Why models are the core of dbt - The Real Reasons

The Big Idea

What if you could turn messy data into clean insights with just a few lines of code?

The Scenario

Imagine you have a huge spreadsheet with messy data from many sources. You try to clean and combine it manually every time you need a report. It takes hours and you often make mistakes.

The Problem

Doing this by hand is slow and confusing. You lose track of changes, repeat work, and errors sneak in easily. It's hard to keep data reliable and up to date.

The Solution

dbt models let you write simple code to transform data step-by-step. They run automatically and keep your data clean and organized. You can track changes and fix errors quickly.

Before vs After

✗ Before

Copy data from source A to B, then clean columns in Excel, then join with source C manually

✓ After

select * from source_a where condition;
-- then in another model: select cleaned columns from previous model join source_c on key

What It Enables

With dbt models, you build reliable, reusable data transformations that update automatically and scale easily.

Real Life Example

A marketing team uses dbt models to combine website clicks, ad spend, and sales data into one clean table for daily reports without manual work.

Key Takeaways

Manual data cleaning is slow and error-prone.

dbt models automate and organize data transformations.

This makes data reliable, reusable, and easy to update.

Practice

1. What is the main role of models in dbt?

easy

A. To transform raw data into useful tables or views

B. To store raw data without changes

C. To create visual dashboards

D. To manage user permissions

Solution

1. Step 1: Understand the purpose of models in dbt

   Models are SQL files that define how raw data is transformed into clean, organized tables or views.

2. Step 2: Identify the correct role from options

   Only To transform raw data into useful tables or views describes transforming raw data into useful tables or views, which is the core function of models.

3. Final Answer:

   To transform raw data into useful tables or views -> Option A

4. Quick Check:

   Models transform data [OK]

Hint: Models transform raw data into tables/views [OK]

Common Mistakes:

• Confusing models with dashboards
• Thinking models store raw data unchanged
• Assuming models manage permissions

2. Which of the following is the correct way to define a model in dbt?

easy

A. models/my_model.yaml containing configuration only

B. models/my_model.py containing Python code

C. models/my_model.txt containing raw data

D. models/my_model.sql containing a SELECT statement

Solution

1. Step 1: Recall dbt model file requirements

   dbt models are SQL files that contain SELECT statements to transform data.

2. Step 2: Match file type and content

   Only models/my_model.sql containing a SELECT statement uses a .sql file with a SELECT statement, which is correct for a dbt model.

3. Final Answer:

   models/my_model.sql containing a SELECT statement -> Option D

4. Quick Check:

   Model = SQL file with SELECT [OK]

Hint: Models are SQL files with SELECT statements [OK]

Common Mistakes:

• Using Python or text files for models
• Confusing config files with models
• Not including a SELECT statement in model files

3. Given this dbt model SQL code:

SELECT user_id, COUNT(*) AS orders_count FROM raw.orders GROUP BY user_id

What will this model produce when run?

medium

A. A table or view with user_id and their total order counts

B. A list of all orders without grouping

C. An error because COUNT(*) is invalid

D. A table with only user_id and no counts

Solution

1. Step 1: Analyze the SQL query in the model

   The query selects user_id and counts orders grouped by user_id, aggregating orders per user.

2. Step 2: Determine the output of the model

   The model will create a table or view showing each user_id with their total number of orders.

3. Final Answer:

   A table or view with user_id and their total order counts -> Option A

4. Quick Check:

   GROUP BY user_id with COUNT(*) = aggregated counts [OK]

Hint: GROUP BY with COUNT(*) gives totals per group [OK]

Common Mistakes:

• Ignoring GROUP BY and expecting raw data
• Thinking COUNT(*) causes errors
• Assuming counts are missing

4. You wrote this dbt model SQL:

SELECT customer_id, date, SUM(amount) AS total FROM sales GROUP BY customer_id

But dbt throws an error. What is the likely problem?

medium

A. SUM(amount) cannot be used with GROUP BY

B. The SELECT includes date but GROUP BY does not, causing mismatch

C. customer_id should be aggregated with SUM()

D. Missing WHERE clause causes error

Solution

1. Step 1: Check SELECT and GROUP BY columns

   SELECT has customer_id, date, and SUM(amount), but GROUP BY includes only customer_id.

2. Step 2: Identify mismatch causing error

   All non-aggregated columns in SELECT must be in GROUP BY. date is missing in GROUP BY, causing error.

3. Final Answer:

   The SELECT includes date but GROUP BY does not, causing mismatch -> Option B

4. Quick Check:

   GROUP BY columns must match SELECT non-aggregates [OK]

Hint: SELECT non-aggregates must match GROUP BY columns [OK]

Common Mistakes:

• Ignoring GROUP BY and SELECT column mismatch
• Thinking SUM() can't be used with GROUP BY
• Assuming WHERE clause is mandatory

5. You want to create a dbt model that builds a monthly sales summary table. Which approach best uses models as the core of dbt?

hard

A. Create a YAML file listing monthly sales without SQL

B. Manually export raw sales data and summarize in Excel

C. Write a SQL model that selects sales data, groups by month, and calculates totals

D. Use a Python script outside dbt to summarize sales

Solution

1. Step 1: Identify how models transform data in dbt

   Models are SQL files that transform raw data into organized tables, like monthly summaries.

2. Step 2: Choose the option that uses dbt models correctly

   Write a SQL model that selects sales data, groups by month, and calculates totals uses a SQL model to group and summarize sales by month, fitting dbt's core purpose.

3. Final Answer:

   Write a SQL model that selects sales data, groups by month, and calculates totals -> Option C

4. Quick Check:

   Models transform data with SQL for summaries [OK]

Hint: Use SQL models to transform and summarize data [OK]

Common Mistakes:

• Using external tools instead of dbt models
• Confusing YAML config with data transformation
• Ignoring dbt's SQL model workflow