Before diving in, it’s worth saying that there are many reasons why an organisation might choose Power BI over Tableau, or vice versa. Factors such as cost, existing technology stacks and organisational expertise all play a role. This post isn’t trying to answer the question of which tool is better for an organisation.
Instead, I want to look at the comparison from a very specific perspective:
- Someone currently at The Data School
- Someone who has more experience with Tableau
- Someone who is still relatively new to both tools
In other words, this is less a technical deep dive and more a reflection on what it feels like to learn and use both tools as a beginner.
First Impressions
Tableau
- As Tableau is used in the application process to The Data School, and is a primary tool here at The Information Lab, I have more experience with it and so can build visualisations faster with it.
- The interface encourages jumping straight into visual exploration, making it easy to experiment and iterate quickly.
Power BI
- Initially felt slightly less intuitive, largely because I have less experience with the tool and am still learning where different features live.
- As a Microsoft product, the ribbon-style interface feels similar to tools like Excel or PowerPoint, which may make it easier for users already familiar with those environments.
- Building visuals often involves working through menus and configuration panes, which can feel a lot slower at first compared to Tableau.
Data Preparation
Tableau
- Tableau Desktop can perform some data preparation (joins, unions, calculated fields), but it generally works best when the data is already relatively clean and well-structured.
- In practice, many Tableau workflows rely on data being prepared beforehand using tools like Tableau Prep.
Power BI
- Power BI includes Power Query, which makes data preparation a core part of the workflow.
- Cleaning steps such as filtering rows, pivoting columns, splitting fields, etc., can all be done directly and easily within the tool.
- This makes Power BI feel more like a one-stop-shop for both data preparation and dashboard creation.
Data Modelling
Tableau
- Data modelling tends to feel less central when starting analysis. It is often possible to connect to a dataset and begin building charts immediately.
- Tableau relationships can use multiple fields (composite keys) between tables, allowing more flexible links between datasets.
- Multiple relationships between tables can exist without needing to force a single active relationship, giving more flexibility when analysing data from different angles.
Power BI
- Data modelling is much more central to the workflow, with clear separation between the data model, report view, and transformation layers.
- Relationships are typically defined using a single column, meaning composite keys often need to be created manually (for example by concatenating fields).
- Only one active relationship can exist between two tables at a time, with additional relationships activated through DAX when needed.
Calculations
Tableau
- Calculations have felt easier for me to write in Tableau. This is partly because I’ve written more of them, but also because the syntax feels closer to natural language and SQL-style logic, which makes them easier to read and structure.
- Conditional logic is very clear. For example, an IF / ELSEIF / ELSE statement can be written naturally in one block, making longer logical calculations easier to follow.
- Table calculations make analysing data directly within the visualisation very easy, allowing quick comparisons such as running totals, differences, and percentages without changing the underlying data.
Power BI
- Calculations in Power BI rely on DAX, which I have found harder to pick up compared with Tableau calculations.
- Writing conditional logic can feel less intuitive. For example, a simple IF / ELSEIF / ELSE structure typically requires nested IF functions, which can quickly become harder to read.
- More complex calculations often require a better understanding of how measures and context work, which adds an additional learning curve when first using the tool.
Dashboarding
Tableau
- Formatting feels much easier and more accessible. Options can be adjusted directly on the view, by interacting with fields in the data pane, or through formatting menus, which makes styling dashboards relatively quick.
- Tableau uses layout containers, which people often either love or hate. They can be a bit finicky, but they allow much greater control over the positioning and sizing of objects on a dashboard.
- Tableau’s flexibility makes it strong for highly customised dashboard layouts, especially when precise control over spacing and alignment is important.
Power BI
- Formatting options can feel more buried, often requiring navigation through several dropdown menus in the formatting pane to locate specific settings.
- Visualisations tend to snap to a grid, and without containers it can sometimes feel less precise when trying to control the layout of objects on the page.
- Power BI includes some useful built-in visual types, such as KPI and card visuals (BANs), which make it very easy to create simple headline metrics.