Skip to content Skip to footer

5 tips and best practices when creating Power BI reports

Designing an effective Power BI report can be a challenge. There are many different ways to approach the design process, and it can be difficult to know where to start. However, there are some general best practices that can help you create a successful Power BI report. In this article, we’ll share some of our top tips for designing effective Power BI reports. By following these best practices, you’ll be able to create reports that are both informative and visually appealing. So let’s get started!

Less is more.

When it comes to Power BI Reports, less is more when comes to visuals on a page. This is because reports that are overloaded with too much data can be confusing and difficult to read. When creating a report, it is important to focus on the most important information and present it in a way that is easy for the reader to understand.

Instead of having tons of filters on every page, consider creating additional tabs and pages geared to a specific role or task.   Users can get lost when there are more than one or two filters on a page or if they feel a page has too much information that is not applicable to them.    

Create pages geared towards something that someone would be looking at.  (e.g. “Executive Overview”, “Sales by Territory”, “Sales by Manager”).  Instead of creating one page with lots of filters and visuals.  

If your users do need to be able to do lots of filters, usually its for exploration.  Create a tab in your Power BI report called “Explore” that has all the option to have all the filters they could possibly want.  Make it easy to find what the users are looking for.  

Don’t reinvent the wheel everytime!

Power BI has some really useful features that allow you to reuse datasets across multiple reports.  Strive to create a “shared dataset” that has all the measures, calculations and data you need to power multiple reports.  Then you can create multiple reports off the same dataset, instead of separately pulling in the same information.

Also, if you feel you need to create lots of reports.  (e.g. a separate dashboard for a specific office or role) consider using Row Level Security instead.   Instead of creating a copy of a dashboard for each audience, you can filter data based on who’s looking at the report.

Row Level Security (RLS) allows you to create one report that filters data depending on who’s looking at it.  It can be setup so an area manager only sees data from their office, a senior manager can see data from multiple locations and a vice president can see things from all offices.

Also create a template that you can reuse when creating reports.  A good template should include the companies branding, color scheme and some useful features like a date table or measure table.

 Do the “eye test”
Studies show that most of the time, the human brain will look at the top left of a report then look down then back to the right.  It’s good to use the “F” approach when designing a report.  All your important visuals should be at the top left. Less important visuals will be on the right or bottom. 
 

 

 Understand the “why” not just the “what”

It’s important to understand the “why” not just the “what” the user is asking for.  For example, if a user asks for a “bar graph that shows sales for the year“, that almost sounds self-explanatory.   However, it might make more sense to ask “why would like see that“?  Or “How will the data be used to drive decisions“?   Then the user might say “I want to see how sales for the entire year so I can see if sales are going up or down each month.”   With that understanding, we’ve learned something.  The user is looking for a trend in sales (whether they go up or down) and will be analyzing the change between month over month. 

So it might make sense to use a line graph to show the trend and either a tooltip or another visual that shows the % change and $ amount change for each month.   Otherwise, they’d be looking at a bar graph, and having to do the math in their heads.   E.g. the user might be looking at the graph and saying in their heads… January was $2.17 million, February was $2.3 million, that means sales went up 130k since January.    Or you can display a visual that displays the difference in sales in February compared to last month, and even the year prior for them so they don’t have to do the math. 

 Make it easier for your end users!  Let them help themselves

Most end users want to be self-sufficient.  One tip is to give them the tools they need to do it!   Teach them what information is being displayed.  Create help information by using buttons in Power BI that bring a popup with information about what is being displayed.    Create a page that explains where the data is coming from and a place they can reach out to for assistance.  Create a tab for notes, and  any updates to the report.    Enable the feature in PBI Desktop and on the service so users can personalize their own visuals.  Teach users where the reset to default button is.  (It is a life saver! saves so many tickets).

 

 

Add Drillthroughs
Rather than displaying an overwhelming amount of data on one page, use the Drill through feature in Power BI to allow your users to add filters and then be redirected to another page with more specific graphs and a table with the underlying information. This helps keep your report focused, but still is able to provide the detail users want.
Maximize performance 
Another key factor to consider when building a report in Power BI is performance.  The last thing you want to do is create a beautiful report, but it never works or it takes too long or hits capacity limit.   Tips for improving performance are to use the Performance Analyzer in Power BI desktop to monitor DAX performance.  When refreshing data use incremental refresh to avoid loading the same data that never changes over and over again.
Need help building reports? AnalyticsPHX can help!