20 Top Tips for Power BI Beginners!

Author: Graeme Hills

Power BI is a great software. Its powerful tools allow you to integrate, interrogate, and visualise your data to spot hidden trends and really understand what your data tells you about your organisation. It is one of the best, or if not, the best Business Information software going.

However, one of the barriers for more user uptake is the initial learning curve. You must learn a new coding language, which is only used for Power BI (DAX) as well as good knowledge of another coding language called M query for transforming your data.

There is so much conflicting information about Power BI that it can be very confusing for beginners.

To help we’ve made a Top Tips guide for people starting to learn Power BI as well as a reminder for more experienced Power BI users!

Our Tips:

1.  Always keep the end-user in mind and make sure that the amount of information on each page is limited, coherent and related. It is better to have a few clear and informative visuals that tell a story than a confusing jumble of graphs.  White space is your friend!

2. Have a clear idea of what the Report is set out to do. If your report is growing organically over time, then don’t be afraid to prune away those visuals that weren’t as informative as you first thought.

3. If your report is getting bloated, consider splitting it into different reports. I would stick to a maximum of 5 pages for each report.

4. Create a new version of your .pbix file every time you make a significant change to the report. That will allow you to go back and compare if something breaks.

5. Layout out DAX code in a user-friendly way with indentation. You will thank me later when you go back to modify some code six months after you have written it.  Which of the following do you find easier to understand?

6. DAX functions should be in capitals. While this is stylistic (you can use lower or mixed case), it again comes down to being able to quickly understand what your code is doing six months after writing it.

7. Use of variables when complex calculation come into place – they are more efficient in terms of code, more readable, and easier to de-bug and reuse.

8. Use comments where the function of the code isn’t obvious or requires business knowledge that a future developer may not have.

9. If you changing the code update the comments!

10. I always import the raw data into one table, then create a separate filtered table for processing and visualisation. This allows you to go back and check the raw data should your report show some odd results.  Ensure that you don’t make any changes to the raw data tables, such as adding columns.

11. I find it useful to number these created tables, such as ’01 Financial Data’. It keeps them at the top of the table list and easily differentiates them from the raw data tables.  Similarly, you can number any calculated columns and measures in a similar way, e.g. ‘01.02 Days Billed Last Week’, by combining the table number and adding a column number.

12 Always make sure you clean your data before hand importing it into Power BI.

13. Add the measure under the table you are creating it for.

14. Although if there are many measures, you may want to have a table or tables to specifically store them.

15. Double check the report after changes are made.

16. Before publishing always double check that:

    1. Every visual is working.
    2. Tables are ordered by the correct column.
    3. Reset any bookmarks.
    4. Save the report on the page it should be opened on.
    5. Date hierarchies have not reverted to dates in your slicers.
    6. Spell-check the report.

17. As best practice, it is always better to extract data from a database rather than from an Excel spreadsheet. The connection to Excel is really fragile.  If the filename or a column name is changed, the report will break.  If a column is added or removed, the report will break.  If the wrong data type is used in a column, e.g., text instead of a date, the report will break.  I have spent may hours fixing reports linked to Excel spreadsheets. 

18. If you must use Excel, then try using data validation and limit write access to the file.

19. Use JSON Themes to keep the layout consistent.

20. When you import tables from a database, Power BI will try to automatically create relationships between them. This can lead to strange results within your data.  I would disable that functionality and only create relationships that you specifically want. 

To Conclude:

Power BI for beginners can be confusing, but with these top tips you’ll be able in a much better position to make exceptional reports for you and your end-user. If you want to know more about how we use Power BI, head to our Power BI specialisms page!

If you have any questions or wish to discuss this topic in more detail, then you are more than welcome to contact me with any of the following contact details at the end of this article.

Graeme Hills PhD.