CloudModernize converts integration, warehouse, and analytics workloads. This is the fourth path: on-premises Informatica Data Quality to IDMC Cloud Data Quality. It reads the Model Repository and generates the corresponding CDQ assets — profiles, rules, and the rule-to-profile assignments that bind them.
The data quality gap
The platform question is settled the day an organization commits to IDMC. What remains is an inventory: hundreds of rules, each encoding a business decision someone made years ago and no one has documented since.
Profiles define what acceptable means for a domain. Rules encode the tests — some written against a regulation, some against a retired system, some against a preference held by an analyst who has since left. Assignments bind rules to profiles. Reference tables carry standardization logic. Mapplets are reused across dozens of rules, where a single misreading propagates silently.
Rebuilt by hand, each object becomes a small act of archaeology followed by a small act of translation. During a manual rebuild the organization runs two definitions of data quality at once and cannot demonstrate that they agree.
One path · Three stages
Reads a metadata export from one Model Repository and returns a complete object inventory, tiered by what converts automatically and what does not.
Builds a structural representation of each repository object and generates the corresponding CDQ asset. A metadata-to-metadata operation throughout.
Converted objects are tested against source behavior before promotion, and the estate lands inside a defined governance structure rather than as loose migrated assets.
The operating model
Count profiles, rules, assignments, and transformations by type from repository metadata.
Sort every object into automated, manual until October, or structural rework.
Generate CDQ assets from the shared structural representation.
Test converted behavior against source before anything is promoted.
What this path covers
Three do not, and they are named. Nothing has been omitted for presentation.
Twenty-nine transformation types span core integration logic and the data quality transformations that carry most rule behavior — Standardizer, Normalizer, Consolidation, Address Validator, Association, Match, Parser, and the rest.
Data Masking and Classification convert automatically through Kestryl, PDI's PII discovery and remediation product. An estate using either needs Kestryl licensed alongside the accelerator.
Labeler, Data Processor / Relational to Hierarchical, and Exception. Automation is scheduled for the October release. Until then PDI's team converts them manually inside the engagement.
Profile definitions convert as first-class objects. Reference table content is exported alongside repository metadata so rules that depend on it can be sized rather than estimated.
Delivery record
Two life insurers and a large public-sector pension plan. Client names are withheld at their request; industry and geography are accurate as stated.
Governed automation
Where this sits
Assess existing logic and dependencies, convert workflows, reconcile outputs, and generate safe test data for repeatable releases.
Evaluate schemas and SQL, automate target conversion, and validate record and metric consistency.
Inventory the rule estate, convert profiles, rules, and assignments, validate against source behavior, and land the result under governance.
Next step
One native Informatica export command, run once per project across a single Model Repository. In return you receive a complete inventory of your estate, tiered by what converts automatically and what does not, and a firm estimate you can take to a budget conversation.