Migrating from dex
Migration Alternatives Guide
As dex transitions, we're committed to helping you continue your data engineering journey with minimal disruption. This guide outlines two recommended migration paths, each designed to match different needs and preferences.
It's important to note that both alternatives will require adaptation and are not exactly the same as dex. In this guide you will find the most similar feature which replaces dex features in each tool, but remember they are not the same.
Alternative 1: dbt Cloud + Fivetran
Best for: Teams that prefer best-of-breed tools and want flexibility in their stack composition. This alternative is also less complex in terms of infrastructure.
Your existing infrastructure can be reused. Since dex already uses dbt under the hood, your models, tests, and Git repository can be connected directly to dbt Cloud with minor configuration adjustments. Your data remains in your cloud storage—no migration required.
What This Alternative Offers
This combination replaces dex with two specialized platforms working together:
Fivetran handles data ingestion (replacing dex Connectors)
dbt Cloud manages transformation, orchestration, testing, and documentation (replacing dex Models, Flows, Tests, and Documentation)
Comparison to dex
Connectors (one-click data ingestion)
Fivetran connectors (500+ sources)
Models (SQL/Python transformations)
dbt models with SQL
Flows (orchestration and scheduling)
dbt Cloud Job Scheduler
Tests (data quality checks)
dbt tests (schema, freshness, custom)
Documentation (lineage and column docs)
dbt Cloud Catalog with auto-generated docs
Version Control (Git integration)
dbt Cloud native Git integration
Key Difference: Unlike dex's unified platform, this requires managing two separate tools. However, you gain access to the largest ecosystem in modern data engineering, with extensive community support and third-party integrations.
Pricing Model and Recommendations
dbt Cloud Pricing
dbt Cloud uses a hybrid pricing model based on developer seats and successful model builds.
Developer Plan (Free)
1 developer seat
Up to 3,000 successful models built per month
Best for: Individual users or small trial projects
Team Plan ($100/developer/month)
Up to 8 developer seats + 5 read-only seats
15,000 successful models built per month included
$0.01 per additional successful model built
Best for: Small to medium teams
Fivetran Pricing
Fivetran uses Monthly Active Rows (MAR) pricing calculated per connection:
MAR = rows inserted, updated, or deleted each month
Pricing is tiered: the more data you move, the lower your per-row cost
Pricing Plans:
Free Plan: Trial period with limited usage
Standard Plan: Pay-as-you-go based on MAR per connection
For most teams migrating from dex, the Team and Standard Plans will be the most suitable option, offering the right balance of features, capacity, and cost. Start with the free plans to test the platform, then upgrade to Team/Standard when you're ready to onboard your full team.
Getting Started
Resources:
dbt Fundamentals Course (free)
You can also use the Fivetran + dbt Cloud integration to automatically trigger dbt runs after connector syncs.
Alternative 2: Databricks
Best for: Teams that want an all-in-one platform with native support for data science, machine learning, and advanced analytics.
Your existing infrastructure can be reused. Your data remains in your cloud storage—no migration required. While dex uses dbt for transformations, Databricks offers multiple approaches including SQL queries, notebooks, and Delta Live Tables. You will need to adapt your existing dbt models to work with Databricks' tooling, which may require more extensive refactoring compared to migrating to dbt Cloud.
What This Alternative Offers
Databricks provides a unified lakehouse platform that combines all capabilities:
Data ingestion and ETL
SQL and Python transformations
Workflow orchestration
Notebooks for collaborative development
Built-in governance and security
Native support for ML/AI workloads
Comparison to dex
Connectors
Databricks connectors + Partner Connect (Fivetran, Airbyte integration)
Models
SQL queries, notebooks, Delta Live Tables
Flows
Databricks Workflows (orchestration)
Tests
Expectations in Delta Live Tables + custom validation
Documentation
Unity Catalog with automated lineage
Version Control
Native Git integration (Repos)
Python Scripts
Databricks notebooks with full Python support
Key Difference: Databricks is a more comprehensive platform that includes dex's capabilities plus advanced features for data science, machine learning, and real-time streaming. It requires more infrastructure knowledge but offers greater power and flexibility.
Pricing Model and Recommendations
Databricks uses a consumption-based pricing model with two cost components:
1. Databricks Platform Cost (DBUs)
Databricks Units (DBUs) measure processing power. Different workload types consume DBUs at different rates:
Jobs Compute: $0.07 - $0.15 per DBU (automated/scheduled workloads)
All-Purpose Compute: $0.40 - $0.65 per DBU (interactive development)
SQL Warehouses: $0.22 - $0.55 per DBU (SQL analytics)
Delta Live Tables: $0.20 - $0.36 per DBU (declarative ETL)
2. Cloud Infrastructure Cost
You also pay your cloud provider (AWS, Azure, or GCP) for:
Virtual machine compute time
Storage (S3, ADLS, GCS)
Data transfer
Important: Infrastructure costs are often 50-200% of DBU costs, so budget accordingly.
For most teams migrating from dex, the Standard Tier will be the most suitable option.
Getting Started
Resources:
Choosing the Right Alternative
Choose dbt Cloud + Fivetran if you:
Prefer specialized, best-of-breed tools
Have a SQL-focused data team
Want the largest community and ecosystem
Need predictable, seat-based pricing for development
Primarily focus on analytics and BI use cases
Choose Databricks if you:
Want an all-in-one platform
Need advanced data science and ML capabilities
Have Python-proficient team members
Require real-time streaming and complex transformations
Want unified governance across all data workflows
Important Note: dex Connectors
dex's connectors are proprietary technology and will not be provided as part of the migration. These connectors were developed internally and remain the intellectual property of dex. Some connectors may not be available in other tools.
If you have unique data sources or prefer to maintain control, you can build custom extraction scripts. This approach works well when combined with a simple orchestration tool.
Last updated
Was this helpful?