Tableau Software: How to harness the power of data through data storytelling

Everyone working in development knows the importance of using reliable data. Organizations use it to measure their impact, raise funds, and chart their own progress.  

There has never been so many readily available data sources to explore, and the volume of data will only continue to grow. By 2020, there will be 50 times more data as there was in 2011, and 75 times more information sources, according to an International Data Corporation study.

Working with data, however, can be overwhelming. That’s why Seattle-based Tableau Software developed a visual analytics tool to help people see and understand their data. With Tableau, companies can tell stories with their data and share their insights.

Devex spoke with Ben Jones, director of product marketing at Tableau Software, about the evolution of data storytelling, and why development professionals should embrace the practice.

Here are some highlights from that conversation:

What exactly is data storytelling, and why is it important?

It’s a hot topic now, but people have been telling stories with data for centuries. As early as the 1700s, William Playfair, a Scottish inventor and business owner, used statistical charts and graphs to analyze the balance of trade data between England and the surrounding countries.

Florence Nightingale, the founder of modern-day nursing, used statistical charts and graphs to convince Queen Victoria to institute sweeping sanitation reform in the field hospitals that the British Army use during Crimean War. The changes dramatically reduce the number of deaths by communicable diseases.

What’s happening now, though, is that so many tools and so much more data have become readily available. While Nightingale had to collect and record the mortality figures by hand, we can now get reliable data from a number of sources. Then we can use tools such as Tableau or Plotly, another free tool web-based tool, to analyze the data.

There is also growing data literacy, particularly among professionals, and a whole variety of disciplines are learning to use and work with data on a more regular basis.

How is data storytelling different from data visualization?

Sometimes when people refer to “data storytelling,” they’re simply referring to the findings and insights they have gleaned from data.

Take, for example, the famous hockey-stick chart showing global temperatures increasing. That was a single view, a single chart. There’s an explanation mark at the end of it, and people are impacted when they see it.

By comparison, “data storytelling” can be understood as a narrative. By this interpretation, data visualizations are to data storytelling as ingredients are to a meal. Data visualizations are the individual elements, and data storytelling is the combination of multiple such elements into a cohesive narrative. That’s a much more specific way to think of this communication technique.  

Why is data storytelling an important tool for global development organizations?

I think it’s helpful for two reasons in global development. First, organizations can use the tool to explain the problems they are trying to solve, raise awareness for their causes and build a base of support.

Second, it can be useful in the communities themselves. Data has the potential to be a universal language because it’s made up of images and numbers. So organizations can use it to reach the very communities they are trying to help.

How is it currently being used by the global development community to drive change?

There are many ways to use it. Data storytelling can help us understand the core issues that global-development practitioners are trying to solve. It can also help communicate those issues to others who can help, and to educate the communities being served.

Data storytelling is a very impactful way to communicate. Reports can be overwhelming in the sense that they can dump a lot of data on people. Videos can be impactful, but oftentimes they only tell anecdotal stories. Data stories can be helpful because they can walk people through the significant findings step by step.

Data storytelling is also useful for a variety of actors. At the leadership level, for example, you have people trying to shape policy and raise funds. They can use data storytelling to accomplish both of those things — to help people understand the nature of the problem and to show what can fix it in an efficient and effective way.

Data storytellers can walk people through a narrative that builds to a powerful and impactful message. There is inherent credibility to what they are showing because it relies on the authority of the data.

On the second level, you have people who are on the ground in communities, working to make real changes on a daily basis. Data storytelling can help them effect the changes they seek.

What are some of the biggest challenges in data storytelling?

I think there is always a question about the reliability of the data. We’ve seen improvements, but there is still a long way to go. This is especially true in some developing areas where ease of recording data is still not where it needs to be.

We saw an example of this during the Ebola outbreak in West Africa. Even sophisticated and capable organizations such as the World Health Organization had to adjust the mortality figures on a fairly regular basis due to the inaccuracies produced at field hospitals.

There are also challenges in data literacy. Do people understand charts and graphs? The answer is: not as much as we would like. We should invest in helping people become more data-literate so that we can all better understand our world and ourselves.

A third challenge is that we are very biased and we tend to jump to conclusions. There are some very typical problems that we run into such as assuming causation when all we can really speak to is correlation. But those are problems that are built into this process of data storytelling: We are imperfect and so is the data. So we are looking at the world through a lens and communicating it imperfectly to others, who may then also understand it imperfectly.  

How can these types of challenges be overcome in practice?

With regard to the cleanliness of data, software such as Tableau lets you easily explore the contours of your data. And because you can look at your data from a variety of angles, you can quickly and easily identify issues in the data. The outliers pop out, and you can investigate potential problem areas in the data.

For example, if you look at data of cars towed in Baltimore, you’ll see lots of “Foeds” — a misspelling of the word “Ford.” But you likely wouldn’t know that the word Ford is spelled many different ways in that data set because it’s typed in. First, the person doing the towing fills it out on a piece of paper, then another person transcribes it. Unless you are able to visualize it, it’s difficult to know the problems with the data set and how to clean it.

As for how to improve data literacy, I think we can do that by presenting the data, then explaining what we are showing and how they can interpret it. It’s not about issuing a report or publishing a chart or graph on the web and walking away. It’s about being there and having a dialogue.

Annotating charts and graphs can also help. That’s what Playfair had to do when he created the very first timelines, explaining to his readers what he meant to place time on the horizontal axis from left to right in the way that we now take for granted. And today, creative data-design studio Accurat does essentially the same thing, adding a section called “How to Read the Visualization” to its phenomenal work such as this piece called “The Life Cycle of Ideas.”  

Follow Focus On: Devex World for more conversations emerging from the global development event of the year.