What’s New in Qlik Cloud

Data Integration

With the sky bruising and summer fast becoming a memory, Qlik is keeping the sun shining just a little longer with a small but valuable set of updates. This month brings smarter data stewardship, new dataset-level data contracts, and a simpler way to move semantic types between environments. All with a clear focus on trust, governance, and reducing manual effort.

1. Data stewardship gets smarter and more flexible

This month Qlik has introduced two complementary improvements to data stewardship in Qlik Cloud.

First, stewardship records can now be stored in Qlik Cloud native storage, as an alternative to Snowflake. When creating a resolution sprint, teams can choose whether resolved records are written to their own Snowflake environment or kept directly within Qlik Cloud. This removes the need to configure an external database connection simply to run a stewardship workflow, which is particularly useful for organisations without Snowflake or those that prefer to keep the process contained within Qlik Cloud.

The second update adds an agentic remediation layer to the stewardship sprint. Previously problematic records were assigned to a human steward. With this update AI can now analyse the issue and propose a corrected value where its confidence is sufficient. The steward remains firmly in control and can accept, modify, or reject every suggestion.

The workflow also makes it clear how each value was resolved, distinguishing between unchanged, AI-resolved, and manually resolved records. A full audit trail records every AI suggestion and human decision, including the actor, timestamp, original value, and final resolved value.

What could this look like in practice?

Imagine we have a customer dataset containing thousands of recurring address, postcode, or formatting issues. Rather than asking a data steward to investigate every record from scratch, AI can propose corrections first. The steward then concentrates on reviewing those suggestions and dealing with the genuinely ambiguous cases.

That can reduce the manual effort involved in repetitive remediation while preserving the human accountability and auditability that data stewardship requires. Combined with native Qlik Cloud storage, it also lowers the setup overhead needed to get the stewardship process running in the first place.

More information:

Reviewing AI fixes in a resolution sprint
Creating a resolution sprint

2. Introducing data contracts for datasets

Qlik Cloud now supports data contracts at dataset level, adding a continuous evaluation layer to data quality management.

A contract lets teams define quality expectations using Qlik Trust Score™ dimension metrics and threshold-based warning ranges. Qlik then evaluates the dataset against those expectations and surfaces a clear in-contract / out-of-contract status through visual indicators.

If a dataset falls outside the agreed thresholds, owners are notified directly, rather than relying on a downstream consumer to discover the problem first. That makes quality degradation much more visible and gives teams a chance to act before the issue spreads further.

Key capabilities include:

• Contract definition and status visibility at dataset level
• Quality expectations based on Qlik Trust Score™ dimensions
• Threshold-based warning ranges
• Visual indicators for in-contract and out-of-contract status
• Notifications when a contract is violated

What could this look like in practice?

A customer dataset might be expected to maintain a minimum level of completeness or freshness. If the relevant Trust Score™ dimension drops below the agreed threshold, the contract moves out of compliance and the owner is alerted.

That gives teams a much clearer way to formalise what “good enough” looks like and spot deterioration before it becomes a consumer-facing problem.

More information: Managing data contracts

3. Import and export semantic types

Semantic types can now be imported and exported directly from the Qlik Cloud interface using a JSON file. This makes it much easier to move semantic type definitions between tenants or replicate the same configuration across development, test, and production environments without recreating them manually.

It is a small update, but a useful one for teams trying to keep data classification and quality rules consistent across environments.

More information: Importing and exporting semantic types