Ask who does carbon projects in Indonesia and you get a list of organisations. That is the wrong unit of analysis. There are five distinct functions in a carbon project, some organisations perform several, and the combinations that are permitted are not the combinations that are common.
The Five Functions
| Role | What it does | Can it combine with others? |
|---|---|---|
| Project proponent | Holds rights, bears risk, receives credits | Yes, with developer |
| Developer | Designs the project, writes the documents | Yes, with proponent |
| Measurement provider | Produces the inventory and monitoring data | Yes, but see below |
| Validation/verification body | Independently assesses | **No** — must be independent of all above |
| Registry | Issues and tracks units | No |
The hard line is the fourth row. A body that helped design a project cannot verify it, and this is where developers most often create a problem for themselves without noticing, by engaging one firm for advisory work and later discovering its verification arm is conflicted out.
Where the Confusion Actually Causes Damage
Proponent and developer merged, with no separate measurement. Common in smaller projects, and it means the party whose revenue depends on the tonnage is also the party producing it. Not prohibited, and it puts the entire evidentiary burden on the verifier, who has limited time and a sampling-based check.
Community as nominal proponent, developer in control. A structure where the community formally holds rights but has no practical say and a small revenue share. It satisfies the paperwork and fails the substance, and it is the structure most likely to collapse mid-crediting-period when the community understands the terms.
Buyer financing development. Legitimate and common, and it needs disclosure. A buyer funding project design has an interest in the tonnage, and the arrangement should be visible rather than discovered.
The Function That Is Thinnest Here
Independent measurement.
Indonesia has proponents with land, developers who can write documentation, accredited verifiers, and registries. What it has least of is organisations whose job is to produce the underlying numbers and who have no stake in how large those numbers turn out.
In most projects, measurement sits inside the developer. The developer is not dishonest; the incentive is simply pointed one direction, and the only counterweight is a verifier checking a sample, years later, against records the developer assembled.
Why This Is Becoming More Visible
Two pressures are converging. Buyers are moving past methodology labels and asking about measurement quality directly, because the label no longer differentiates. And regulators tightening registry requirements are asking for traceability that casually-kept records cannot supply.
Both push toward the same structural answer: measurement that can be traced to a party that did not need the answer to come out a particular way.
What a Proponent Should Take From This
Before signing anything, map who is performing which of the five functions, and write it down. Then check two things: that nobody is performing the verification function alongside any other, and that the party producing the measurements has a documented protocol somebody else could audit.
Projects that can produce that map in one page tend to move through validation without structural surprises. Projects that cannot usually discover the conflict at the point where fixing it is most expensive.
Selling into a compliance market?
Compliance buyers examine sampling design, uncertainty and data traceability directly, and a later finding of ineligibility returns the liability to them. TREEO builds the evidence file that answers it.
