
That choice shapes onboarding speed, security ownership, operating cost, and how fast models become dependable production pipelines. dbt Labs reported more than 5,000 Cloud customers in early 2025, a vendor-reported figure, so plenty of teams are making this call.
This guide is for analytics engineers and data platform leaders. It compares the two options and gives you a framework for choosing. One naming note: dbt's current docs call the managed service the "dbt platform." We use "dbt Cloud" because most teams still search for it that way.
Key Takeaways
- dbt Core is the free, open-source CLI. You own infrastructure, orchestration, CI/CD, secrets, and docs hosting.
- dbt Cloud is a managed platform with a browser IDE, scheduler, and collaboration features, priced by subscription and usage.
- Decide on engineering capacity, security needs, existing orchestration, and total cost of ownership, not license price.
- Hybrid setups work only if you assign clear ownership of each layer.
dbt Core vs dbt Cloud: Quick Comparison
| Dimension | dbt Core | dbt Cloud |
|---|---|---|
| Cost | Software is free (Apache 2.0); you fund infrastructure and maintenance | Free Developer plan; Starter at $100 per user/month; custom Enterprise pricing |
| Setup | You install Core and a warehouse adapter in your own environment | Hosted, browser-based, managed services |
| Development | Local IDE and CLI, full flexibility | Browser Studio IDE, plus a platform CLI and a VS Code extension for local work |
| Scheduling | Needs an external scheduler | Built-in scheduler (cron, job-completion, API triggers) |
| CI/CD | Self-configured Git and CI | Integrated CI checks on GitHub and GitLab |
| Docs | Static site you host yourself | Catalog, built from job metadata |
| Security | Your credentials, access, and networking | SaaS controls; SSO and RBAC at Enterprise; PrivateLink at Enterprise+ |
| Best fit | Platform-capable teams needing control | Mixed-skill teams wanting speed and less upkeep |

Plan details come from dbt's pricing page as of November 2026. Confirm current limits before you buy.
What Are dbt Core and dbt Cloud?
What Is dbt Core?
dbt Core is the open-source command-line framework for developing, testing, documenting, and running SQL transformations in your warehouse. You install it with a warehouse adapter, for example python -m pip install dbt-core dbt-postgres. Core and adapters release separately.
The trade-off is plain. You get local control, flexible tooling, custom CI/CD, and private deployment. You also own:
- Python environments and version compatibility
- Credentials and secrets
- Orchestration and monitoring
- Hosting generated documentation
- Upgrades
Use Cases of dbt Core
Core suits technically strong teams that already run Airflow, Dagster, Prefect, GitHub Actions, or Jenkins. It also fits private-network or heavily customized environments.
Accor Digital & Tech is a good example. After a study phase, it chose open-source tools: dbt Core for local transformation and testing, Airflow for scheduling, and Astronomer Cosmos to run dbt projects as Airflow task groups.
What Is dbt Cloud?
dbt Cloud is a managed SaaS platform around the dbt engine. It adds hosted development, job scheduling, CI, Catalog, a Semantic Layer, APIs, and admin controls.
You get faster onboarding, standardized environments, centralized visibility, and less platform maintenance. You also take on:
- Subscription and usage charges
- Plan limits (Developer includes 3,000 successful models a month; Starter 15,000)
- Dependence on a SaaS vendor
- Separate orchestration for ingestion and downstream systems

