisher** | The system or channel that exposes the approved representation | That the published copy still carries the evidence, scope or provenance behind it |

NIST’s manufacturing traceability meta-framework is useful here because it is designed around organising, linking and querying verifiable traceability data across disparate manufacturing ecosystems. It does not require one central repository. OECD work on lithium and nickel supply chains reaches a compatible practical conclusion from another direction: real traceability can be partial, heterogeneous and assembled from different records and actors.

That is why “source of truth” is too vague on its own. A supplier portal can be where you receive a fact. A PIM can be where you govern it. Neither answer tells you where the fact became true.

Once the fact-origin question is settled, the next internal governance decision belongs on the separate guide to [which internal system should own the fact](https://activatedigital.ai/knowledge/product-data/which-system-owns-each-fact).

## Evidence object and scope have to travel together

A value without scope is rarely enough.

“Factory: Porto” might refer to the site that made the finished item, the site that made a component, a supplier’s registered office or a facility used in an earlier season. “Weight: 420 g” might mean net product weight, packaged weight or a different variant. A test report may be authentic and still describe the wrong model, batch, sample or method.

The evidence test is therefore not simply “Do we have a document?” It is:

- **subject:** what product, component, material, company or site does the evidence describe?
- **granularity:** is it model, variant, batch, item, facility or another level?
- **time:** when was the value established and does the relevant version still apply?
- **method or basis:** was the value assigned, measured, calculated, declared or determined under a rule?
- **identifier join:** can the evidence be joined unambiguously to the downstream record that will carry the claim?
ESPR itself illustrates why granularity matters. The framework allows product-passport information to operate at model, batch or item level depending on the product-specific rules. The Batteries Regulation also distinguishes information about a battery model from information tied to an individual battery. Those are legal examples from particular regimes, not a universal rule for every product fact.

**Evidence rule:** a genuine document can still be the wrong evidence for the assertion you are making if its subject, scope or identifier join does not match.

That is also why a signed or machine-readable artefact does not prove the physical fact by itself. It can prove who issued a statement and whether the artefact has been altered, but the downstream claim still depends on correct binding and an appropriate evidence scope.

## Where the fact becomes true tells you where to start, not how far to go

Locating the fact-origin point narrows the search. It does not automatically answer how much of the chain a particular legal, assurance or due-diligence task needs to cover.

A batch number, for example, becomes a usable fact when a production or packing process assigns units to a lot. A production site becomes relevant when the product or batch is actually allocated to and made at that facility. Once you know that, you know the kind of record and actor you need to look for. You have not yet decided every upstream evidence requirement around the product.

That distinction matters because due-diligence regimes can require chain information even where one product characteristic could be measured or checked another way. EUDR is a useful comparator: it has its own information, geolocation and risk-based due-diligence requirements. It is not a DPP regulation and its requirements should not be generalised to unrelated product categories.

If the next decision is [how far upstream the fact must go](https://activatedigital.ai/knowledge/guides/how-far-back), use the upstream-depth owner rather than turning the fact-origin question into a universal chain-depth rule.

## Can the fact be established another way?

Sometimes yes. The important word is **same**.

An alternative route is useful only if it establishes the same fact, for the same subject and scope, to the standard needed for the downstream use. A different piece of evidence that merely looks plausible is not an alternative route.

A practical way to sort facts is:

| Fact type | Typical origin | Can a downstream party establish it again? | Main caution |
|---|---|---|---|
| **Assigned identity** | A manufacturer or business assigns the model or trade-item identity | Usually not by inventing a replacement. The identity may be observed on the product or documents, then checked against controlled records | Renumbering, private-label remapping and reused model names can break the join |
| **Production lot** | Units are assigned to a batch or lot during production or packing | Usually not if the original lot relationship was never preserved | Repacking, aggregation and relabelling can destroy the historical relationship |
| **Measured product characteristic** | The relevant product configuration is measured | Often yes, if the same definition, unit and configuration can be measured again | Net, gross, packaged and variant-specific values are different scopes |
| **Test result** | A stated sample is tested under stated conditions and method | A new test can establish a new result. It does not retroactively prove what happened in the original test | Sample identity, representativeness, method and date matter |
| **Production-location record** | Production is allocated to and carried out at a facility | Usually only through records that bind the product or batch to the site | A company address, supplier relationship or certificate with wider scope is not automatically site evidence |

This distinction stops two opposite mistakes. The first is chasing suppliers for a fact that can be established directly with better evidence. The second is trying to infer a process-held fact from the finished product when the decisive record only existed upstream.

