DP-900 · 4 Describe an analytics workload
Data visualization types
Exam objective: Identify appropriate visualizations for data
Picking a visualization means matching the chart to the question being asked, such as a trend, a proportion, or a relationship between measures.
Picking a visualization is really about matching the chart to the question you are asking of the data.
| You want to show | Use |
|---|---|
| Many related values at once | A table |
| One important figure | A text card |
| Values across discrete categories | A bar or column chart |
| A trend, often over time | A line chart |
| Categories as a proportion of a total | A pie chart |
| A relationship between two numeric measures | A scatter plot |
| Values across geographic areas | A map |
In Power BI, the visuals in a report are not independent: selecting a value in one, for example a city in a column chart, filters and highlights the matching data in every other linked visual on the same report page.
Power BI also layers AI on top of these basics. Smart narratives write a text summary of what a visual shows and update it as the data changes. The Q&A visual turns a plain English question into a chart answer straight from the underlying semantic model. Key influencers shows which factors most strongly drive a chosen metric, and a decomposition tree lets you drill down across several dimensions at once to see what is contributing to a value.
On the exam, match the comparison being asked about (trend, proportion, correlation, category or location) to the chart built for it.
Key points
- Bar and column charts compare values across discrete categories; line charts show a trend, usually over time.
- Pie charts show categorized values as a proportion of a total; scatter plots show a relationship between two numeric measures.
- Maps compare values across geographic areas.
- Selecting a value in one Power BI visual filters and highlights the matching data in other visuals on the same report.
- AI visuals include smart narratives, the Q&A visual, key influencers and the decomposition tree.
Exam trap
A pie chart and a line chart are easy to mix up when a question mentions only over time versus share of total. A trend belongs to a line chart, a proportion of a whole belongs to a pie chart.
Check yourself
An analyst wants to check whether higher marketing spend is associated with higher revenue across many campaigns. Which visualization should they use?
Go deeper on Microsoft Learn
Checked against Microsoft Learn on October 1, 2026.