Multiple data sources in Tableau - Deep Dive

Overview - Multiple data sources

What is it?

Multiple data sources in Tableau means connecting and combining data from different places or files into one view. This lets you analyze information that lives in separate systems together. For example, you might join sales data from one file with customer info from another. Tableau helps you blend or join these sources smoothly to create meaningful reports.

Why it matters

Without the ability to use multiple data sources, you would be stuck looking at pieces of information separately. This makes it hard to see the full picture or find important insights that come from combining data. Multiple data sources let businesses make smarter decisions by connecting dots across different systems, like sales, marketing, and inventory.

Where it fits

Before learning about multiple data sources, you should understand how to connect to a single data source in Tableau and basic data visualization. After mastering multiple data sources, you can explore advanced data blending, calculated fields across sources, and performance optimization.

Mental Model

Core Idea

Multiple data sources let you combine separate sets of data into one analysis to get a fuller, richer story.

Think of it like...

It's like cooking a meal using ingredients from different stores; you bring them together to make one dish that tastes better than any single ingredient alone.

┌───────────────┐   Connect   ┌───────────────┐
│ Data Source A │────────────▶│ Tableau View  │
└───────────────┘             └───────────────┘
       ▲                             ▲
       │                             │
┌───────────────┐   Connect   ┌───────────────┐
│ Data Source B │────────────▶│ Tableau View  │
└───────────────┘             └───────────────┘

Multiple sources feed into one combined view.

Build-Up - 7 Steps

1

Foundation

Connecting to a single data source

Concept: Learn how to connect Tableau to one data source to understand the basics of data import.

Open Tableau and choose a data source like an Excel file or database. Connect and load the data to see its fields. This is the first step before combining multiple sources.

Result

You see your data fields in Tableau ready for visualization.

Understanding single data source connection is essential before combining multiple sources.

2

Foundation

Understanding data joins basics

Concept: Learn what joining data means and how it combines tables from the same source.

Joining means linking tables by matching columns, like joining customer IDs in two tables. Tableau lets you choose join types: inner, left, right, full outer.

Result

Tables combine into one larger table based on matching keys.

Knowing joins helps you understand how data from one source can be combined before moving to multiple sources.

3

Intermediate

Using multiple data sources in Tableau

Concept: Learn how to connect and use more than one data source in a Tableau workbook.

In Tableau, add a second data source by clicking 'Add' in the Data pane. Each source appears separately. You can create views using fields from either source.

Result

You have two or more data sources loaded and ready to use in your workbook.

Knowing how to add multiple sources opens the door to richer analysis combining different data.

4

Intermediate

Data blending to combine sources

🤔Before reading on: do you think data blending merges data like a join or works differently? Commit to your answer.

Concept: Data blending lets you combine data from different sources on a common field without physically joining tables.

When you use fields from two sources in one view, Tableau blends them on a shared key (like Customer ID). The primary source drives the view, and the secondary adds related data.

Result

You see combined data from both sources in one visualization without merging tables.

Understanding blending clarifies how Tableau handles multiple sources without complex joins.

5

Intermediate

Differences between joins and blends

🤔Quick: Is data blending done before or after aggregation? Commit to your answer.

Concept: Joins combine data at the row level before aggregation; blends combine aggregated data after each source is processed separately.

Joins happen in the data source, merging rows. Blends happen in Tableau after aggregation, linking summarized data. This affects performance and results.

Result

You understand when to use joins or blends based on data structure and analysis needs.

Knowing this difference helps avoid wrong results and performance issues.

6

Advanced

Using calculated fields across sources

🤔Before reading on: Can you create a calculation that uses fields from two different data sources directly? Commit to your answer.

Concept: Calculated fields usually work within one data source; blending limits cross-source calculations.

You can create calculations in the primary source using its fields. To combine fields from different sources, you often need to blend or use data prep tools before Tableau.

Result

You know the limits of calculations across sources and how to work around them.

Understanding calculation limits prevents confusion and guides data preparation.

7

Expert

Optimizing performance with multiple sources

🤔Quick: Does blending multiple large data sources always improve performance? Commit to yes or no.

Concept: Using multiple sources can slow down Tableau if not managed well; optimization is key.

Techniques include using extracts, minimizing data in secondary sources, indexing keys, and avoiding complex blends. Sometimes, pre-joining data outside Tableau is better.

Result

Your dashboards run faster and are more responsive despite multiple sources.

Knowing optimization strategies is crucial for real-world scalable Tableau solutions.

