Independent science. Public records. Animal by animal.
The methodology, data architecture, and governance beneath the verified low-carbon claim. Designed for the scrutiny that climate disclosure frameworks increasingly demand.
Bioeconomy Science Institute
The Life Cycle Assessment work beneath KaiProva’s verified low-carbon claim was commissioned with the Bioeconomy Science Institute (formerly AgResearch) and published as AgResearch report RE450/2024/054, Mazzetto et al. November 2024. The methodology covers 10–18 month young dairy beef finished on grass across 54 validated scenarios: 3 animal classes (steers, heifers, bulls) × 2 birth weights × 2 calving seasons × 9 ages × 3 average daily gain levels.
The result: a measured carbon footprint of 6–8 kg CO₂e per kilogram of beef, 32–48% below the New Zealand pastoral beef average reported by Beef + Lamb New Zealand (Class 9 average ~11.50 kg CO₂e/kg LW). Even at the protocol’s conservative floor — 10-month animals at 700 g/day — verified certificates measure 44% below that average. Stronger growth strengthens the certificate further.
Methodological frame: IPCC AR6 GWP100, IDF 2022 allocation, NZ GHG Inventory. Comparator dataset: Beef + Lamb New Zealand. KaiProva and Beef + Lamb NZ are partners in the Dairy Beef Opportunities Project.
Why averaging doesn’t work
Most carbon systems average animals across an industry. Good operators get lost in the average. Below-average operators benefit from a number they did not earn. The verified claim ends up being aggregate, indistinct, and easy to contest.
KaiProva measures animal by animal. Each enrolled animal’s carbon footprint is computed from its individual weigh history through the AgResearch LCA scenarios. The verified certificate carries that individual number through to the buyer.
This is what makes the verified claim defensible. Operators who run efficient pasture systems and get animals to processing earlier produce certificates that show it. The system rewards real practice, not averaged practice.
Biosecurity infrastructure as the chain anchor
The chain-of-custody behind every verified animal runs on the public livestock biosecurity infrastructure that already exists in each jurisdiction:
- NLIS — National Livestock Identification System (Australia)
- NAIT — National Animal Identification & Tracing (New Zealand)
- BCMS — British Cattle Movement Service (UK)
- USDA-ADT — Animal Disease Traceability (US)
- SISBOV — livestock identification (Brazil)
- ICAR-compliant systems globally
The tag in the animal’s ear is the receipt. The platform consumes identity from these systems but does not replace them. An auditor wanting to verify a specific movement record queries the public registry independently — the chain-of-custody is never the platform’s private record, it’s the public infrastructure.
This is the architectural commitment that lets the platform port across geographies as the standard scales. New geography = science commission + biosecurity integration, not platform rebuild.
How the data behind the claim is held
The methodology is only as trustworthy as the data architecture underneath it. KaiProva’s data posture is built to four principles:
Your data stays yours. Farmer data is scoped to the Organisation that owns it. No request, query, or background job ever reads or writes data scoped to one Org on behalf of a user who is not a member of that Org. Cross-Org reads are not permitted by design. When an animal moves to a new owner, the receiving Org sees the weigh history (date, liveweight, growth rate) but operator identity from the prior Org is stripped; only auditors, processors and buyers with appropriate platform access see the full chain unfiltered.
Append-only audit log. Every verification-anchored event — enrolment, weighing, transfer between Orgs, lapse, re-enrolment, kill closure, claim issuance — is written as an immutable audit row. Never updated, never deleted. If a correction is needed, a new row records the correction; the original stays. Retention is indefinite for verification-anchored events; statutory minimums apply elsewhere.
HMAC-signed partner integration. All inbound partner data (kill records from processors, weigh data from sensor partners, allocation feeds from dairy companies) is signature-verified before any processing. Unsigned data does not enter the verification record. Partners who won’t sign don’t get a webhook endpoint — we ingest by other means.
Public-records anchoring as the source of truth. See §3 — the chain rides on biosecurity registries, not the platform’s private record. The platform consumes identity, not movement.
Operator data stays with the operator
The data architecture honours producer data sovereignty in three ways. Farmer data is held by the Organisation that owns it, not aggregated into a platform-owned dataset. The chain-of-custody runs on public biosecurity records that are independently queryable, not on a private registry. And data is named alongside animals, land, and claims as something the standard protects in its own right.
This sits alongside the relational articulation in our kaupapa — the Māori data sovereignty principles (mana motuhake o ngā raraunga) the same commitment is grounded in. Both layers run beneath the platform: read our kaupapa →
What we align with
The methodology and reporting framing is designed to align with the major frameworks shaping global procurement and climate disclosure:
- GHG Protocol Scope 3 Livestock Sector Guidance — intensity-based reporting (kg CO₂e per kg of product)
- ISO 14067:2018 — carbon footprint of products, methodology principles
- SBTi FLAG Guidance — Forest, Land and Agriculture targets and reporting framework
The verified certificate carries this alignment statement explicitly. Procurement standards teams, CSRD reviewers and FLAG-aligned buyers can read the certificate against the frameworks they report under and see the structure they expect.
Multi-stakeholder from inception
A verification standard’s independence is its most load-bearing claim. KaiProva is built on the model that took Bonsucro, FSC and RSPO to durable global scale: a defined standard that participants conform to, governed multi-stakeholder rather than by the company that operates it.
From inception, KaiProva is governed by a founding board drawn from pilot participants — dairy company, processor, buyer-pull integrator, science partner, producing operator, independent seat, with KaiProva holding a single seat. From Year 3, the governance moves to a formal multi-stakeholder commission structure with an independent secretariat, rotating chair, and public annual report on standard conformance. KaiProva retains the operating licence from the commission — structurally and legally distinct from the standard itself.
That is the model that produces a defensible answer to "who controls the standard?" for a global buyer’s procurement standards team. Without it, a verification platform is a proprietary certification programme operated for the benefit of one company. With it, the standard is independent of its operator by design.
Honesty about scope
The methodology aligns with the global frameworks named above. It is not certified by them. Alignment means the methodology’s structure follows the framework’s structure; certification requires a formal audit by the framework body, which has not happened. We say this plainly on the verified certificate and in every external artefact.
The Bioeconomy Science Institute LCA is commissioned research, not peer-reviewed journal publication. Commissioned research is funded by a specific client (Alps2Ocean Foods and Fonterra) and produced under contract; peer review is independent verification of the methodology by an external academic panel. We commission the science with rigour, but we don’t describe it as peer-reviewed because it is not.
The platform’s broader compliance posture — SOC 2, ISO 27001 — is not formally pursued in 2026 but is being designed with control shapes that align with those audits, with a view to pursuing them as the platform scales. We operate as if pursuing the standards by 2028 even though we do not yet claim them.
These distinctions matter. A wrong word on a buyer-facing certificate can invalidate a premium claim. We choose the honest word every time — alignment, not certification; commissioned research, not peer-reviewed; designed to align with, not compliant with. The verified claim is stronger because of the discipline, not weaker.