Power BI basics for programme dashboards

What is Power BI, and when does an NGO need it?

Power BI is a Microsoft tool for turning data into interactive visuals. A project manager can open a dashboard and see progress against targets, filter by township and spot problems, without scrolling through spreadsheets.

Microsoft describes Power BI Desktop as a free Windows application that lets you connect to data, transform it and create interactive visual reports, and notes that the Power BI service lets you publish and share them. For many NGOs, the main use is to put MEAL and programme data from Excel or KoboToolbox in one view.

But a dashboard is not always the answer. If one person reads one table a month, an Excel chart is enough. Power BI makes sense when many people need the same up-to-date view, when data comes from several files or when manual reporting takes too long. See our guide to dashboards for MEAL for how to decide.

Reports and dashboards: what is the difference?

Microsoft's terms are useful to know.

Microsoft notes that unlike reports, dashboards do not support filtering or slicing. This is why you build the full analysis in a report and pin only the key visuals to the dashboard. It also says that, because a dashboard is limited to one page, a well-designed one contains only the most important elements of the story.

Step 1: Prepare your data

Power BI cannot fix messy data. Start with a clean table in Excel: one header row, one record per row, no merged cells and consistent spelling. Use drop-down lists when you collect data, and follow the checks in our guide on data quality checks. If you can build a pivot table from the data, Power BI can use it. Our guide on pivot tables for monitoring data shows how to test this.

Remove names and contact details that the dashboard does not need. Read data protection for NGOs before you publish anything that holds personal data.

Step 2: Build your first report

Microsoft's getting-started guide follows these steps. Use them as a plan.

1. Install Power BI Desktop, or sign in to the Power BI service in a browser.
2. Connect to your Excel file. In the service, you can add the file as a semantic model.
3. Select fields in the Fields or Data pane. Power BI creates a visual from your choice.
4. Change the visual type in the Visualizations pane: bar chart, line chart, map or table.
5. Add filters, such as year or township, so users can explore the data.
6. Save the report and give it a clear name.

Choose visuals by question. Use a line chart for change over time and a bar chart for comparing places or groups. Use a card, which is one big number, for a headline figure.

Step 3: Pin key visuals to a dashboard

Open the report and, on each visual, choose Pin visual. Pick a new dashboard or an existing one. Microsoft says that after you pin, the tile stays up to date with the data, but if you later change the visual type in the report, the pinned tile does not change, so you pin it again.

A good programme dashboard has five to eight tiles. For example: people reached against target, activities completed by month, budget spent against plan, number of complaints received and closed, and a map or table by township. Put the most important figure at the top left.

Step 4: Plan for refresh and sharing

A dashboard is only useful if the data is fresh. Microsoft's learning path includes data refresh and scheduled refresh as steps after publishing. Decide who updates the source file, how often, and who checks it. A weekly update by a named person works better than a vague plan.

Also check sharing rules before you promise anything. Microsoft states that to view dashboards shared by colleagues, a person needs a Power BI Pro or Premium Per User licence, or the content must be shared from a Premium capacity. Licence terms change, so check current details and costs. Where internet access is limited, consider exporting a PDF snapshot for partners.

Common mistakes

• Too many visuals. A crowded dashboard hides the message.
• No clear audience. Decide who will use it and what decision it supports.
• Messy source data. Fix it first.
• No owner. Name someone to maintain it.
• Exposing sensitive data. Share only what each viewer should see.

Start small

Pick five indicators, one audience and one data source. Build a first version in a week and ask users what they would change. Improve it for a month, then add more. A small dashboard that people use beats a large one they ignore.

How EMPO can help

EMPO is a reliable, Myanmar-based training and consultancy provider and one of the country's leading names in practical NGO training, with 1,200+ people trained and 15 partner organisations. We offer MEAL and data training online or in person across Myanmar, in English and Burmese. See our training services or contact us to plan a session on your own data.

Sources

• Dashboards for Business Users of the Power BI Service — Microsoft Learn
• Tutorial: Get Started Creating in the Power BI Service — Microsoft Learn