# Should we round our carbon estimates up to be safe?

Canonical: https://senseible.earth/answers/should-carbon-estimates-be-conservative

Not for an inventory. The GHG Protocol asks for accuracy: no systematic overstatement or understatement. Use the best estimate and state the uncertainty. Conservativeness is a principle for carbon credit projects under ISO 14064-2, where overclaiming reductions is the risk. Never inflate a baseline to show a bigger cut later.

## Where to be careful

- Missing months: estimate from similar months and label it.
- Unknown factors: use a recognised default and cite it.
- Reduction claims: understate rather than overstate.

## What verifiers want

A clear note of each estimate, the method and the source. A padded figure is as wrong as a low one.

## What Senseible does here

Senseible keeps each record linked to its source document and marks estimates as estimates, so a verifier can see what was measured and what was assumed.

- [Estimation note for the verifier](https://senseible.earth/answers/estimation-note-for-verifier)
- [Inventory uncertainty](https://senseible.earth/answers/ghg-inventory-uncertainty-calculation)

Sources: [GHG Protocol Corporate Standard](https://ghgprotocol.org/corporate-standard), [ISO](https://www.iso.org)

Last reviewed 2026-10-23.