Under the Hood

Tableau connects to each data source independently. For joins, it pushes the merge operation to the database or data engine before loading data. For blending, Tableau queries each source separately, aggregates data, then links results on common fields at the visualization layer. This separation allows flexibility but can limit cross-source calculations.

Why designed this way?

Tableau was designed to be flexible with many data types and sources. Joins require data to be in the same system, which is not always possible. Blending allows combining data from different systems without moving or merging data physically, trading some functionality for flexibility.

┌───────────────┐       ┌───────────────┐
│ Data Source A │──────▶│ Tableau Data  │
│ (e.g., SQL)   │       │ Engine        │
└───────────────┘       └───────────────┘
                             ▲
                             │
┌───────────────┐       ┌───────────────┐
│ Data Source B │──────▶│ Tableau Data  │
│ (e.g., Excel) │       │ Engine        │
└───────────────┘       └───────────────┘

Data engine blends aggregated results for visualization.

Myth Busters - 4 Common Misconceptions

Quick: Do you think data blending merges data tables like a SQL join? Commit yes or no.

Common Belief: Data blending is just like a join but for multiple sources.

Reality: Blending happens after aggregation and does not merge rows like joins do.

Why it matters: Assuming blending works like joins can lead to incorrect data analysis and unexpected results.

Quick: Can you create calculated fields using fields from two different data sources directly? Commit your answer.

Common Belief: You can freely create calculations combining fields from any data sources in Tableau.

Reality: Calculations usually work within one source; cross-source calculations are limited and often require blending or prep.

Why it matters: Expecting cross-source calculations without preparation causes errors and confusion.

Quick: Does adding more data sources always improve dashboard performance? Commit yes or no.

Common Belief: More data sources mean better insights and faster dashboards.

Reality: More sources can slow performance if not optimized properly.

Why it matters: Ignoring performance impact leads to slow, frustrating dashboards.

Quick: Is it always better to blend data rather than join? Commit your answer.

Common Belief: Blending is always the best way to combine data from multiple sources.

Reality: Joins are better when data is in the same system; blending is a workaround for separate systems.

Why it matters: Choosing blending over joins blindly can cause inefficiencies and limit analysis.

Expert Zone

1

Blending uses the primary data source as the anchor, so the choice of primary source affects results and performance.

2

Joins push computation to the database or extract engine, while blending does computation in Tableau, impacting speed.

3

Data blending aggregates data before combining, so detail-level joins are not possible with blending.

When NOT to use

Avoid blending when data resides in the same database; use joins instead for better performance and richer analysis. For very large datasets, consider data warehouse solutions or Tableau Prep to combine data before Tableau.

Production Patterns

Professionals often use blending for quick ad hoc analysis across systems but rely on pre-joined or prepared data extracts for production dashboards. They optimize blends by limiting secondary data size and carefully choosing primary sources.

Connections

Data Warehousing

Builds-on

Understanding multiple data sources in Tableau helps appreciate why data warehouses centralize data to simplify analysis.

ETL (Extract, Transform, Load)

Complementary process

ETL prepares and combines data before Tableau, reducing the need for complex blending and improving performance.

Cooking Recipes

Similar pattern

Combining ingredients from different stores to make a meal is like blending data from multiple sources to create one report.

Common Pitfalls

#1 Trying to join tables from different data sources directly in Tableau.

Wrong approach:

Drag tables from different sources into the join area and expect a join to work.

Correct approach:

Use data blending or prepare joined data outside Tableau before importing.

Root cause:

Misunderstanding that joins require tables to be in the same data source.

#2 Creating calculated fields that reference fields from two different data sources directly.

Wrong approach:

SUM([SourceA].[Sales]) + SUM([SourceB].[Profit]) in one calculated field.

Correct approach:

Use blending to combine data or create separate calculations per source and combine results in the view.

Root cause:

Not knowing Tableau limits cross-source calculations.

#3 Adding many large secondary data sources without optimization.

Wrong approach:

Blending multiple large Excel files without extracts or filters.

Correct approach:

Use extracts, filter data, or pre-aggregate before blending.

Root cause:

Ignoring performance impact of multiple large sources.

Key Takeaways

Multiple data sources let you combine data from different places to get a fuller picture in Tableau.

Joins combine tables within the same source before analysis; blending links aggregated data from separate sources after loading.

Blending is flexible but limits cross-source calculations and can impact performance if not optimized.

Choosing the right method and optimizing data sources is key for fast, accurate Tableau dashboards.

