The Complete Guide to Designing Power BI Reports

Introduction

Design is often overlooked in the process of building reports but can be a one-way ticket for ensuring your audience not only read what you share with them but remember it as well.

In this guide, you will learn why you need to implement design in your Power BI reports and how to do so effectively.

Additionally, we'll walk through the step-by-step process we use to build and design reports, along with our top design tips to ensure you're meeting best practices.

For reference, all the Power BI reports and visuals used in this handbook were built using the Numerro Toolkit, allowing for an effortless build/design process and ensuring we met design best practice standards.

‍

What we'll cover

Part 1 - Why you Need Design

Part 2 - Design Principles

Part 3 - Step-by-step Process to Building & Designing Reports

Part 4 - Design Tips Checklist

Part 5 - Power BI Report Example

‍

Why you Need Design

Design is what makes your reports stand out and your information more memorable.

The goal of implementing design is to make it easy for your users to understand the data, remember the data and take action from the data.

‍

Design Helps Unlock Insights

Implementing design into your reports makes the data easier to understand and interpret. This helps users to quickly grasp the information and enables them to unlock new insights. Through unlocking new insights, users can spot new data-driven opportunities for you or your business.

‍

Design Improves User Experience

Well-designed reports improve the end-user experience by making the information easier to digest and navigate, easing the frustration of the users.

‍

Design Adds Trust & Credibility

Design helps you to build a reputable and credible brand. Creating well-designed reports ensures that you are projecting a professional image that can be trusted. This trust will allow your users to believe and have confidence in the insights that you are presenting.

‍

Using design can save you money, time, and your reputation. Well-designed reports help your business to quickly identify trends and insights, enabling your users to find and act on data-driven insights faster. By speeding up this process you are saving money and time. Professional report design will also save your clients from getting confused and frustrated when looking at the data, thus protecting your reputation.

‍

Design Principles

Before we look at the process for building and designing your report, it's important for you to first understand the fundamental design principles.

Following these principles will help you to create clear, concise, and well-designed reports.

‍

Simplicity

Applying the principle of simplicity across your reports will allow you to tell a clear data-story that users can quickly understand and take action on. Prioritize creating a simple but accurate report that presents a carefully selected amount of data to avoid clutter and complexity.

‍

Consistency

Applying the principle of consistency across the foundations of your reports will help your audience to compare and understand information faster, whilst eliminating the risk of miscommunication. The quicker your audience can process the information you show them, the easier it will be for them to engage with and then retain the insights.

Some key consistency best practices:

• The way you represent your data in metrics, KPI's and units should be consistent across reports, using the same naming structures to avoid confusion.
• Use a consistent color scheme across your dashboards, using no more than ideally 6 colors as doing so can lead to more visual noise.
• Ensure that any icons or images that you use serve a purpose across all your dashboards, and are not just relevant to one department.
• Use the same date and text-based formats, such as DD/MM/YYYY.
‍

Integrating the principle of accuracy across your reports will help you to create the clearest depiction of the data through your insights. It's best practice to set up systems to govern the layout, structure, hierarchy, and proportions of your reports and insights. Relying on these systems will help to ensure accuracy as well as reducing the risk of deviation from design best practices.

‍

Step by Step Process to Building and Designing Reports

‍

Step 1: Defining your insights

The first step in the process is to define the insights that you want your report to drive to have a clearer understanding of the direction you're going to take when building.

‍

Audience

‍

To determine your insights, you need to first define your audience. Unless you can clearly communicate to your intended users, your report will be useless.

By understanding what your audience wants before deciding upon your insights, you'll be able to create valuable, relevant, and memorable visualizations. Considering your audience first will make sure that you don't risk spending time and money creating an irrelevant report.

‍

Purpose

‍

Defining the purpose of your report follows from identifying your audience. A clear purpose will provide you with the guidelines needed to help you focus on answering the report's most vital questions. To find the purpose of your report, start by asking yourself these questions:

Why are you building this report?

What does your audience want to see in this report?

What information do you want the user to gain from seeing this data?

What are the most important questions you're trying to answer?

‍

Overarching Insights

After you have determined the audience and purpose of your report, you'll then be able to identify the most valuable insights you want to drive.

The great thing about using Power BI is that you can visualize data more engagingly than Excel and other tools. However, it can be easy to get carried away trying to apply too many visuals to your report.

To avoid a cluttered report, it's best practice to decide upon 6 to 10 data points that you're going to visualize.

‍

Example Report Showcasing 9 Insights

‍

In the above example report, the following 9 insights are shown:

• Total Revenue - KPI
• Total Profit - KPI
• Total Orders - KPI
• Total Customers - KPI
• Total Quantity - KPI
• Performance Ratio - 100% Stacked Bar Chart
• Sales Performance Revenue by Month - Area Chart
• Region Performance split by Category - Stacked Bar Chart
• Quantity & Sales by Category - Multi-Card Row

We recommend you make a list of the 6-10 insights you want to drive as this will make it easier for you to decide on the data and match each insight with the relevant visual.

If you are looking to create a multi-page report, we recommend you make a list of all the insights you want to drive before categorizing them by relevance e.g. HR insights and Sales insights. After you have listed and categorized you can work on prioritizing the 6-10 insights per page.

Now you have a clear set of insights you want to drive, you can look into which data sets you are going to choose.

‍

Step 2: Deciding on the Data

Identifying the right data set to use ensures that your report will include all the information you need to create your insights.

