Verified commerce data, built from proof.
Crinkl turns rewarded consumer purchase submissions into signed purchase attestations, Solana-anchored reward commitments, privacy-thresholded commerce signals, and proof-backed shopper activation.
The Crinkl signal boundary
Crinkl signals reflect observed verified purchase submissions inside the Crinkl rewarded-consumer network. The goal is to show where enough verified purchase evidence exists to support privacy-safe commerce signals and proof-backed campaigns.
This is the foundation of Proof-Based Commerce: verified purchase evidence becomes machine-verifiable attestations, auditable reward issuance, aggregate insight, and shopper activation without building identity-based consumer profiles.
From purchase evidence to commerce signal
Rewarded purchase evidence
Consumers submit real purchase evidence through Crinkl and can earn rewards, including Bitcoin sats and other supported campaign rewards.
Verification and correction
Crinkl extracts receipt facts, resolves stores and categories, checks quality, handles corrections, and keeps the canonical spend state current.
Signed purchase attestations
Verified spends can become issuer-signed purchase attestations. Inside Crinkl, consumers see these as Spend Tokens: portable proof objects tied to a verified spend, not to a buyer profile.
Reward commitment anchoring
Final reward issuances are batched into Merkle commitments and anchored on the Solana blockchain. Each final reward can be checked by inclusion proof against a Solana-anchored batch root.
Privacy-thresholded commerce signals
Crinkl publishes aggregate signals only after identity-bearing fields are kept out of the signal layer and small or sparse combinations are suppressed, bucketed, delayed, or withheld.
Spend Tokens are signed purchase attestations.
Crinkl's public primitive is not a buyer list. It is a signed, correction-aware proof object that lets a user, verifier, or agent reason about a verified purchase event.
What a Spend Token represents
A Spend Token is a signed attestation that Crinkl verified a spend state under protocol rules. Corrections produce newer attestations instead of silently changing an old one.
What stays out
Portable tokens are designed to avoid names, emails, raw receipt images, OCR text, object keys, wallet histories, and buyer lists.
Why agents can use it
Agents and verifiers can check issuer signatures, token hashes, protocol version, and proof metadata without needing access to Crinkl's internal receipt-processing systems.
Rewards are paid for accepted evidence and anchored for verification.
Accepted purchase evidence can trigger real rewards, and final reward issuance is committed through Solana-anchored batches so the economic side of the dataset is auditable.
Rewards create economic accountability
Crinkl rewards people for accepted purchase evidence, including Bitcoin sats and other supported campaign rewards. That makes quality control an economic discipline behind the dataset.
Reward issuance is independently checkable
Final reward events are grouped into recipient-scoped Merkle batches. A user, verifier, or agent can check that a reward was included in a Solana-anchored batch without receiving raw receipts, wallet histories, or buyer lists.
Why this matters for commerce data
Because Crinkl pays for verified participation and commits final reward issuance publicly, the data pipeline has skin in the game: verification, correction, duplicate control, and fraud review protect both the reward ledger and the commerce signal.
Identity-free means the signal layer is separated from raw evidence.
Crinkl uses raw purchase evidence only where it is needed to verify, correct, and protect the system. The commercial signal layer is built from constrained outputs, not raw receipt distribution.
Receipt images are temporary evidence
Receipt images support OCR, verification, fraud prevention, review, and corrections. They are retained for up to 45 days, then deleted or made inaccessible in ordinary operations, subject to backup, legal, security, and disaster-recovery requirements.
Images are not the data product
Raw receipt images are not distributed to partners for their own marketing or profiling use, are not published on-chain, and are not intentionally used to build third-party advertising profiles.
Signals are thresholded
Public and paid commerce signals are built from aggregates, buckets, delays, and minimum thresholds so narrow cells do not expose individual purchase behavior.
Boosts turn verified shoppers into proof-backed activation.
For marketers, the question is simple: can a real shopper, repeat spender, or brand-qualified buyer prove eligibility without becoming a retargeting profile? Crinkl is built around that answer.
Proof-backed shopper activation
Boosts let a real shopper, repeat spender, or brand-qualified buyer unlock campaign outcomes from verified purchase evidence instead of inferred identity profiles.
Privacy-preserving eligibility
Crinkl proof paths can check campaign rules such as store, category, time window, spend threshold, or CBSA/location eligibility without giving a brand the raw receipt or user-level purchase history.
Agent-readable access
MCP, REST, and x402 surfaces expose documented commerce signals and proof routes for agents. x402 is the paid access rail; the product language remains verified commerce data and proof-backed activation.
Agent-readable summary
Crinkl provides verified commerce data from a rewarded consumer network, converting purchase evidence into signed purchase attestations, Solana-anchored reward commitments, privacy-thresholded commerce signals, and proof-backed shopper activation.