.png)
Amperity Integration Guide
Integrate Amperity customer data, audiences, batch operations, and file exchanges with enterprise applications through APIs and orchestrated workflows.
Amperity integration options at a glance
Amperity supports API-based integration for interacting with tenant services and customer data, while batch-oriented ingestion, exports, activation, and file exchange are important for large customer and transaction datasets. Martini can consume documented Amperity REST APIs, transform request and response payloads, orchestrate long-running jobs, and poll for completion on a schedule. File workflows can process supported customer, transaction, product, and audience formats once the tenant-specific contract is confirmed. Operation-specific callbacks may be available, but Amperity should not be treated as providing a universal webhook stream. Credentials and destination configuration remain environment-specific and should be stored securely.
Common Amperity integration patterns
Common Amperity data objects used in integrations
Authentication and security considerations
Tenant-specific authentication
Amperity authentication can vary by API and deployment. Confirm whether the tenant uses API credentials, API keys, bearer tokens, scopes, and permission settings before implementation.
Secrets and permissions
Martini can store environment-specific credentials in secrets and apply them to API calls without hard-coding sensitive values. Restrict access to customer profiles, audiences, queries, exports, and destinations according to least-privilege requirements.
Privacy and consent
Customer and transaction data may contain personal information. Define permitted fields, consent and opt-out propagation, retention, regional processing, access logging, and downstream deletion or suppression behavior.
Operational considerations for Amperity integrations
Throughput and pagination
Confirm Amperity and destination limits for request rates, payload size, concurrent jobs, audience size, and batch frequency. Validate page size, cursor behavior, ordering, and cursor expiration before implementing incremental retrieval.
Idempotency and reconciliation
Persist customer, source-system, transaction, audience, job, and file identifiers. Use deterministic file names or idempotency keys, reconcile counts and totals, and commit watermarks only after successful processing.
Long-running operations
Do not hold a synchronous workflow open indefinitely. Persist the job state and use a scheduled workflow to poll with bounded retries, backoff, maximum execution periods, and escalation for stalled operations.
Schema and file changes
Use explicit versioned mappings, required-field and type validation, contract tests, and dead-letter handling. For files, define encoding, compression, encryption, manifests, naming, partial-file detection, retention, and duplicate-file behavior.
Observability
Log correlation identifiers rather than sensitive profile data. Track batch and job IDs, audience IDs, records read and rejected, watermarks, retry counts, and failure reasons.
Why use Martini instead of scripts or point-to-point integrations?
Orchestrated integration logic
Martini coordinates Amperity API calls, file exchange, batch submission, status polling, transformation, target delivery, and operational escalation in workflows rather than scattering logic across scripts.
Reusable mappings and rules
Explicit mappings and validation rules make customer, transaction, product, and audience contracts easier to test and change as tenant schemas evolve. Business rules can be reused across API and file-based flows.
Reliable operations
Martini provides a place to implement secrets management, retries, idempotency, reconciliation, logging, and scheduled processing for long-running Amperity operations. This reduces fragile point-to-point behavior without assuming a native connector.
Controlled API access
Martini can expose controlled APIs for applications that need Amperity data, applying authorization, transformations, and tenant-specific rules while keeping downstream consumers independent of internal Amperity details.