Tableau Interview Questions
Updated: Jun 20
What is LOD?
Tableau Level of Detail (LOD) expressions provide a way to calculate aggregation at different levels of granularity than the view’s default aggregation. This is particularly useful when you want to perform an analysis at a level of detail that isn't available in the original data source.
LOD expressions can be used to create complex calculations that are not possible with regular aggregations, such as nested aggregations, moving averages, and percent of total calculations. LOD expressions can be defined using the curly braces { } syntax and they can be created at three levels of granularity:
Fixed Level of Detail (Fixed LOD): This allows you to compute an aggregation at a specific level of granularity, regardless of the level of detail of the view. Fixed LODs are created using curly braces followed by the keyword "FIXED" and then the dimensions that define the level of detail.
Include Level of Detail (Include LOD): This allows you to include additional dimensions in the view without affecting the aggregation level of the existing dimensions. Include LODs are created using curly braces followed by the keyword "INCLUDE" and then the dimensions to be included.
Exclude Level of Detail (Exclude LOD): This allows you to exclude dimensions from the view without affecting the aggregation level of the remaining dimensions. Exclude LODs are created using curly braces followed by the keyword "EXCLUDE" and then the dimensions to be excluded.
By using LOD expressions, you can customize your view and perform more sophisticated analyses without having to modify the underlying data source or schema. LOD expressions can be used in various types of calculations, including basic arithmetic calculations, date calculations, and string calculations. They can also be combined with other Tableau functions and operators to create even more complex calculations.
Utilize a Speedy LOD articulation
You can make a Decent LOD articulation without expecting to enter the full estimation into the computation exchange.
There are two methods for making a speedy LOD computation.
In the Information sheet, control-click drag the action you need to total onto the ideal aspect. Another field shows up as a Proper LOD computation.
The total in the total articulation will come from the default conglomeration on the action. This is normally Aggregate. To change the conglomeration or in any case alter the LOD, right snap on the new field and alter the estimation.
Or on the other hand, in the Information sheet, select the action you need to total and afterward control-click the aspect you need to total on.
Right-click on the chose fields and select Make > LOD Estimation...
(Discretionary) Adjust the LOD in the estimation proofreader.
Select alright .
How to Improve Tableau Dashboard Performance?
Tableau dashboards can sometimes suffer from performance issues, especially when dealing
large data sets or complex calculations. Here are some tips to improve dashboard performance:
Optimize data sources: Ensure that your data source is optimized for Tableau by minimizing the number of rows, reducing the number of columns, and removing any unnecessary calculations.
Use data extracts: Data extracts can improve performance by pre-aggregating the data and reducing the amount of data that needs to be processed. Extracts can also be refreshed on a schedule, reducing the load on the data source.
Limit the number of worksheets: Having too many worksheets on a dashboard can slow down performance. Try to limit the number of worksheets and use filters or parameters to allow users to interact with the data.
Simplify calculations: Complex calculations can slow down performance. Simplify calculations by breaking them down into smaller parts or using LOD expressions.
Use filters and parameters: Filters and parameters can improve performance by reducing the amount of data that needs to be processed. Use them to allow users to focus on specific aspects of the data.
Avoid using live connections: Live connections can be slower than data extracts, especially when dealing with large data sets. Use data extracts whenever possible.
Optimize dashboard design: Dashboard design can also affect performance. Avoid using large images or background colors that may slow down rendering.
Upgrade hardware: If performance is still an issue, consider upgrading hardware such as RAM or CPU to improve processing speed.
By following these tips, you can improve the performance of your Tableau dashboards and provide a better user experience for your audience.
Â
Types of Tableau Filters:
Extract Filters
Data Source Filters
Context Filters
Dimension Filters
Measure Filters
From the original data source, a limited subset of data is extracted using the Extract filters in Tableau. Tableau provides us with two options when we connect to a data source: Live or Extract. Tableau makes a local copy of the subset of data it receives when we extract data from a data source in its repository. Extract filters are used to apply a filter on data that has been extracted from the data source. Data is taken from the data source and added to the Tableau data repository for this filter. Datasource Filters - Extract filters also refer to datasource filters. They work with the extracted dataset as well. However, the sole distinction is that it supports both live and extract connections. Contextual filters: Contextual filters are used on the data rows prior to the application of any other filters. They can only be used on specific sheets, and they are only applicable to views. In Tableau, they explain how to aggregate and disaggregate data. Dimension Filters - Worksheet dimensions can be filtered using dimension filters. The top or bottom conditions, formula, and wildcard match are used as dimension filters. Measure Filters - The values in the measurements are subjected to the application of measure filters.
what are tableau blending limitations?
While Tableau is a powerful data visualization tool, it does have some limitations when it comes to data blending. Here are a few common limitations:
Data structure: Tableau has difficulty blending data with different structures or granularities. For example, if you have two datasets with different levels of detail or incompatible dimensions, blending them can be challenging.
Performance: Blending large datasets can impact performance. When blending data from multiple sources, Tableau may need to process and join a significant amount of data, which can result in slower query performance and increased memory usage.
Aggregation: Tableau's data blending is limited when it comes to aggregating data. Blending works best when you have detailed or granular data at the row level. Aggregating data before blending can sometimes lead to incorrect or unexpected results.
Data freshness: Blended data sources in Tableau do not automatically update in real-time. If the underlying data in one of the sources changes, you may need to manually refresh the blended data to see the updates.
