Every product attribute,
segmentable in real time.
Bring your catalog into SegOps. Declare your own product schema, import any feed with AI-assisted field mapping, and segment products with the same DSL you use for users — all live against your event stream.
A catalog that moves with the business.
Most PIMs are static spreadsheets dressed up as databases. Product Intelligence is a live, queryable layer that earns its keep every time you ship a campaign.
Tenant-defined schemas
Declare exactly the attributes your catalog needs. Backed by a flexible JSON-typed store — add or remove fields without a database migration.
AI field mapping
Drop in a CSV from any source. Claude proposes a mapping into your schema and flags anything ambiguous. You approve before import runs.
One DSL, end to end
Product segments use the same condition language as user segments. Learn it once; reuse it everywhere.
Live against events
Product memberships update as inventory, price, or attributes change. No nightly batch, no stale audiences.
Cross-pillar joins
Build user segments off product behavior, and product segments off user behavior. The graph is one graph.
Versioned schemas
Every schema change is captured. Roll back if a field rename breaks downstream segments — without losing history.
Any feed in.
Clean catalog out.
Bring product data from Shopify, an ERP export, or a partner feed. The wizard detects columns, proposes a schema mapping with our intelligence agents, surfaces conflicts, and scores each product's AI visibility the moment the import lands.
- CSV, Parquet, or live APIFirst-party Shopify connector plus a generic CSV/Parquet ingest. Build your own with the API.
- AI-assisted field mappingOur intelligence agents read sample rows, recommend a mapping, and flag ambiguous columns for review.
- Versioned importsEvery import is auditable. Re-run, diff, or roll back without losing history.
Questions about Product Intelligence.
What does the product schema look like?
Tenant-defined. You declare the attributes that matter for your catalog (title, price, vendor, taxonomy, custom fields), back it with a flexible JSON-typed store, and add or remove fields without a migration. Same shape across SKUs, variants, and parents.
How does the AI field mapper work?
Drop in a CSV from any source. Our intelligence agents read the first few rows, suggest a mapping into your product schema, and surface unmapped columns for review. You approve, run, and the import lands in the analytics warehouse with a full audit trail.
Can product segments be used in user segments?
Yes. Product segments and user segments share the same DSL, so a user segment can reference "users who viewed any product in {segment}" — the join happens at query time over the live event stream.
How does this work with Shopify?
First-party Shopify ingestion is included. Products, variants, and inventory sync continuously; you can extend the default schema with custom attributes without forking the integration.