Best PowerBI Data Visualizations

After creating dashboards for 600+ clients as part of my consultancy I have created the list of recommendations for creating the best visualizations for Power BI. I have used these visualization principles to create Power BI dashboards for clients like Google, Teleperformance and the UK Ministry of Defence.

In this article I will show the screenshots of some of best visualization for Power BI that I created and explain when those visualizations are most effective. We will discuss each data visualization type in detail and the situations when they are most effective.

The short answer is that the best visualization for Power BI is the one that clearly answers the business question asked. Therefore, the majority of this article will focus on identifying the best visualization for every type of questions asked.

Let’s begin by discussing the what Power BI is and how it helps you to create best visualizations.

What is Power BI

Power BI is a market-leading data visualization software that was created by Microsoft. You can think of Power BI as Excel on steroids. 

Power BI also enables you to create the pivot charts, same way as Excel. However, you can also write more complex formulas, create better visualizations and analyse larger amounts of data.

In addition, Power BI has made big noize in the industry by positioning itself as the low-code technology. This enables you to create best visualisations for your Power BI report without writing any code!

‍

Here is an extensive guide on Power BI, popular features, components and real-world case studies. Feel free to read it in more details to learn more about Power BI and evaluate it for your organization.

‍

Criteria to Determine Best Visualization for Power BI 

Best Visualization for Power BI needs to fulfil the following criteria:

• Be easy to read so that your audience can quickly understand the communicated analysis
• Be intuitive to understand so that no technical knowledge is needed to interpret the analysis.
• Be visually engaging so that your audience don’t get bored looking at your graphs. 

Now that we have established the criteria, let’s look into each data visualization type and discuss the most appropriate time to use each one of them.

15 Best Visualizations for Power BI

Cards - Quick Communication of Key Metrics

Card visualization is one of the best visualizations for Power BI when you want to quickly communicate the most important KPIs. 

Imagine yourself as a CEO of a company. At any time you will have 3-4 metrics that will tell you whether your business is doing good or bad. When you open your Power BI dashboard you often want to quickly learn your performance on these key metrics. This is exactly what card visualizations are used for.

Card visualizations are often placed at the top and are the first thing that your users read after landing on your dashboard.

The downside of card visualizations is that they provide no context for your analysis. For example if your card visual shows income of $3M USD, is it good or bad? If you previously made $10M USD in the same time period, this is really bad but if you made $2M USD it is good!

‍

This brings us to the next visualisation type that allows you to add context to your numbers.

‍

KPI Cards - Context-Driven Insights

KPI Cards are the best visualization for Power BI when you want to save space and concisely present key insights. They have the same use case as KPI cards while enabling you to add extra formatting features. 

Power BI recently introduced the Card (new) visuals that enable developers to add trends, arrows and many other useful formatting elements!

I would encourage you to use the KPI cards if your users ask you to add more context to your cards graphs!

Line Charts - Change Over Time Visualization

‍

Line charts are the best visualization for Power BI when measuring the change over time for 2 reasons:

1. Even when the change over time is minimal, the line slope will make it clear whether your numbers are going up or down. 
2. Even when your lines are close together you will still be able to easily see which line is higher. This is not the same with other type of trend charts.

Line charts are only used to visualise change over time. Do not use them unless you X axis values imply some sort of trend! You don’t want to mislead your audience into believing that such trends exist when in reality they don’t.

Vertical Bar Charts - Data Distribution Analysis

Many data analysts would argue that vertical bar charts are the best visualization for Power BI when you analyse distribution. 

Every one of us has seen a tonne of vertical bar charts throughout our lives which means that we immediately know how to interpret them. As a result, vertical bar charts are easy to read and don’t require any technical knowledge to interpret!

Vertical bar charts are also effectively used when you need to visualize trends. For example, consider the cash flow vertical bar chart below.

‍

The rule of thumb for using vertical bar charts is that your items in the X-axis must be sorted in a logical order. For example months in the X-axis should be sorted chronologically. Without this condition vertical bar charts become really difficult to read!

Vertical Line and Column Charts - Dual Trends

Vertical line and column charts are the best visualisation for Power BI when you have 2 separate trends to show. This visualization type allows you to have 2 separate Y axis which is useful when you have 2 different units of measurement.

Consider the chart above where we are visualizing attendance and attendance %. This chart allows us to use the columns for attendance and the line as percentage making both metrics easy to read!

Horizontal Bar Charts - Categorical Breakdown

Horizontal bar charts are the best visualization for Power BI when analysing the categorical breakdown of your KPI. They offer multiple benefits over vertical bar charts:

1. It is much easier to read long category names. Consider the chart above. Imagine how hard it would be to tilt your head and try to read these category names in a vertical bar chart. 
2. The horizontal bar charts don’t imply any logical order of the categories. Therefore, if your categories have no logical order to them, the vertical bar charts would work great!
3. Horizontal bar charts look tidy even when you have a lot of categories.  

