Skip to main content

Smart Narrative (preview)

Many times we see users use PowerPoint, text boxes, interactive headers, or custom visuals to add narrative to their data.  All these options either lack interactive data, interactive and customizable text, or useful automatic insights. We are very excited to announce the preview of Smart narrative visual which can help you with this.

The Smart narrative visualization helps you quickly summarize visuals and reports by providing relevant out-of-the-box insights that are customizable.

Using this feature creators can add narratives to their report to address key takeaways, point out trends, edit the language, and format it to fit a specific audience. Instead of pasting a screenshot of their report in PowerPoint with key takeaways added, they can now add narratives to the report that are updated with every refresh. Your end-users can use the narratives to understand their data, get to the key points faster, and explain the data to others.

Since this feature is in preview, you will need to first turn on the feature switch by going to File > Options and Settings > Options > Preview feature and make sure Smart Narrative visual is turned on:

Clicking on the new smart narratives icon in the visualization pane creates a narrative based on all the visuals on the page.

For example,  in this report about online sales, you can click on the icon to automatically generate a summary of these visuals talking about revenue, website visits, and sales. Notice that Power BI automatically does trend analysis to show that Revenue and Visits have both grown and it even calculates what kind of growth it sees – in this case, a 72% increase.

You can also right-click the visual and select ‘summarize’. This will generate an automatic summary of that visualization. For example, when you right-click -> summarize on the scatter chart showing the various transactions, Power BI analyses the data and shows which city/region has the highest revenue per transaction and the highest number of transactions. It also shows the expected range of values for these metrics, so you can understand most cities were below $45 in revenue per transaction and had less than 10 transactions.

The summary is highly customizable where you can add new text or edit the existing text using the same controls available in the regular text box. For example, you can make the text bold or change the text color.

You can also customize the summary and add your own insights by adding dynamic values. You can map text to existing fields and measures, or use natural language to define a new measure to map to text. For example, if you want to add information about the number of returned items, you can use the add value experience as shown in the gif. We have integrated the Q&A experience to add dynamic values. As you type, you’ll get suggestions in a drop-down just as in a Q&A visual and you can just save this as a value.  So, in addition to being able to ask questions of your data in Q&A, the scope has been expanded to create your own calculations without even doing DAX.

You can format the dynamic values, for example, to show as currency, specify decimal places, thousand separator, etc. You can do this by clicking directly on the value in the summary to format it or clicking on the edit button corresponding to the value in the review tab of the text box control.

The summary is dynamic and automatically updates the generated text and dynamic values when you cross filter. For example, if you select Electronics in the donut chart, the rest of the report is going to cross filter and the summary will also cross filter to focus on the Electronics products too.  In this case, the visits and revenues have different trends, so the text gets updated to reflect that. And the count of returns value you added also gets updated to $4196.

You can also do more advanced filtering. For example in this visual that looks at trends of multiple different products, if you are only interested in the trend across a certain quarter, you can select the relevant data points to get the summary updated for that fragment.

Let us know what you think!

Make sure to try out this visual while it is in preview. We would love to hear what you liked about the feature and how we can improve it!  If you have any feedback for the team, please comment in our community post.