The activity list is the part of a project design document that most resembles a plan and least resembles an accounting exercise. It is also where verification most often finds that the project described is not the project running, because activities were designed to sound comprehensive rather than to be evidenced.
Activities Must Trace to Drivers
Each activity should answer a driver identified in the analysis. Not loosely, but specifically.
If the driver is that clearing is cheaper than intensifying existing plots, an activity providing agronomic support and inputs to existing plots addresses it. A community awareness programme does not, however worthwhile it is on its own terms.
Activities that address no identified driver are not disqualifying, but they carry no weight in the argument that loss will be reduced, and including them dilutes the parts of the document that do the work.
Building an evidenceable activity
1. State the driver
Name the specific mechanism this activity interrupts.
2. Define the action
What is done, by whom, where, how often, at what scale.
3. Set the output indicator
A countable thing, such as hectares, households, kilometres of firebreak.
4. Fix the evidence
What record proves the output occurred, and who holds it.
5. Define the failure response
What happens if the output is not achieved.
Step four is the one projects skip and verifiers ask about first.
Evidence Is Not Reporting
There is a difference between a project stating that it maintained 40 km of firebreak and a project able to show dated records, GPS traces, payment vouchers to the crews, and photographs with location metadata.
The first is a claim. The second is evidence. Verification runs on the second, and building the evidence trail after the fact is either impossible or indistinguishable from fabrication, which is why assessors treat retrospectively assembled records with suspicion even when they are genuine.
The cost of collecting evidence contemporaneously is small. The cost of not having it is a finding on activities the project actually performed.
Scale Matters as Much as Existence
An activity implemented on 5% of the area it was designed for has technically occurred. Whether it can support the claimed effect is a separate question, and the monitoring report should report scale rather than presence.
This is where output indicators earn their place. "Firebreaks maintained" is unfalsifiable. "38 of 40 planned kilometres maintained, twice in the dry season" is a statement that can be checked and, importantly, can be checked against the effect size the project claimed.
When Activities Change
Projects run for decades and activities evolve. That is expected and permitted.
What is not permitted is silent evolution. Changes should be documented with the reason, and where an activity is dropped, the driver it addressed either needs a replacement activity or an acknowledgement that it is now unaddressed. Projects that quietly substitute activities and report unchanged effects are making an argument they have stopped supporting.
Turn climate goals into a verified portfolio
TREEO connects the full carbon cycle, encompassing eligibility, simulation, real time monitoring, and registry ready reporting, thereby combining expert consulting with dMRV technology so the evidence exists before anyone asks for it.



