dbt Analytics Engineering Certification Exam

Last checked November 10, 2026 against the official dbt Labs exam page.

1. Question 1Developing and optimizing dbt models

   A data team maintains an events model with roughly 500 million rows. Each run should add only the newest events, keep the existing history, and avoid rescanning the whole source table. Which materialization fits best?

   1. AAn incremental model whose is_incremental() filter selects only the newest rows on each run
   2. BAn ephemeral model, because dbt inlines it as a common table expression in each query that uses it
   3. CA table that is fully rebuilt on every run, so the output always mirrors the current source data
   4. DA view, because the warehouse recalculates the underlying query each time the model is read
   Show answer and explanationHide answer

   Correct answer: A

   An incremental model builds the full table once and, on later runs, its is_incremental() block limits the work to new rows, so history is kept without a full scan. A view stores no data, an ephemeral model is only inlined SQL that is never persisted, and a full-rebuild table reprocesses every row on every run.

2. Question 2Managing dbt models governance

   A downstream team depends on the exact column names and data types of a customer mart. The team wants the build to fail if a column is renamed or its type changes. Which configuration does this?

   1. AA model contract with enforced set to true, and a declared data_type for each column in the YAML
   2. BA detailed description on each column, so that schema changes are visible in the generated documentation
   3. CA not_null generic test on every column, so any schema change is reported during dbt test
   4. DA table materialization, so that the warehouse keeps a copy of the previous schema for comparison
   Show answer and explanationHide answer

   Correct answer: A

   With an enforced model contract, dbt compares the model's columns and declared data types with the YAML during the build and fails when they do not match. Null tests check values rather than structure, descriptions only document the model, and a table materialization does not keep schema history.

3. Question 3Debugging data modeling errors

   A model fails with a Database Error about an invalid identifier in a SELECT that uses several ref() calls and macros. The engineer wants to see the SQL that dbt actually sent to the warehouse. Where should the engineer look?

   1. AThe seeds folder, where dbt saves a copy of each model's SQL so it can be reloaded after a failure
   2. BThe raw model file, because the warehouse executes the Jinja inside it exactly as it is written in the source
   3. CThe dbt_project.yml file, which keeps a copy of each model's SQL after the most recent successful run
   4. DThe compiled and executed SQL under the target directory, with Jinja already expanded into plain SQL
   Show answer and explanationHide answer

   Correct answer: D

   dbt writes the Jinja-expanded SQL to target/compiled and the statement it executes to target/run, which is what the warehouse actually received. The raw model file still contains Jinja the warehouse never sees, and dbt_project.yml and the seeds folder do not store model SQL.

4. Question 4Troubleshooting and optimizing dbt pipelines

   A nightly dbt build reports an error on a staging model. Several mart models that read from it show as skipped. Which nodes did dbt skip because of this failure?

   1. AThe upstream source tables feeding the failed staging model, since dbt re-checks their freshness
   2. BThe models and tests downstream of the failed staging model that depend on its output
   3. CEvery model in the same schema as the failed staging model, regardless of how they depend on it
   4. DOnly the generic tests attached to the failed staging model, while its child models still run
   Show answer and explanationHide answer

   Correct answer: B

   When a node fails in dbt build, dbt skips the descendants that depend on it, while parents that already completed are unaffected. Upstream sources are not skipped, schema membership does not decide skipping, and child models do not keep running when their parent fails.

5. Question 5Implementing dbt tests

   A model's order_id column must hold a distinct value on every row, and it must never be null. Which approach is the standard way to check both rules?

   1. AA snapshot on the model, because snapshots reject duplicate and null keys before they reach the target table
   2. BThe unique and not_null generic tests declared under the order_id column in the model's properties YAML
   3. CAn accepted_values test that lists every valid order_id, so that nulls and duplicates show up as unknown values
   4. DA severity of warn in the model config, so that duplicate or null rows are logged without failing the build
   Show answer and explanationHide answer

   Correct answer: B

   The unique and not_null generic tests are built-in checks declared on a column in YAML, and they fail when duplicate or null values exist. Snapshots record history rather than validate keys, accepted_values cannot practically list every valid ID, and severity only changes how a failure is reported.