Editing the Narrative

One of the biggest challenges in the data world is communicating the result of our data models to those in the business world who reside outside the world of technology. While we certainly understand the ins and outs of our own work, the biggest obstacle is making sure others find our work tangible and more importantly valuable to their own perspective within the business or organization.

The business-facing front of this conversation between the technical side and the business side is what we can call the narrative. They don't have to represent two people or business groups either. They can even represent a single person with the skills and ability to bridge the gap between the two sides. Here's how Merriam-Webster defines the term narrative.

narrative: noun, nar· ra· tive

1. Something that is narrated: story, account
2. A way of presenting or understanding a situation or series of events that reflects and promotes a particular point of view or set of values.

However, I don't feel it's worth getting too hung up on the definition of this term. The bigger questions we want to answer are:

• Why is it important to communicate the narrative of data models?
• How can we do this in an effective way?

Smart Narrative Visual in Power BI

Within Power BI, the latest versions of the desktop application offer a standard smart narrative visual that we can use to communicate the outcomes and key points of our Power BI data model from the Power BI Dataset. Here's the latest video in my Power BI Weekly series that walks through how to use the smart narrative visual in an impactful way.

On a high level there are two parts to using the smart narrative visual:

Part One: Adding the Visual

When we add the visual to the page in Power BI, it automatically creates a written text narrative describing key trends, outliers, and anomalies in the data model represented by the Power BI Dataset. It's an example of built-in AI within Power BI (different from building the models ourselves), which runs automatically behind the scenes for us to create a narrative without any of our input, apart from creating the initial Power BI Dataset beforehand.

Part Two: Editing the Visual

Once we create our smart narrative visual, we can then edit it (hence the title of this week's newsletter edition). This is the key part of using this visual; not only can Power BI provide insights into our data, but we can also make it our own. This means we can:

• Delete the text and any related fields we don't want to include in our narrative.
• Edit any of the text and related fields.
• Add any additional text and related fields to help enhance the existing insights the smart narrative visual already provides.

Any of the edits above allow us to communicate the narrative within Power BI on a high level with aggregated data summaries, as well as on a detailed level with individual data points.

Coming Up!

It's the first week after Thanksgiving in the US, so this newsletter is a short one because I took a break over the last week. I also scheduled my posts here on LinkedIn before I took this breather as well!

I'm excited to continue to share the latest videos in my Power BI Weekly serial course. I'm also excited to begin filming (this week!) the time series models course for the LinkedIn Learning library that uses Excel, R, and Power BI to play around with, build, and share the outcomes of time series models like ARIMA. Stay tuned for more details!

-HW