Voxonomics
Methodology / Evidence

Blockchain data is not automatically economic evidence.

A ledger can tell you exactly what happened at protocol level. Economic measurement still has to establish scope, meaning, provenance, transformations and what the available evidence cannot support.

SourceProvenanceCanonical measurementReproducibility

Why raw blockchain data is not enough

The strength of public blockchains is that many underlying records can be independently observed. The difficulty is that an observable record does not automatically tell us its economic meaning.

A transfer can be a payment, a bridge movement, a self-transfer, an internal accounting operation or part of circular activity. A DEX trade can represent real price discovery or wash activity. A validator identity can be technically distinct while sharing common control with others.

Economic measurement therefore has to preserve the raw fact while separately documenting the rules used to interpret it.

The Voxonomics evidence chain

1
SourceThe chain record, node response, event log, governance record or approved external dataset.
2
Preserved raw evidenceThe source material is retained or referenced with enough identity and integrity information to identify what was actually observed.
3
ObservationThe source fact is represented in a structured form without silently changing its native meaning.
4
Canonical measurementThe observation is processed under the metric's declared definition, scope, units, exclusions, direction, horizon and transformation rules.
5
Published analytical outputOnly eligible measurements can flow into a parent index, VTS or other official output under the publication rules.

What provenance should answer

A useful provenance record should make it possible to answer questions such as:

  • Which network and canonical scope was measured?
  • What block, slot, time range or other observation boundary applies?
  • Which node, provider, indexer, contract or external dataset supplied the observation?
  • What was the raw value and unit?
  • Which extraction query, adapter or transformation version was used?
  • When was the evidence retrieved?
  • Was the observation complete, estimated, revised, disputed or anomalous?

Edition 3.1 formalises these requirements because a number without scope and lineage cannot be independently interpreted with confidence.

Canonical sources and source independence

Direct chain data is preferred where a fact can be computed deterministically, but not every economic property lives in a single RPC field. Open-source indexers, protocol APIs, public datasets and specialist providers can be appropriate where methodology and provenance remain sufficient.

Multiple endpoints do not necessarily mean independent evidence. Two providers can depend on the same upstream index or classification set. Voxonomics therefore distinguishes transport redundancy from methodological independence and cryptographic or deterministic verifiability.

Why scope causes so many errors

Cross-chain systems make double counting easy. A bridged asset can exist as an origin-chain lock and a destination-chain representation. A base layer and a rollup can both expose activity tied to the same economic flow. Ecosystem-wide aggregates are not directly comparable with a single-chain scope unless that difference is declared.

The canonical measurement object must therefore state exactly what execution, settlement and asset boundaries are included.

Unknown is not zero

One of the most important evidence rules is that unavailable or unobservable data cannot be silently converted into ordinary measured zero. Edition 3.1 distinguishes functional absence, technical equivalence through another adapter, temporary source failure, unobservable evidence and work not yet implemented.

Failing honestly is part of the measurement system. A missing observation should reduce coverage, confidence or publication eligibility where appropriate — not produce a convenient number that looks complete.

How corrections should work

Historical data can change because labels improve, providers revise records, chain reorganisations are reconciled or adapter defects are discovered. Voxonomics treats corrections as new versioned observations rather than silently overwriting prior history.

A meaningful correction record identifies the prior value, new value, reason, affected interval, source or adapter version and downstream impact.

Reproducibility

For a fixed evidence snapshot and methodology version, a third party should be able to reproduce the declared result. That requires more than a final CSV. The reproduction package needs source identities, adapter and framework versions, benchmark inputs, transformations, calculation configuration and deterministic handling of precision and rounding.

A cryptographic hash proves commitment to bytes. It does not by itself prove that the evidence is complete, economically relevant or correctly interpreted. Provenance and methodology remain necessary.

Why this matters for blockchain analytics

Dashboards often compete on speed and breadth. Voxonomics is aiming at a stricter question: can the economic claim be traced, challenged and reproduced? That requirement makes evidence architecture part of the product rather than hidden backend plumbing.

Read the primary methodology

See the 30 canonical submetrics, What Is Voxonomics? and the Edition 3.2 whitepaper, especially its data-source, provenance, adapter, missing-data and reproducibility sections.