When a Verra grouped project has hundreds of sites, does the verifier check every one?
Not necessarily. Where the number of instances makes individual checks unreasonable, the verifier documents statistically sound evidence-gathering methods. The number of instances included and eligible for credits shall be proportional to the percentage of instances sampled and found to conform to the Verified Carbon Standard (VCS) rules.
What the Standard says (sections 4.1.30 and 4.1.31)
The proportional rule is in 4.1.30(2), for instances being validated. Section 4.1.31 adds the statistically sound and documentation duties for the verification report.
- Where an individual assessment of each initial or new instance is unreasonable because of numbers, the validation and verification body (VVB) shall document and explain its evidence-gathering methods.
- The methods shall be statistically sound.
- The number of instances included in the project and eligible for monitoring and for credits shall be proportional to the percentage of instances sampled by the VVB and found to conform.
- The verification report shall explain the evidence-gathering methods for the greenhouse gas (GHG) statement of reductions, and any changes forced by findings shall be documented.
Reading the proportion rule
Our reading: the Standard does not give a formula or a sample size. It says the eligible count follows the conformance rate in the sample. The verification body chooses the method, so ask the VVB in advance how it plans to sample and what it will do if sites fail.
What this means for an association
- Weak sites cost everybody: failures in the sample lower the number of instances that can earn credits.
- Records matter for every site, since any site can fall in the sample.
- Fixing records before the audit is cheaper than a lower conformance rate after it.
Example of the proportion, our reading
- A project proposes 1,000 new instances. The VVB samples 100 of them (10%).
- 90 of the 100 sampled conform, a rate of 90 / 100 = 90%.
- On a proportional reading, 90% of 1,000 = 900 instances would be included and eligible, and 1,000 less 900 = 100 would not.
Numbers are an example. The Standard sets no sample size or formula, so the real method is the verification body's.
What Senseible does here
Senseible reads fuel invoices and electricity bills into monthly records that you can keep with your project evidence. Its in-house check is not an accredited third-party verification, and Senseible is not part of Verra's process.
Sources: Verra, VCS Standard v5.0 (16 December 2025)
Last reviewed 2026-11-05. Check the current VCS Standard, Registration and Issuance Process and Program Guide on Verra's site before you rely on this; Verra updates them periodically.