Your first Power BI dashboard: from spreadsheet to clear report

By Axyra Nova Editorial · 12 October 2026 · 5 min read · Data and AI

Power BI is Microsoft's tool for turning data into interactive reports and dashboards. It connects to spreadsheets, databases and many online services, lets you clean and combine data, and produces visuals people can explore. For analysts and small business owners alike, it is a powerful step up from static spreadsheet charts.

This guide walks through building your first Power BI dashboard, focusing on clarity rather than complexity.

Start with questions, not charts

The most common dashboard mistake is starting with visuals. Instead, write down the questions the dashboard should answer:

• How are sales trending month by month?
• Which products and regions perform best?
• How many orders are delayed?
• Are we on track against this quarter's target?

Each question becomes a visual or a number on the page. If a visual does not answer a question someone actually asks, leave it out.

Getting Power BI

Power BI Desktop is the free application for building reports on Windows. Publishing and sharing reports online uses the Power BI service, which may require a licence depending on how you share. Microsoft's Power BI fundamentals documentation explains the components and how they fit together.

Step 1: Load your data

In Power BI Desktop, choose Get data and select your source: an Excel workbook, a CSV file, a database or an online service. Start with something simple, such as a sales spreadsheet with dates, products, regions and amounts.

Step 2: Clean data with Power Query

Before loading, Power BI opens Power Query, where you can clean and shape data:

• Remove unnecessary columns.
• Set correct data types for dates, numbers and text.
• Fix inconsistent values, such as trimming spaces and correcting spelling.
• Remove blank or duplicate rows.
• Split or merge columns where needed.

Each step is recorded and repeated automatically when the data refreshes, so you clean once and benefit every time.

Step 3: Build a simple data model

If you have more than one table, such as sales and products, connect them with relationships. A common pattern is a star schema:

• A fact table with transactions, such as sales.
• Dimension tables describing things, such as products, customers and dates.

Relationships link them through shared columns, such as product ID. A dedicated date table makes time-based analysis much easier.

Step 4: Create measures

Measures are calculations written in a formula language called DAX. Start with simple ones:

• Total sales: the sum of the amount column.
• Order count: the number of orders.
• Average order value: total sales divided by order count.

Measures calculate correctly whatever filters are applied, which is what makes dashboards interactive.

Step 5: Choose clear visuals

Match each question to a suitable visual:

Keep it simple:

• Prefer bar and line charts; they are easy to read.
• Avoid 3D effects and decorative visuals.
• Use colour sparingly and consistently, highlighting what matters.
• Label axes and titles in plain language.

Step 6: Lay out the page

A clear layout guides the reader:

• Put the most important numbers at the top.
• Group related visuals together.
• Leave white space.
• Add slicers for common filters, such as date range and region.
• Make sure text is readable at normal zoom.

Good layout follows the same principles as good interface design. The Nielsen Norman Group's usability heuristics, such as consistency and minimalist design, apply to dashboards too.

Step 7: Test with real users

Before sharing widely, show the dashboard to one or two people who will use it. Ask them to answer their usual questions using it. Watch where they hesitate. Adjust labels, order and filters based on what you learn.

Step 8: Publish and refresh

When ready, publish the report to the Power BI service and set up a refresh schedule so data stays current. Decide who should have access and share accordingly, keeping sensitive data restricted.

A worked example

A small chain of pharmacies tracked sales in a shared spreadsheet that was updated weekly. Managers spent hours each month building charts for a review meeting.

An analyst built a Power BI report connected to the spreadsheet. In Power Query, she standardised branch names and product categories that had been typed inconsistently. She created a date table and measures for total sales, transactions and average basket size. The single report page showed:

• Three cards at the top: sales this month, change from last month and average basket size.
• A line chart of weekly sales by branch.
• A bar chart of top product categories.
• Slicers for branch and date range.

Managers now opened the report before meetings instead of building charts, and discussions focused on what the numbers meant rather than on producing them.

Six months later, the analyst added a second page for stock levels, using the same layout and colours so managers could find their way around immediately.

Common beginner mistakes

• Too many visuals on one page.
• Skipping data cleaning, leading to split categories.
• No date table, making monthly comparisons awkward.
• Using calculated columns where measures would be better.
• Forgetting who the audience is.

Learning more

Microsoft offers free Power BI learning paths, and our learning platform includes the Power BI course in its data track. Strong spreadsheet and SQL skills make Power BI much easier; see our guides to Excel pivot tables and SQL for data analysis. Browse the full learning platform for more.

Frequently asked questions

Is Power BI free?

Power BI Desktop is free to download and use for building reports. Sharing reports through the online service may require a licence.

Do I need to learn DAX?

Basic measures are enough to start. Learning more DAX helps with advanced calculations later.

Can Power BI connect to Google Sheets?

Power BI can connect to many sources, including web-based files, though some connections need extra setup. Check Microsoft's documentation for current options.

How many visuals should a dashboard have?

As few as needed to answer the key questions. Many good dashboards have between four and eight visuals on a page.