This page does not decide the SME practicality of each route. For that, use the separate guide to [what a downstream business can observe itself](https://activatedigital.ai/knowledge/evidence/small-business-product-fact-observability).

## Worked fact-origin examples

The framework becomes clearer when the roles are kept separate.

### Product model identity

A model identifier becomes true when the responsible business defines and controls that model identity. The usual holder is the manufacturer or brand owner, although an importer or private-label operator may control the commercial identity in some arrangements. Evidence can include the controlled product master, technical documentation and the identifiers used on packaging or catalogues.

The downstream job is not to create a new identity because the original one is inconvenient. It is to preserve the join between the identifier, the relevant revision and the product being described. ESPR and the Batteries Regulation both show why product identity matters in passport architectures, but the detailed identifier requirements remain product-specific.

### Batch or lot identity

A batch or lot becomes true when a production or packing process assigns units to that lot. The manufacturer or packer will usually hold the production record. A downstream business may also hold the batch number on a label, delivery record, ERP or WMS record.

Those downstream copies are useful only while the batch relationship survives. If goods are repacked, split, aggregated or relabelled without preserving the link, the historical fact cannot safely be recreated from a plausible-looking number later.

### Product weight

Weight is a good example of a fact that can often be established by more than one route. A manufacturer may hold a specification or weighing record, but a downstream party can also measure the product again.

The alternative route works only when the definition is the same. A fresh measurement of net product weight does not validate an old value that meant shipping weight. Nor does a measurement of one variant establish every variant.

### Test result

A test result becomes true when an identified sample is tested under a stated method and conditions. The test report is the evidence object. Its scope is the sample, product or batch actually covered, together with the method, date and relevant conditions.

Repeating the test can be a legitimate alternative route, but it creates new evidence. It does not prove that an earlier report covered a different batch or that the earlier sample was representative.

### Regime-specific geolocation information

EUDR shows a different kind of problem. It requires defined information and geolocation for relevant commodities and products within that regime, alongside due-diligence steps. That information depends on a defensible link to where production occurred. It is not something a downstream business should infer from a finished product simply because a supplier, warehouse or corporate address is known.

The lesson is not that every product needs EUDR-style data. It is that a fact can be process-held and legally scoped in a way that makes the original evidence relationship decisive.

## Downstream reuse gate

A fact is reusable downstream when the evidence still supports the assertion you are about to make. Before copying a value into another system, file or passport, run six checks:

- **Same fact:** are both sides using the same definition?
- **Same subject:** does the evidence describe the product, material, site or company you are asserting about?
- **Compatible scope:** does model, variant, batch, item, facility and time coverage fit the downstream claim?
- **Working identity join:** can you connect the evidence to the target record without inference?
- **Evidence still applicable:** is the source current for the relevant revision, method and period?
- **No unsupported strengthening:** are you reusing what the evidence says rather than turning it into a broader claim?
If all six pass, reuse is usually a governance and transformation problem rather than a fresh evidence-establishment job. If one fails, the right action may be to get stronger evidence, establish the fact by a legitimate alternative route or leave it unestablished.

This is where the fact-origin question hands over. Once the value, subject, scope and join are sound, the next question is [how the evidence should travel downstream](https://activatedigital.ai/knowledge/evidence/how-product-evidence-travels-downstream) without losing those properties.

There is no universal registry that resolves every product fact to one authoritative holder. The practical answer is therefore not to nominate one universal source. It is to make the origin, evidence object, scope and join explicit for each fact that matters.

## What would change this page

The framework would need revisiting if a product regime or shared infrastructure materially standardised authoritative fact origin, evidence or reusable proof across product categories. Product-specific rules can already change the answer for particular facts, so examples should be rechecked when their governing regime changes.

## Sources

- [Regulation (EU) 2024/1781 establishing a framework for ecodesign requirements for sustainable products (ESPR), European Union](https://eur-lex.europa.eu/eli/reg/2024/1781/oj/eng) EU law
- [Enhancing Resilience Through Traceability: Insights from Lithium and Nickel Supply Chains in Latin America and Southeast Asia, OECD](https://www.oecd.org/en/publications/enhancing-resilience-through-traceability_a898687d-en.html)
- [NIST IR 8536: Supply Chain Traceability: Manufacturing Meta-Framework, NIST](https://csrc.nist.gov/pubs/ir/8536/final)
- [Regulation (EU) 2023/1542 concerning batteries and waste batteries, European Union](https://eur-lex.europa.eu/eli/reg/2023/1542/oj) EU law
- [Regulation (EU) 2023/1115 on deforestation-free products (EUDR), European Union](https://eur-lex.europa.eu/eli/reg/2023/1115/oj)