The EU value-chain ceiling limits what large customers may require from suppliers of up to 1,000 employees. Directive (EU) 2026/470 applies to financial years beginning on or after 1 January 2027.The EU value-chain ceilingSee what applies to you →

Four states between a spreadsheet and a report you can defend.

Rows move from the file you already have to a filed figure through gates that are allowed to refuse. The ledger does the remembering, so next year is an update rather than a rebuild.

1 of 4

Import

Spreadsheets and CSVs in, mapped once per source and remembered. Rows land in the ledger and stay out of the totals until you accept them.

  • Every rejected row gets a named reason rather than a generic failure: a required field missing, an unknown unit, a period already locked, a date outside the period, a site code that matches nothing.
  • Every rejected row stays on screen with its reason on every plan. On a paid plan the error report also downloads verbatim, so you can send it back to whoever produced the file instead of retyping the problem.
  • Duplicate suspects are flagged in the ledger beside last year’s figure, not on a separate screen.
  • A one-million-row file read in nineteen minutes on a developer machine: 950,000 rows imported and 50,000 quarantined by design, with every cap holding and announcing itself rather than truncating quietly.

2 of 4

Resolve

Every row is matched to a factor from a licensed, versioned publisher, and the resolver records why it chose that one.

  • You see which factor, which dataset version, which valid window and which licence.
  • The resolver records its own deliberation: what decided it, which region level it used, how many candidates it weighed and the runners-up it set aside. We know of no other tool that keeps that record. The lineage drawer surfaces the decision today; the runners-up are recorded and the screen for them is still being built.
  • A factor outside the reporting period is never used quietly. It is excluded before ranking, and if nothing in period remains the row fails with that named reason rather than silently borrowing last year’s number.
  • The snapshot is frozen with the figure, so next year’s library cannot move this year’s number.
  • AR4, AR5 or AR6 selectable, with per-gas masses stored natively wherever the publisher provides them.

3 of 4

Review

The completeness register tells you what you still owe, generated from your own sites, declarations and assets rather than hand-maintained.

  • It attaches rows that already exist to the gap they fill, so the list shrinks as you work.
  • The same total is shown split by what stands behind each figure, in the product’s own five tiers: primary, supplier-reported, activity-based, spend-based, estimated.
  • What we cannot calculate for your country yet is marked as our gap and counted separately, so it is never mistaken for something you failed to supply.
  • Review is a filter on the ledger, not a second screen, because reviewing means reading rows in context.

4 of 4

Publish

Pre-flight separates blockers from advisories and links each to its fix. Filing writes an immutable version.

  • A blocker stops the filing. An advisory is disclosed on the report rather than hidden. Each links to the screen that fixes it where such a screen exists, and says so plainly where none does.
  • Re-filing writes version N+1 under the same report; the version it replaces stays readable at its own number.
  • Locking a period refuses backdated edits with an explanation rather than a generic error.
  • GRI 305 and ESRS E1 from one dataset, with stated omissions printed on the face of the report.

Coverage, stated including the gaps

  • Emission factors from 11 named public sources, every licence recorded and cited on the report. 3 more sit in the registry and are deliberately not served: one is not licensed to us, one is under a licence review, and one says nothing about redistribution at all.
  • Countries you can open an account for: Australia, France, Germany, New Zealand, United Kingdom, United States. The list is generated from the factor library rather than curated, and a country qualifies only when it can supply grid electricity, natural gas, diesel. Registration is refused elsewhere rather than producing a figure we cannot stand behind.
  • Scope 3: 6 of 15 categories modelled, each naming its method. The other 9 print as stated omissions.
  • Market-based Scope 2 falls to a disclosed grid fallback, because the residual mix dataset is not licensed to us.
  • Unlimited users on every tier including the free one. Auditor access does not consume a seat.

How the calculation works, in detail