What's new in dbt Cloud - August 2024

Believe it or not, it's time for our regular product update launch post! Feels like we were just announcing a bunch of new features in dbt Cloud... but our product teams are 👏 shipping 👏, and so here we are again.

Before we dive in, we’d be remiss if we didn’t plug Coalesce—the annual Analytics Engineering conference where thousands of data professionals and thought leaders like yourselves will come together in beautiful Las Vegas, Nevada to learn, share, and grow. 🌱 We have an incredible agenda that’s quickly coming together, and we hope to see you there starting October 7.

Alright, on to the announcements!

📈 dbt Semantic Layer

Learn about the latest features powering the dbt Semantic Layer. If you’re not sure where to start with your semantic strategy, check out our recent blog post on the Five use cases for the dbt Semantic Layer.

🔐 Access controls: Admins can now federate access based on specific teams or groups of users by configuring multiple underlying data platform credentials and mapping them back to service tokens used for authentication in downstream apps. This means finance teams and data teams, for example, can have partitioned and granular access to underlying data. Group-level access controls are now GA and available for Enterprise customers.

🌎 Integrations: Our Microsoft Excel integration is now in Preview! You can find it in the AppSource Marketplace, or get started with our Docs. This integration is available for both Excel Online (in-browser via Microsoft 365) and Excel Desktop.

🗄️ Declarative caching improvements: Declarative caching now supports query time filters. Now, if you filter on a dimension that’s already included in the cache at query time, we automatically add that filter to the cached data set. This means faster delivery times, regardless of how you continue to splice your data. Declarative caching is GA for all Semantic Layer customers.

🔢 Cumulative metrics now support granularity options: Cumulative metrics are non-additive, making them tricky to calculate flexibly. Previously, if you wanted to roll up a cumulative metric to a different grain (e.g.: by week or by month), you’d need to create an entirely new metric for each grain. Now, you can keep your code DRY and streamline workflows by setting your own aggregation behavior when you request a cumulative metric by a new granularity. Check out the docs to learn more.

🐍 Python SDK: With the Python software development kit (SDK) for the dbt Semantic Layer, you can interact with semantic layer APIs and query metrics and dimensions more easily using Python. This allows you to build data products to serve internal stakeholders, or build experiences for external users, like your customers, more quickly. Read the docs to learn more.

🔀 Metrics in CI: You can now automatically test your semantic nodes (metrics, models, and saved queries) during your code review process! Just add validation checks in your CI job using the dbt sl validate command. You can also validate modified semantic nodes to guarantee the code changes you make to dbt models don't break your metrics. Check out the docs to learn more.

🔎 dbt Explorer

dbt Explorer is your command center for navigating, understanding, and improving your dbt projects. Dive into what’s new:

🗺️ New landing page: dbt Explorer has a new-and-improved landing page designed to be the perfect jumping off point for diving deeper into your dbt Cloud projects. Get an at-a-glance view of important context about your projects including descriptions, latest changes, and latest issues, or scan the mart and public models tables to understand relative usage and lineage to keep pipelines healthy. You’ll also notice improved filtering and sorting of models in the model filter view. Read the docs to learn more.

🦘 Navigate from the Cloud IDE into Explorer: Jump straight from the Cloud IDE into dbt Explorer to quickly visualize lineage and dependencies for the model, seed file, or snapshot you’re developing. (You can also navigate to the Cloud IDE from dbt Explorer in one click 🙌.) Check out the docs for more info.

🏃‍♂️Developer productivity

Check out these new features designed to help supercharge developer productivity:

😎 Environment-aware dbt snapshots: Snapshots in dbt were originally created before we introduced the concept of deferral, and so that meant that, previously, regardless of what environment (dev, staging, prod) you were in, if you ran dbt snapshots, it would build to the same target_schema. Now, dbt snapshots can now be configured to be environment-aware, making testing and developing snapshots much safer and easier. Now available in dbt Cloud for customers who are running on “Versionless.” Read the docs to learn more.

🔀 The dbt Cloud CLI now supports parallel command execution, just like the dbt Cloud IDE. This means you can run multiple different dbt commands such as dbt build and dbt compile at the same time. Read more about parallel execution in our docs.

🧺 The dbt Cloud CLI now also supports SQLFluff commands, allowing you to easily lint your SQL files for improved consistency and readability. More info in the docs.

⚠️ Notifications for warnings: You can now choose to get notified via Slack or email if a job run encounters warnings from tests or source freshness checks — in addition to already available options for notification on job success, failure, or cancelation. See the docs for more.

Platform & Partner

We have lots of exciting announcements about platform improvements and partner milestones to share:

☁️ Support for Azure deployments: dbt Cloud multi-tenant can now be deployed natively on Microsoft Azure! This is in addition to the current support for AWS deployments, bringing the same powerful dbt Cloud experience to even more data teams — regardless of your choice of cloud. Azure multi-tenant support is currently in Preview, and available to dbt Cloud Enterprise customers. Azure deployments can be hosted in Europe, with more deployment regions to follow in the coming months.

💫 New dbt Cloud UX: You’ve likely already noticed, but dbt Cloud has gotten a glow-up, including an updated navigation panel and account home page (now in Preview). These updates make it more intuitive to navigate between different areas of dbt Cloud. For instance, you can now quickly configure and find your favorite projects in the project switcher, while the new account homepage gives enterprise customers an at-a-glance view of each of the projects they have access to.

⚙️ Environment-level permissions: You can now enable developers to edit and trigger jobs in specific environments types (such as development or staging), while restricting them from accessing others (production) — giving you more granular control over permissions. More info in the docs.

☑️ Multi-factor Authentication (MFA): We've rolled out access to multi-factor Authentication to all dbt Cloud accounts, making it easier to enhance the security of your accounts with an extra layer of protection. See our docs for more.

👩‍💻 IAM User authentication is now generally available for Amazon Redshift, giving you the option to further enhance account security. Check out the docs for more; also note that this will require you to be using “Versionless” in dbt Cloud.

👾 PATs (Personal Access Tokens) used for dbt Cloud’s Azure DevOps integration have been reduced in scope. The integration uses the service user to to generate a limited PAT, that now has only the following scopes: code_full, project and build_execute. These PATs are valid only for 5 minutes, and become invalid after a single API call. See our ADO docs for more.

✨ Polaris Catalog integration partner: dbt is proud to be an integration partner for the Polaris Catalog launch! Using the catalog, teams can read and write from engines to a single catalog with centralized access controls. Read Snowflake’s launch blog to learn more.

See you in Vegas!

As always, we’re looking forward to your feedback and hope that these features help improve your analytics workflows and collaboration across your company.

We can't believe we're T minus ~60 days until Coalesce 2024 (😱!) and we CANNOT WAIT to be together with you all in Vegas (or online!) and share some of the exciting new features we’ve been cooking up! If you haven’t registered yet (what are you even doing), you can do that here. Hope to see you all real soon!