Joins and relationships: Tableau's blending functionality is not as flexible as traditional database joins or relationships. It may not support complex join conditions, such as multiple keys or non-equijoins, which can limit the ability to blend certain datasets.
Calculated fields: Blending data sources in Tableau can be challenging when it comes to creating calculated fields that involve data from multiple sources. Some calculations may require complex workarounds or the use of data blending extensions.
What are tableau scenario based interview questions?
Here are a few example scenario-based interview questions that you might encounter in a Tableau interview:
Scenario: Imagine you are working on a project for a retail company, and they want to analyze their sales data to identify trends and patterns. How would you approach this task using Tableau?
Scenario: You are given a dataset containing customer information, including their demographics and purchasing history. How would you use Tableau to segment the customers based on their behavior and create a visual dashboard to present the findings?
Scenario: A company wants to track the performance of its marketing campaigns across different channels. How would you use Tableau to create a dashboard that provides insights into the effectiveness of each campaign and helps optimize the marketing strategy?
Scenario: You have been given a dataset with information about website visitors, such as their location, time spent on the website, and actions taken. How would you use Tableau to analyze this data and create visualizations that provide insights into user behavior?
Scenario: Imagine you are working with a large dataset that contains information about a company's inventory, including product names, quantities, and sales. The company wants to identify which products are running low on stock and need to be replenished. How would you use Tableau to analyze this data and create a report that highlights the products that require immediate attention?
What are Tableau level of detail limitation ?
Tableau has a concept called "Level of Detail" (LOD) expressions that allow you to perform calculations at different levels of granularity in your data. LOD expressions are powerful and flexible, but they do have some limitations. Here are a few important limitations to be aware of:
Aggregation limitations: LOD expressions can only be used to aggregate data at a higher level of detail than the level of detail of the view you are working with. You cannot use LOD expressions to compute aggregations at a lower level of detail than what is currently displayed.
Fixed level of detail: LOD expressions define a fixed level of detail, meaning they cannot be dynamically changed based on user interactions or filters applied to the view. LOD expressions are evaluated independently of any filtering or user interactions.
Context filters: When using context filters in Tableau, LOD expressions are computed before the context filter is applied. This means that LOD expressions are not affected by the context filter, and the filtering is applied after the LOD calculations.
Performance considerations: Depending on the complexity and number of LOD expressions used in a workbook, performance can be affected. Using LOD expressions extensively or in complex calculations can result in slower query performance and increased memory usage.
Cross-database limitations: If you are working with multiple data sources in Tableau, LOD expressions may have limitations when it comes to blending data across different databases. Some LOD expressions might not work as expected when blending data.
What are limitation of data blending in tableau
Data blending in Tableau is a powerful tool that allows you to combine data from multiple sources into a single view. However, there are some limitations to data blending that you should be aware of.
Data blending is not a join. When you blend data, Tableau does not actually join the data together. Instead, it creates a separate view for each data source and then merges the views together. This means that you cannot use all of the features of Tableau joins, such as filtering and aggregation.
Data blending can be slow. If you are blending large datasets, it can take Tableau a long time to create the blended view. This is because Tableau has to load and process all of the data from both data sources.
Data blending can be inaccurate. If the data in the two data sources is not compatible, it can lead to inaccurate results. For example, if the two data sources have different levels of granularity, Tableau may not be able to correctly aggregate the data.
Data blending can be difficult to troubleshoot. If you are having problems with data blending, it can be difficult to troubleshoot the issue. This is because Tableau does not provide a lot of information about how it is blending the data.
Overall, data blending is a powerful tool that can be used to combine data from multiple sources into a single view. However, it is important to be aware of the limitations of data blending so that you can use it effectively.
Here are some additional limitations of data blending in Tableau:
You cannot publish a blended data source as a single data source on the server. Instead, you need to publish the two data sources separately on the same server and then blend the published sources.
The primary data source must be a cube data source. If you are using a cube data source to blend with other sources, the cube data source must be the primary data source.
Any data brought in from a secondary data source will automatically be aggregated. This means that you cannot use non-additive aggregates in secondary data sources.
What is similarity between Level of detail (LOD) expressions in Tableau and window function in sql?
Both LOD expressions and window functions allow you to calculate values at different levels of aggregation. For example, you could use an LOD expression to calculate the total sales for each region, or you could use a window function to calculate the running total of sales for each region.
Both LOD expressions and window functions can be used to calculate subtotals and grand totals. For example, you could use an LOD expression to calculate the total sales for all regions, or you could use a window function to calculate the grand total of sales for all regions.
Both LOD expressions and window functions can be used to deal with different levels of granularity in the data. For example, you could use an LOD expression to calculate the total sales for each region, regardless of the other dimensions in the view.
However, there are also some key differences between LOD expressions and window functions:
LOD expressions are specific to Tableau, while window functions are a part of the SQL standard. This means that you can only use LOD expressions in Tableau, while you can use window functions in any SQL database.
LOD expressions are evaluated at the data source level, while window functions are evaluated at the visualization level. This means that LOD expressions are calculated once, when the data is loaded into Tableau, while window functions are calculated each time the visualization is rendered.
Overall, LOD expressions and window functions are both powerful tools for analyzing data. However, they have different strengths and weaknesses, so it is important to choose the right tool for the job.