It’s been more than a decade since I took my first statistics course in college. Unlike for many, my introduction to statistics brings back happy memories of an enthusiastic professor who jaunted up and down the stairs of the lecture hall. It’s not easy to get excited about beginning concepts in distributions and hypothesis testing, but he pulled it off. I grew interested in working with and understanding data which eventually led to many years of graduate school. I had no clue back in college that statistics—or more generally, using data—would be so popular now. I just liked to play with data. And there’s a lot of data to play with these days.
Every day I read or hear about companies and organizations that use data in some way. There’s a wide array of applications: improving business, providing better service to customers, helping to make the lives of others easier, or communicating complex processes. There’s an excitement. People want to gain insights from all this data they collected.
There’s a gotcha though, and it’s a big one. You can’t just take a stream of data, plug it into the most expensive software you can find, and gather instant results—regardless of whether you’re one person or a big organization. It’s never that easy. Anyone who tells you otherwise either doesn’t know what he is talking about or is trying to sell you something.
As someone focused on data visualization, I would love to build a dashboard or develop an interactive tool that enables people to understand their data in an instant. No background needed. However, you have to learn how to use the tool before anything worthwhile comes of it. You must know what data represents and how to analyze and interpret.
When you start to look at how an entire organization can grow more fluent in the language of data, you introduce other challenges. Those in management have different responsibilities than those working on the floor, but there must be a proper foundation for everyone to work together in an effective way.
Zach and Chris Gemignani, co-founders of Juice Analytics, help groups with these challenges every day, and now they educate others with Data Fluency. The two brothers and their team have been consulting long before “big data” became a thing, before Google’s chief economist Hal Varian said that the job of a statistician is sexy, and before I started FlowingData. The Gemignanis’ experience shows in their articles online and in this book. Their advice is practical but general enough so that you can apply frameworks to your own situation.
When I first searched for “data visualization” years ago, the Juice Analytics site was one of the first ones I found and still subscribe to today. So I was excited when Zach and Chris agreed to write Data Fluency. However, this isn’t a book about visualization. It certainly covers the topic, but Data Fluency provides a wider view.
When you have visualization floating around in your organization—reports, talking slides, and data displays—does it actually matter if no one looks or gets anything out of it? It ends up in the recycle bin or as background noise. You can have the most efficiently designed charts in the world, but at the end of the day, you need people to pay attention. The goal is to bring data closer to the front so that everyone from management on down can make better informed decisions.
At the same time, there is no promise of a panacea or a new tool to make all data problems go away. It’s a realistic view that stems from the Gemignanis’ experience. They understand that often a lot of moving parts in groups might move slower than others or are difficult to change. I’m just a one-man show with FlowingData, but in my own consulting work, I understand the pains of bureaucracy all too well. The key is to work with the areas that do change and go from there. Data Fluency is an excellent guide to figuring out how you can do this.
Sitting here, thinking about what data might look like another decade from now, I can only imagine more of it, at a more detailed level. In the present day, the rate of collection far exceeds the rate at which we can understand. However, the growing rate at which people want to understand is a different story. So the more people who can learn the language of data now, the better we will be for it later.