✨ New & Improved

• New navigation bar: Sub-menu will replace the existing tabs for a better UX
• Redshift's ingestion limitation: ingestion time reduced by at least 5 times

🛠 Fixes

• Fixed several bugs around ingestion jobs for BigQuery, dbt

✨ New & Improved

• QuickSight integration: you can now integrate and visualise your dashboards on Quicksight
• Datasource Timezone 🕛: you can now precise the timezone of your datasources
• Display Timezone 🕛: you can now choose which timezone you wish to see all data on Sifflet's UI
• Improved dbt integration: Sifflet can now ingest dbt artifacts when not generated in the right order
• Incidents:
  • Page loading time optimised
  • Timeline tab has been moved to the Overview tab
• Catalog: added a Reset button to clear more easily your filter

🛠 Fixes

• Fixed a bug where data assets with long names will not display properly on some cases
• Fixed a bug where data were not pulled correctly from BigQuery on some cases
• Fixed a bug where a Completeness rule could not run on some edge cases (with qualifications)

✨ New & Improved

• LookML repository SSH and token connection. You can now connect to your LookML repository using both SSH or a token
• New Rule status "Requires your attention": this new status will appear when there is an issue in your rule setup that requires your attention. For instance, running a rule requiring a date column on a table that doesn't have one will result in this status.
• Freshness rule new graph: Freshness graph is now different than the completeness graph and provides clearer information

🛠 Fixes

• Fixed a bug where the graph will not show while using the multi dimension analysis (group by) in some edge cases

✨ New & Improved

• Dashboard page revamp: Our dashboard page comes now with deeper insights and a more personalized experience for the user.
• Pinning: Users can now pin their most useful data assets, data quality rules, or incidents and easily navigate to them from the My summary page on the dashboard.
• "Assigned to" Filter in Incidents page: Users have an additional filter "Assigned to" which allow to filter your incidents on certain users
• 7 latest rule run Status: you can now see the 7 latest run status for each rule instead of the last one

🛠 Fixes

• Fixed a bug where rules will result in a technical error in few instances
• Fixed an issue where the role 'Viewer' was missing some access rights (e.g. access to the failing rows of a rule)

✨ New & Improved

• ML-rules data querying optimisation: Sifflet now optimises the data sources querying for the models training in your datasource by querying only the new data ingested since the last rule execution

🛠 Fixes

• Fixed an issue where rules ran with technical errors in a few instances
• Fixed an issue where the rule schedule will sometimes not update

✨ New & Improved

• Power BI integration (Beta): Supporting Microsoft Power BI as a bi integration. Sifflet will be able to catalog dashboards, tiles, and datasets as well as full lineage to the actual datasources.

• Bigquery integration test : advanced testing to make sure that Sifflet and Bigquery are properly configured.

• Dataset selector optimization : improved user experience for the dataset selector used in both filters and quality rule creation stepper.

• Metadata pulling optimization: performance optimization for the jobs pulling metadata from sources.

🛠 Fixes

• Graph misinterpretation when min, max, and predicted value are the same.
• Fixing the preview issue when trying to sample a Snowflake table.
• Display a comprehensive message when a multi-dimension rule fails because of a high dimension cardinality.

✨ New & Improved

• Multi-dimensional monitoring (Beta): Users can now monitor their data assets at a more granular level by breaking down the monitoring based on any dimensional/low-cardinality fields.
  This type of monitoring can be applied to any data quality template.
• Multiple UX improvements: including a better UX when deploying new versions of the product and a redesign of notifications cards and an improvement of the data quality rules monitoring page.

🛠 Fixes

• Fixed a bug on the catalog search preventing from considering the tags filter.

✨ New & Improved

• Multi-schema/datasets integration: Users can simultaneously integrate Sifflet to multiple schemas (and datasets in the case of BigQuery) of the same data source.
• Data quality rules batch editing users can simultaneously batch edit schedules, descriptions, tags, and alert destinations for multiple rules.
• Auto-coverage rules frequency: Users can now schedule Auto-coverage suggested rules with cron expressions.

🛠 Fixes

• Fixed bugs preventing the run of ML-based rules in some specific circumstances.

✨ New & Improved

• Data assets usage: Users can better understand their most critical data assets based on their usage. This feature ranks the data tables of the same data source based on the number of reading queries and gives the breakdown by the user account. This feature is GA on BigQuery only.
• Time offset: you can now insert a time offset for your monitoring rules in case of a constant time delay when your dataset gets updated. You can find more details here
• Slack and email alerting: we streamlined the way alert notifications are now sent. Previously only monitoring rules with criticality "Critical" will send notifications. All monitoring rules can send notifications as long as the destination is defined.
• UI/UX improvements: a new design for both the asset page and rule pages for a better navigation on the UI
• Performance optimization on connections to data warehouses and BI tool Integration Tableau

🛠 Fixes

• Fixed a display bug when hovering on data assets on the lineage view
• Fixed an issue that prevented the creation of a new user

✨ New & Improved

• SSO (Single Sign-On): you can now enable SSO on Sifflet. For more information, you can refer to this page
• CLI: you can now interact programmatically with Sifflet. For more information, you can refer to this page
• Monitoring new templates: 4 new templates have been added: "Duplicate Count", "Duplicate Percentage", "Null Count" and "Null Percentage". You can now monitor - with a dynamic threshold - the count or percentage of Null values in a field and the count or percentage of duplicates in a field. You can find these template in Category "Field Profiling"
• Several UX enhancements across the UI, such as ability to see the lineage directly in an incident report

🛠 Fixes

• Fixed a bug preventing some incidents to be created following a rule failure