So often data is overwhelming. It comes to executives in an incomprehensible wall of data, as a spreadsheet, or as a lengthy deck with poorly conceived charts and graphs that are equally confusing. It doesn’t have to be that way.
Great data storytelling shows a meaningful and clear relationship with the reality it represents. People should be able to quickly and clearly understand how that data relates to a goal or a strategy, so they can actually take action to reach that goal or implement that strategy.
Over the last two decades, XPLANE has developed a framework that allows us to turn complex data into stories that are at the right level and with the right narrative arc for the audience to clearly understand the story.


Framework for establishing the audience’s needs

Great data stories are more than using a lot of data and a dozen ways to slice and dice it. In fact, telling data stories requires restraint and alignment on selecting and showing only the data that supports the needs of the audience.

The key to crafting the right story doesn’t start with the data; it starts with the users and a people-centered design process. Consider the following questions: Who are the users? How will they use data? What are their goals? What are their priorities? Understanding the user helps us hone in on the KPIs and data points that will support the narrative.

Then, good data storytelling follows a fairly simple narrative arc: setting the stage, establishing the tension or the challenge, and then bringing a resolution. That requires us to take our understanding of the user and focus on three key questions:

  • Where does the audience need to focus? This is the starting point, setting the stage for the audience; it should help them quickly see the landscape or thescene where they are beginning. 

  • What do they need to do? Define the action. What do people need to act on? What are opportunities and challenges for them? What is the pressing problem?

  • How do they do it? This is the diagnosis, the resolution. It looks to the data to find the solution and to resolve the challenge.


Once you answer these questions, you have the elements of your story and a clear sense of the business objectives driving decision-making. The next challenge is to approach the design in a way that clearly represents the narrative driving the data story and allows users to quickly and intuitively make the right decisions.

Designing data stories

We start the process by looking at how to group the data. So often in large organizations, managers develop their own reports using different data sources and methodologies. This often results in a set of reports that can’t roll up into a cross-organizational view that is useful for an executive. 

To develop a common language, we ask our clients to identify what data story would be useful. By framing things this way, it frees up their thinking, getting away from the columns and rows in the spreadsheet, and starts them think about how the data might relate to the broader business objective. We can then begin to select KPIs and groupings of data that will support that objective.

Focus on taxonomy

Once we have a sense for the elements that make up the common language, we focus on taxonomy. That means understanding the relationships and hierarchy of information and putting the data into information categories. For this to be successful, we again have to rely on understanding users.

Different users in different roles will need to see data at different levels of granularity. We need to know how and when different people use it, so we can organize data with fidelity to support their needs. For example, an executive might only need a few key data points to know if the new strategy is working, but a CFO might want to know what aspects of the strategy are going well and what action needs to be taken to affect the outcome.

Visualizing the data story

Finally, we think about how we represent the data on the page. What’s the right chart or graph to express the data? While a scatter plot might be a great way to a see distribution across a few dimensions, it’s not the best way to view trends over time. In addition, cognitive considerations are made around color and space. Which colors will elicit the reactions we are looking to communicate? A lot of warm colors on a page might create alarm even when that’s not the intent; a page with too much information or too many things going on might create confusion about what to focus on. Finally, considerations such as type and labeling are important to establish the proper relationship among elements and provide clarity and context where needed.