Pie Charts - Intuitive Distribution for Small Categories

Pie charts are arguably the most hated visualisation method in the data analysis community. Many would advise you to never use pie charts at all. There are several arguments that support this point of view:

1. Pie charts look messy when they have a lot of categories for the breakdown. You often can not see the category names when you have lots of categories in the pie chart. 
2. Pie charts make it difficult to visually compare segment sizes. It is often easier to read the same information in a horizontal bar chart. 

Personally, I am not a pie chart hater. I feel pie charts just get misused a lot but the visualisation itself isn’t flawed. Pie charts also offer several advantages:

1. You can show both % and real numbers in the pie chart data labels. You can’t do this in any type of bar chart.
2. In my opinion pie charts are the most intuitive chart to read when you want to see distribution for 2-3 categories. They are great to visualise the answers to Yes/No questions or showing Male/Female split.

Donut Chart - Space for Additional Insights

Donut charts are just another version of a pie chart. They have exactly the same benefits and drawbacks as pie charts which we won’t repeat in this section. The only difference is empty space in the middle of the chart. 

You can make use of this empty space by placing a card visual inside. This way you will see the total for the analysed metric and it’s distribution across 2-3 categories.

‍

Treemap - Button-Like Filtering

‍

Another variation of a pie chart is a treemap. Again, it has all the same benefits and drawbacks.

An advantage of a treemap is that it offers a button-like look. This means treemaps an intuitive visual for not only visualising distribution but also filtering all the other graphs.

A disadvantage of a treemap is that the small boxes make it difficult to read the text. An abundance of colors when you have many categories also makes them look intidy. 

I would still use treemaps when you have 2-4 categories and you want to create interactive buttons to filter other graphs.

Table/Matrix - Most Common Visualization

I feel like tables are the most underrated visualization for Power BI! Tables offer several key advantages as compared to other visualizations in Power BI!

• They look clean. No matter how many categories you have they look tidy and easy to read!
• They are extremely customizable. You can format the numbers inside of your tables more than in any other visualization!  For example you can put the negative numbers in brackets, show both % and $ amounts in the same column, etc.
• Even small numbers are visible. In most data visualizations small numbers get lost and become barely visible but not with tables!

I would recommend to use the tables for detailed breakdowns and visualizing more than 3 metrics on the same graph. 

For example the table below allows the users to have a clear visibility of numbers no matter if they are small or large. It also makes the analysis easy to read even though there are multiple metrics in the same graph.

Funnel - Visualizing Step-by-Step Processes

Unsurprisingly a funnel chart is the best visualization for Power BI when there is a funnel-like business process. A very common mistake that I see is that funnel charts are used as another version of a bar chart. This is misleading the users into thinking that there is a funnel-like relationship between the categories when there isn’t one.

I would encourage the use of funnel charts for visualising business processes like sales pipeline or a hiring pipeline. 

In addition, funnel charts allow to neatly compare the values in each funnel step vs the previous step. This offers a tidy way of visualising conversion rate in Power BI.

Bubble Maps - Engaging Geographic Insights

Bubble maps are the best visualization for Power BI when your goal is to show the geographic distribution in an engaging way. Maps make your Power BI page look a lot more engaging in general!

Power BI uses bing maps integration which means that simply having city and country names in enough to create a map. 

The disadvantage of bubble maps is that bubble sizes may be difficult to compare to each other. Because of this it is ofter a good idea to supplement a map with a table. 

An advantage of bubble maps is that you can turn your bubbles into mini pie charts adding an extra layer of analysis to the same visual!

Filled Maps - Colour Density based Geographic Comparisons

Filled map are the best visualization for Power BI when you want a tidy way to show geographic distribution. In my opinion they look tidier than bubble maps since many bubbles overlapping often overwhelm the users. I feel that it is easier to compare the color gradient than bubble sizes.

I would encourage you to use filled maps wherever you need to compare different regions of the same country. I would also encourage you to add a table next to your map to make the numbers easier to read.

The Best Visualization for Power BI Depends on Context

As you can see the choice of the best visualization for Power BI depends on the business question. I hope that this article helps you determine which Power BI visualization are best for your particular use case and that's why you need PowerBI Data Connectors.

If you are still unsure how to effectively visualise your data in Power BI, it might be worth it for you to speak to a professional Power BI consultancy. After completing dashboards for 600+ clients we can quickly understand your data, business questions and the effective way to visualise it and even run automations on it using Konnectify.

I hope that you enjoyed this article and hoping to see you again on the Konnectify blog!