Skip to main content

dbt Cloud Explore (dbt Explorer)

In the open-source version of dbt (dbt Core), you generate documentation by running dbt docs generate and hosting the static HTML site yourself.

In dbt Cloud, documentation has been completely reimagined into a feature called dbt Explorer. It is a fully interactive, always-up-to-date data catalog and lineage tool built directly into the dbt Cloud platform.

Topics Covered​

  • Accessing dbt Explorer
  • The Interactive Lineage Graph
  • Column-Level Lineage
  • Project Health & Metrics

1. Accessing dbt Explorer​

dbt Explorer is available to all users in a dbt Cloud account (even those with "Read-Only" seats).

  1. Log into your dbt Cloud account.
  2. In the top navigation bar, click on Explore.
  3. You are immediately presented with a Google-like search bar. You can search for any model, source, seed, metric, or exposure in your entire project.

Because it is powered by dbt Cloud's metadata API, dbt Explorer is always perfectly in sync with your latest production run. You never have to manually generate or host static HTML files.

2. The Interactive Lineage Graph​

The core feature of dbt Explorer is the Lineage Graph. While the open-source dbt Docs has a static lineage graph, dbt Explorer's graph is highly interactive.

If a new Data Engineer (e.g., Rohan) joins your team in Bengaluru and needs to understand how the fct_daily_revenue table is built, he can open it in Explorer.

  • He can click on upstream staging models to instantly see their descriptions and code.
  • He can filter the graph by Tags (e.g., only show models tagged finance).
  • He can filter by Resource Type (e.g., hide tests, only show models and exposures).
  • He can instantly see the Status of the last run. If an upstream model failed last night, it will be highlighted in red directly on the graph.

3. Column-Level Lineage​

This is the most powerful feature in dbt Explorer (available on Enterprise plans).

Traditional lineage shows you that Model A feeds into Model B. But what if you only care about a specific column, like customer_email?

In dbt Explorer, you can click on the customer_email column inside dim_customers. The UI will isolate the lineage graph to show you exactly which raw source columns feed into that specific field, and exactly which downstream dashboard exposures consume that specific field.

Use Case: PII Auditing If the Security team asks: "Where does the raw phone_number column from the Salesforce API end up?" You simply click the column in Explorer, and you can instantly prove that it is masked in the staging layer and never reaches the final marts models.

4. Project Health & Performance Metrics​

Because dbt Explorer is tightly integrated with your dbt Cloud job history, it acts as an operational dashboard for your data pipeline.

When you click on a model (like fct_orders), you don't just see the SQL code. You also see:

  • Execution Time History: A chart showing exactly how long this specific model took to build over the last 30 days. If the model suddenly starts taking 10 minutes instead of 2 minutes, you can spot the anomaly immediately.
  • Test Coverage: A clear indicator of how many unique and not-null tests are applied to the model, and whether they passed in the most recent run.
  • Model Size/Rows: (If your data warehouse supports it), Explorer can display exactly how many rows were inserted or merged during the last execution.

Summary​

  • dbt Explorer is the modern, interactive replacement for static dbt Docs, built directly into dbt Cloud.
  • It is always up-to-date with your latest production runs.
  • It provides Column-Level Lineage for precise impact analysis and PII auditing.
  • It surfaces operational metrics like historical execution times and test success rates directly alongside the documentation.