Understanding these concepts helps you build powerful reports that connect data across your business.

Practice

1. What is the main purpose of using multiple data sources in Tableau?

easy

A. To create multiple dashboards separately

B. To combine data from different places into one view

C. To increase the size of a single data file

D. To export data to Excel automatically

Solution

1. Step 1: Understand the concept of multiple data sources

   Multiple data sources allow Tableau to bring together data from different origins.

2. Step 2: Identify the main benefit

   This lets users analyze combined data in one view without merging files physically.

3. Final Answer:

   To combine data from different places into one view -> Option B

4. Quick Check:

   Multiple data sources = Combine data [OK]

Hint: Multiple data sources combine data from different places [OK]

Common Mistakes:

• Thinking it only creates separate dashboards
• Believing it increases file size
• Confusing it with exporting data

2. Which of the following is the correct way to create a relationship between two data sources in Tableau?

easy

A. Drag a field from one source to a matching field in the other source

B. Use the Data menu to export both sources to CSV

C. Manually merge the data files outside Tableau

D. Create a calculated field with SUM() in both sources

Solution

1. Step 1: Identify how Tableau links data sources

   Tableau uses relationships by matching fields between sources to combine data logically.

2. Step 2: Recognize the correct method

   Dragging a field from one source to a matching field in another creates this relationship inside Tableau.

3. Final Answer:

   Drag a field from one source to a matching field in the other source -> Option A

4. Quick Check:

   Relationship creation = Drag matching fields [OK]

Hint: Drag matching fields to create relationships [OK]

Common Mistakes:

• Trying to export data to CSV for relationships
• Merging files outside Tableau instead of inside
• Using calculated fields incorrectly for relationships

3. Given two data sources: Sales and Targets, blended on 'Region', what will happen if a region exists in Sales but not in Targets when blending?

medium

A. The region will be excluded from the view entirely

B. Tableau will throw an error and stop loading data

C. Sales data for that region will show, but Target values will be null

D. Target values will be duplicated for that region

Solution

1. Step 1: Understand blending behavior with unmatched keys

   When blending, if a key exists in the primary source but not in the secondary, Tableau shows primary data but leaves secondary fields blank.

2. Step 2: Apply this to the region example

   Sales data for the unmatched region appears, but Targets show null because no matching region exists there.

3. Final Answer:

   Sales data for that region will show, but Target values will be null -> Option C

4. Quick Check:

   Blending unmatched keys = Primary shows, secondary null [OK]

Hint: Unmatched blend keys show primary data, secondary null [OK]

Common Mistakes:

• Assuming unmatched regions are excluded
• Expecting errors on unmatched keys
• Thinking secondary data duplicates

4. You created a blend between two data sources but notice the data is not matching correctly. What is the most likely cause?

medium

A. You used a calculated field instead of a dimension

B. You forgot to refresh the data extract

C. The dashboard has too many sheets

D. The linking fields have different data types or mismatched values

Solution

1. Step 1: Identify common blending issues

   Blending depends on matching fields; if data types or values differ, matches fail.

2. Step 2: Recognize the cause of mismatched data

   Different data types (e.g., string vs number) or inconsistent values prevent correct blending.

3. Final Answer:

   The linking fields have different data types or mismatched values -> Option D

4. Quick Check:

   Blending mismatch = Data type/value mismatch [OK]

Hint: Check data types and values of linking fields first [OK]

Common Mistakes:

• Blaming dashboard complexity
• Ignoring data refresh needs
• Confusing calculated fields with linking fields

5. You want to analyze sales data from two different databases: one with daily sales and another with monthly targets. How should you combine these in Tableau to compare performance effectively?

hard

A. Use relationships to link the data sources on the month field and blend the data

B. Join the two databases on the day field using an inner join

C. Export both datasets to Excel and merge manually before importing

D. Create separate dashboards for daily sales and monthly targets without linking

Solution

1. Step 1: Identify the best way to combine different granularity data

   Daily sales and monthly targets have different time levels, so joining on day would exclude unmatched data.

2. Step 2: Use relationships and blending on month field

   Linking on month lets Tableau combine data logically without losing unmatched days, enabling effective comparison.

3. Final Answer:

   Use relationships to link the data sources on the month field and blend the data -> Option A

4. Quick Check:

   Different granularity = Use relationships and blend [OK]

Hint: Link on common granularity (month) and blend data [OK]

Common Mistakes:

• Joining on day excludes unmatched data
• Merging outside Tableau wastes time
• Keeping dashboards separate loses combined insight