Use Cases of dbt Cloud
Cloud fits growing or mixed-skill teams that want to spend time on modeling, not developer environments. SpotOn migrated its models to dbt Cloud after adopting Snowflake. Its dbt Labs case story reports model creation dropping from days or weeks to hours and a 750% increase in engineering contribution. Those are company-reported results, not an independent benchmark.
dbt Core vs dbt Cloud: What Is Better?
Neither wins outright. Both run the same transformation logic. The real question is who owns the operating model and how much control you need at each layer.
When dbt Core Fits
Choose Core when:
- You have platform engineering capacity
- You need private or customized deployment
- You already run an external orchestrator
- You require Git or CI tooling that Cloud doesn't support
- Infrastructure control matters more than convenience
When dbt Cloud Fits
Choose Cloud when:
- You need a fast start and centralized development
- You want built-in dbt scheduling and simple collaboration
- You'd rather not manage upgrades and the platform around dbt
Check Security Before Choosing Cloud
Confirm that Cloud meets your requirements for data residency, private connectivity, identity management, auditability, and regulated workloads.
dbt Labs reports SOC 2 Type II and ISO 27001:2022. Enterprise accounts can pick a region (North America or EMEA), and PrivateLink and IP restrictions are listed for Enterprise+.
dbt says it aligns with HIPAA Security standards, which is not a HIPAA certification, so regulated teams should ask about BAA terms directly.
Calculate Total Cost of Ownership
Don't compare a free license against a subscription. Core costs include:
- Engineering time to build and maintain the platform
- Orchestration, CI/CD, secrets management, and monitoring
- Documentation hosting and infrastructure
- Onboarding and support
Cloud costs include seats, usage above plan limits, and any orchestration you still run alongside it. No authoritative benchmark gives a standard dollar comparison, so model your own.

Hybrid Models
Teams can develop locally and run in Cloud, or the reverse. Assign explicit owners for development, execution, CI, orchestration, and metadata. Mismatched runtimes or versions cause drift between local, CI, and production.

Decision Checklist
- Who maintains the runtime?
- Who responds when a job fails?
- Must the orchestrator control upstream and downstream tasks?
- Which tools must stay inside your network?
- How much customization do you need?
AICG's strategy and roadmap work covers this decision alongside team structure and a prioritized backlog of data products. Its dbt health check reviews architecture, CI/CD maturity, warehouse spend, and run performance.
Real-World Examples and Decision Scenarios
| Factor | Accor Digital & Tech | SpotOn |
|---|---|---|
| Choice | dbt Core + Airflow + Cosmos | dbt Cloud |
| Trigger | Managing pipelines at scale | Snowflake adoption, existing models to migrate |
| Outcome | Tests after each model | Days/weeks to hours; +750% engineering contribution |

Accor shows Core working when an orchestrator is already central. SpotOn shows Cloud working when onboarding and contribution speed matter most.
Situational cues:
- Control and customization dominate: Core
- Managed delivery and collaboration dominate: Cloud
- Responsibilities split across the stack: a deliberate hybrid, often with outside help
Hybrid setups still need solid CI/CD. AICG's implementations use slim CI with state:modified+, deferred builds, source-freshness gates, and nightly full builds in GitHub Actions. If you're weighing dbt architecture, CI/CD, or implementation, talk with AICG's senior-led data and AI team about building production systems your team can own.
Conclusion
dbt Core and dbt Cloud share the same transformation foundation. They differ in platform ownership, operational responsibility, cost structure, and control.
The better choice is the one that matches your team's priorities: faster delivery, reliable pipelines, governance, and predictable costs. AICG's DBDeux is built for teams that outgrow either dbt Cloud or self-managed Core and need a managed path forward.
Frequently Asked Questions
What is dbt Core?
dbt Core is the open-source CLI framework for transforming, testing, documenting, and managing warehouse data models. Your team operates the surrounding infrastructure.
What is dbt Cloud?
dbt Cloud is a managed platform built around dbt's engine. It adds hosted development, scheduling, collaboration, CI/CD, and documentation or metadata features.
Is dbt Core free?
The software is open source and free. Production use still costs money for infrastructure, orchestration, CI/CD, monitoring, secrets management, and engineering time.
Is dbt Cloud free?
There's a free Developer plan: one seat, 3,000 successful models a month, one project, and no API access. New accounts get a 14-day full-featured trial; Starter is $100 per user/month. Verify current limits on dbt's pricing page.