Migrate your Synerise data

The primary customer identity object in Synerise. Accessed via the Profile Management API (89 endpoints). Exportable to CSV/JSON/JSONL with a 10M record cap per job. Default attributes are email, firstName, lastName, city, birthDate, phone. Custom attributes are fully supported but names become immutable after creation.

Behavioral signals tracked per profile — page.visit, product.view, added-to-cart, transaction, and 40+ other event types. Batch event ingestion is supported via POST /v4/events. Events are assigned to the client's clientId or UUID. Historical event export is available through the Analytics Suite API.

Companies are linked to Profiles via the profile.assigned-to-company event. There is no standalone Company object — company data lives as profile attributes or schema records in Brickworks. We map these by tracking the assignment event and merging company fields into a structured schema.

Transaction records are created via POST /v4/transactions or POST /v4/transactions/batch. They contain line items, totals, and timestamps. Exportable via Data Management API. Transaction events (transaction, cancelled-transaction) are automatically tied to the profile that made the purchase.

Product/item feeds managed in the Data Modeling Hub. Exports available as CSV, JSON, or JSON lines. Item feeds are used by AI recommendation models (personalized, visual similarity, top items). Catalog records can include arbitrary custom fields per catalog schema.

Segmentation returns true/false values per profile. Export includes segment membership flags. Custom segments built in the Segmentation builder may have complex rule logic that requires expression evaluation during import to reproduce membership on the destination platform.

Any profile attribute beyond the defaults can be custom-named. The critical constraint is that custom attribute names cannot be changed after creation. We audit all custom attribute names during scoping and flag immutability risks before migration begins.

Workflows are defined in Automation Hub with trigger nodes, conditions (Profile Filter), and actions (Send Email, Send SMS, webhook calls). Workflows are fire-and-forget — completion of an action node does not gate workflow progression. Active workflow state cannot be migrated; we export workflow definitions and re-activation is a separate step post-cutover.

Recommendation configurations (personalized, visual similarity, last seen, top items, item comparison) are trained on catalog feeds and profile event history. We export recommendation model configurations and re-import them to the destination's recommendation engine where equivalents exist. Visual similarity models trained on Synerise's proprietary image embeddings require re-training on the destination platform.

Brickworks schemas are arbitrary record structures that store any data type. Exportable from the Data Modeling Hub. Schema definitions (field names, types, constraints) must be recreated on the destination. Singleton schemas and 'one to many filtered' field types have specific handling requirements.

Campaign definitions — email, SMS, push, WhatsApp — are accessible via the Campaigns API (100 endpoints). We export campaign configurations including templates, audience rules, and scheduling. Active campaign state (sends in progress, queue depth) cannot be migrated atomically.

Profile tags are exported as true/false boolean attributes in the profile export. Tag names are free-form strings. We preserve full tag sets per profile and import them as tag fields or custom properties on the destination platform.

Synerise supports computed aggregates and expression results attached to profiles. These are exported as scalar values. Since aggregates are computed from underlying events, destination recomputation may produce different values if the event history is not fully replicated.