The dataset you choose determines which insights you can drive, and if you don't choose correctly you will be unable to show what you originally intended which can result in a redundant report. This is why it's essential to make sure you're using the right data before you go into building the report and risk wasting time.

To evaluate whether you're using the right data in your report, identify the key pieces of information you want the audience to retain and confirm your data can provide them.

Additionally, it's within this step that you'll spend time modelling data should you need to.

‍

Step 3: Choosing your Visuals

Once you've listed the insights you want to drive, covered in the previous section, you can begin to match them with the visual that will best represent them.

To support this we've created a list of visual types and their use cases to ensure you're selecting the correct visual based on the insight.

‍

Comparisons

Comparison visuals compare data between different categories. ‍

Data Over Time

Data over time visuals represent the spread of data over a period of time and are displayed to identify trends or changes.

Correlation

Correlation visuals are used to find a correlation between different variables.

Distribution

Distribution visuals are used to show how often values occur in a dataset.

Part-to-whole

Part-to-whole visuals show the breakdown of elements that add up to a whole.

Ranking

Ranking visuals showcase an ordered list based on a unique data point. ‍

By following the guidance above, you should now have a clear idea as to which visual best suits each of the insights you're looking to showcase in your dashboard.

In the next section, we'll walk through how to take your chosen visuals and begin to build a logical report layout and structure.

Step 4: Positioning of Visuals

Where you position your visuals in your report is critical.

A consistent layout and grouping relevant metrics together will help your audience understand and absorb the data quickly.The correct layout ensures your dashboard is easy to understand and has a logical flow between different insights, which is important as users tend to process information from top to bottom.

Grouping relevant metrics together, such as KPIs, adds further to the logical report flow and the ease of user insight interpretation.

Referring to the dashboard example below, we can begin to segment our dashboard to ensure the correct layout is implemented.

When determining where to allocate each of your visuals, it's best practice to follow the guidance below:

‍

Top

The top of a dashboard should include high-level insights represented as visuals such as KPIs or Gauges and are best kept to 2-3 squares as demonstrated by the red section in the canvas grid above.

‍

Middle

The middle of a dashboard should represent trend-based data including activity-based metrics, and visuals that demonstrate data over time. This section is best suited for larger visuals and is best kept to 4-6 squares as demonstrated by the purple section in the canvas grid above.

‍

Bottom

The bottom of a dashboard is reserved for granular metrics such as specific KPIs, or Tables, and is best kept to 3-4 squares as demonstrated by the blue section in the canvas grid above.

An Example Report Showcasing A Logical Layout

‍

This step-by-step process has given you the foundations needed to build a well-designed report. The next section will look at the additional design tips that you should incorporate into your reports.

‍

Step 5: Design Tips Checklist

In this section, we'll go through a checklist of data visualization design fundamentals to ensure you're implementing best practice at each stage of the build and design process.

‍

The final stage of building a report is the design checklist. Thankfully, with the Numerro Toolkit design best practice is already built-in, meaning when you're building reports using the toolkit, you're automatically implementing design best practice into your reports.

However, there are a few tips to follow during the build process. For convenience, we've laid out a simple checklist to follow when building your report.

‍

Tip #1 - Ensure you're visualizing 6-10 insights per page

It's important not to go overboard with the insights, as the best way to help users understand what the data is trying to tell them, is to embrace simplicity through design. Prioritize 6-10 insights you wish to visualize per report page.

‍

Tip #2 - Match the right insights with the correct visual

Different visuals suit different insights. Ensure you are matching the right insight with the correct visual using the choose your visuals section of this guide.

‍

Tip #3 - Use a logical layout and group related metrics

Follow the structure provided in the choose your layout section of this guide to ensure your users can easily navigate through your report in a logical format. It's also important to group related metrics together, for example, grouping KPIs or Donut visuals together.

‍

Tip #4 - Ensure the correct sizing of visuals

Use the image below to support with the sizing of visuals, as different visuals represent different levels of detail, and so their size needs to reflect this.

‍

Tip #5 - Apply suitable margin & padding

When working with boxes, be generous with the margin outside of the boxes between visuals, and then padding within each visuals box. This allows the content in the box more room to breathe and becomes more legible for the end-user.

‍

Tip #6 - Apply a visual hierarchy

Visual hierarchy is about making elements of different importance stand out accordingly on the dashboard. In this example, the numeric value is more important than the textual description, so you can increase the font size and the font-weight to make it stand out. The subheading can be de-emphasized through smaller font size, lighter color, and wider letter spacing.

Tip #7 - Add clear headings and labels

Using clear headings and labels helps to add clarity and context to the information you've provided. This principle will also give your audience the ability to extract valuable insights at a glance, eliminating any confusion.

For additional tips around visuals, see our complete guide to Power BI visuals.

Tip #8 - Use a Color Palette

Build your own color palette to stay consistent when using colors across your report. Using a color palette helps you understand which color to use for certain aspects of a report. As seen below, you can also leverage the Power BI customize theme framework to help you build your theme/color palette.

‍

Tip #9 - Use a consistent theme

Ensure you stay consistent with your theme's color palette across your visuals, and limit the number of colors per visual, maximum 6. For example, in the example above we've remained consistent with the use of color across our visuals, but also limited the amount of color used as to not overwhelm users.

‍

Tip #10 - Use distinguishable colors with enough contrast

Use colors that are easy to distinguish between one another, e.g. using contrasting colors compared to shades of the same color, as this can make it hard to tell the difference.