Foundation

Homegrown intelligence for scan + evidence integrity.

Structured analysis that turns raw capture into durable, auditable records without expanding the blast radius.

Ingestion decision engine

Server-side rules that hard-gate evidence ingest without slowing operators.

Require client_id

Require program_id

Require visit_id

Require evidence_hash

Max evidence size (bytes)

Allowed categories (comma-separated)

Allowed scopes (comma-separated)

Rules load from the AXIS API.

Descriptive dashboards

Scan density, compliance, and exception frequency per program/client.

—

Diagnostic + prescriptive

Root causes, anomalies, and next-best-actions with evidence anchors.

—

Feature flags

Enable intelligence per client/program. Default off.

Client

Program

Intelligence enabled

Scoped flags are stored locally; wire to API to persist.

Signal inputs

Normalized inputs across capture and verification.

• Barcode, QR, NFC payloads

• Evidence metadata (no raw image storage by default)

• Operator confirmations + exceptions

• Client, program, store, and visit scopes

PBNA priority intelligence

Preference toward PBNA segments and execution outcomes.

• CSD, Energy, Hydration segment coverage

• Merchandising compliance and display verification

• Store-level timelines and OOS risk signals

• SLA and delivery variance tracking

Core analytics

Detection and enrichment engines.

• UPC/SKU normalization + catalog matching

• Anomaly detection (mismatch, velocity, scope)

• Shelf intelligence summaries (facings, alignment, cleanliness)

• Confidence scoring + review flags

Output artifacts

Human + machine readable outputs.

• Signed ScanEvents + EvidenceEvents

• Structured PDFs (barcodes/QR rendered)

• At-risk flags and mitigation logs

Guardrails

Intelligence that respects C2 policy.

• Append-only records

• Scope required on every request

• Quarantine only on failed analysis

References

Technical baselines for intelligence layer buildout.

Scan analytics spec Capture policy Intelligence plan