Skip to main content

Data Flow

A data flow is a naming system plus a state machine plus a maturity map. All three are required before a device becomes an economic actor.

Naming System

Who named each thing shapes who captures the value.

TermWho Coined ItWhat the Name HidesPower Concentration
"Smart device"Consumer electronicsWhether it signs, earns, or can be owned by its userOEM
"Hotspot"HeliumThat it is a token-mining node with routing dutyProtocol operator
"Interrogator"RFID industryThat it is a notary of physical eventsWarehouse or integrator
"Gateway"TelecomThat it is a signing authority and time witnessTelco
"Digital twin"Industrial softwareWhether the twin is cryptographically boundEnterprise software vendor
"Sensor network"Academic IoTWhether readings are signed or tradeablePlatform aggregator
"Machine identity"Peaq / DePINThat a DID is the machine's wallet addressProtocol

The naming gap: the industry does not have a shared term for "signed event from an attested reader." Every vendor calls it something different. Until that term is agreed, audits stay manual.

Data Model

Entities, relationships, state transitions.

EntityRelationshipsState Transitions
Asset1:N tags, 1:N sensors, 1:1 custodian at a timeCreated → custodied → transferred → retired
Tag1:1 asset, N:1 family, readable by many interrogatorsIssued → activated → read → revoked
Sensor1:1 asset or 1:1 environment, emits readingsCommissioned → streaming → drifted → replaced
InterrogatorN:1 operator, deployed at 1 location, reads N tagsDeployed → operating → tampered → retired
Edge Signer1:1 interrogator or gateway, holds a machine DIDProvisioned → signing → compromised → rotated
Event1:1 signer, 1:N tags/readings, timestampedCaptured → signed → anchored → settled
OperatorN:1 network, owns N devices, earns per eventOnboarded → active → slashed → paid
Settlement1:N events, triggered by a smart contractPending → executed → finalised

Data Footprint

Maturity across the five layers for each major domain. 1 = schema only, 5 = closed feedback loop.

Data DomainSchemaDataAPIUIFeedbackMoat Signal
Asset identity (RFID)44332GS1 schema exists, settlement thin
Environmental sensing34333Many vendors, no common API
Custody events22111Almost no standard — greenfield
Machine identity (DID)43322Peaq leads, adoption early
Positioning (RTK)44443GEODNET mature, feedback improving
Settlement proofs33322Chain-side solid, UI side weak

Where moats live: Custody events (score 1.4 avg) is the widest opportunity. Nobody owns the schema for "signed custody handoff." First mover wins.

Decisions Data Drives

Good data turns tacit judgements into automated decisions.

DataDecisionActionImpact of Bad Data
Temperature logAccept or reject a wine shipmentRelease payment, or trigger claimHealthy cargo rejected, spoiled accepted
Custody chainAuthenticate provenanceMint NFT, or refuse listingCounterfeit enters secondary market
Device attestationTrust a readingAccept event, or quarantineOracle manipulation, systemic fraud
Event timestampOrder handoffs across bordersSettle cross-border dutyTax disputes, customs holds
Operator uptimePay network rewardsDistribute tokensFree riders, honest operators leave

Context

Questions

Which data domain, if standardised tomorrow, would unlock the most venture opportunity?

  • Why does custody events sit at 1.4/5 when every enterprise wants it?
  • What would a GS1-equivalent standard for signed custody handoffs look like?