Readers:Today, I am featuring an interview with Filippo Mastroianni.
Filippo was born in Milan, Italy. He is a Data Analyst and Consultant for The Information Lab Italy. Filippo helps organizations make data-driven decisions, helping clients make sense of their data and embracing a new analytical culture. Filippo graduated with a thesis on Data Journalism and discovered his great passion for Data Visualization, which he approaches from a humanistic perspective. Since 2016, he regularly collaborates with Il Sole 24 Ore.
Filippo has authored several Viz of the Day Tableau workbooks, and currently is a Tableau Ambassador and Former Tableau Featured Author.
I find Filippo’s work to be very exciting and informative. It really shows his great passion for data visualization. I hope you enjoy this interview and his work as much as I did discussing it with him.
An Interview With Filippo Mastroianni
Michael: Hi Filippo, it is very nice to finally meet you. If you don’t mind I am going to dive right into the questions.
Michael: Back in 2016 (and overlapping into 2017), you published a data visualization that chronicled the last 50 years of the government of Italy. On Tableau Public, your dataviz was Viz of the Day as well as it became a Greatest Hit.
The first question I have about your dataviz is what is your process behind creating a data visualization like this? How do you determine which data to use and where do you find this data?
Filippo: This is one of my favorite vizzes because it is the first one that I published on Il Sole 24 Ore, the most important Italian daily business newspaper. On the InfoData Blog we are really free to choose our topics, but in this case, the visualization was influenced by actual events. The government Renzi ended on December 7 after the failure of the constitutional referendum. My intent was to analyze the last 50 years of the Italian Republic. We’ve seen alternating at Palazzo Chigi 21, different Prime Ministers and 43 governments, one every 1.16 years. This is way too many. I was inspired by Güney Soykan’s project Face of a Nation, and I thought to make it interactive with Tableau. I started with his creation to make the background image (even if I needed to change it a bit because he missed a few of the Prime Ministers).
This is the first tip I offer your readers: look around and find inspiration from other Tableau users or dataviz projects.
It is always difficult for me to explain how I build my viz, which design choice I take, but I can try. There isn’t a single way to start a dataviz project. Sometimes I start with an idea, I know what I want to show. So I look for data and draw a basic sketch of how I want to visualize my information. Other times I begin my project from a dataset and I use Tableau to make what is called data-discoveryand find something interesting that I didn’t notice.
There are a lot of open data portals and resources where you can find data. Working for Il Sole 24 Ore I often look at Istat or Eurostat data.
Michael: I notice you seem to follow a particular style with your Tableau data visualizations [NOTE: A screenshot snippet of some of the Tableau workbooks Filippo has created are shown at the end of this blog post]. Does it help you to follow a particular style (e.g., color, fonts, layout, etc.)? When we look at paintings by famous artists, you can see a distinctive style like Monet or Picasso.
Filippo: As I already said we are very lucky to have a huge data visualization catalog such as Tableau Public. You can look around and find inspiration from other Tableau users to built your own personal style. This is my first piece of advice. This allows people to immediately recognize your work. I started to always use the same background because it was what was needed for my newspaper, but now I feel that it represents my style. Essentially, find your own personal way to visualize data.
I always use the same background and limit myself to 1 or 2 fonts per workbook, trying to use the least amount of colors as possible. Tableau helps us a lot with the color palette they provide. The color palette, “Tableau 10”, is more than enough in most cases. Obviously be careful that colors are consistent.
If you create something for a media organization the most important thing is to grab the reader’s attention. I try to be more creative and “think outside of the box”. When I create a dataviz for Il Sole 24 Ore my purpose is to not only to give the information. I want to try to have people play with my data. Creating engagement with my vizzes is the key in the world of communication. Data visualization is science but also art, and for us Italians, the second one is something natural. As you can see in the work of some famous data visualization specialist, such as Giorgia Lupi. Her work is also being bought by MOMA (New York’s Museum of Modern Art), to be part of their permanent art exposition.
Anyway, most of all it is essential that design is at the service of data and not the contrary.
Michael: How did you get started in the area of data visualization?
Filippo: I was still attending the university when I started looking at this area of specialty. I was studying Science in Communication and I was looking for a new journalistic approach. At that time I read a lot of books about investigative journalism and data journalism, For example, Simon RogersFacts are Sacred, or Alberto Cairo’sThe Truthful Art. My professor advised me to try Tableau to create a little project for my thesis. And that’s how I started. Then it was all about passion for this tool and for data visualization and journalism. I think today data journalism is the most important weapon that investigative journalism can handle to find and share the truth with the people. Data journalism became the main topic of my thesis, titled “Data Journalism: the Precision Journalism in the Age of Big Data”.
I started from a humanistic background and my first job in data visualization was for a newspaper. This means that the world of data is opened to everyone, including people without a strong technical background. And I find this amazing. Data is for everyone. Obviously, I’m also studying and building stronger technical skills, because my new job as a consultant needs it.
Michael: You are also an Alteryx Consultant. I personally struggle with trying to sell the notion of our business partners doing their own data prep versus an enterprise level data warehouse where data has been normalized, enhanced, and in terms of things like addresses, validating against a stand like CASS (USPS Postal Standards) to ensure consistency and accuracy.
Do you find having business partners do their own self-service data prep a hard proposition to sell?
Filippo: I don’t think that the idea of self-service data prep is hard to sell. I think that Alteryx can really help organizations and analyst to easily prep, blend, and analyze data using a repeatable workflow, then deploy and share analytics at scale for deeper insights in less time. The idea is the same for Tableau: work on data without a strong technical background, without knowing how to code and in a faster manner. It’s true that maybe there isn’t the same “wow effect” that Tableau has when you try to sell it.
Michael: Tableau is currently beta testing its own data prep product called Maestro. What are your thoughts on this and how do you think it will affect the use of Alteryx in companies?
Filippo: After looking a bit at the alpha and beta versions, I can say that Maestro is a simple data prep workflow tool. It will fill a vacancy that Tableau has in regards to ETL. You will able to do Joins, Unions, pronunciations grouping, cleaning up spelling, prefiltering and more. It’ll be easy to learn and can do any data preparation that can be done today in Excel. Compared to other data preparation software, Maestro seems to be really intuitive.
Alteryx is something different. It is an Analytics tool and can do all that you will be able to do with Maestro in an excellent manner. But it also can do a lot more. I use Alteryx for scraping data for example. And this tool also has key capabilities for everything related to spatial and predictive analysis. Something more than just simple data prep. Maestro will help Tableau users to make it easier to do their data prep, but Alteryx opens new possibilities and permits people to do things that could never be done without coding knowledge, such as Python or R.
They are simply different tools and cover different tasks.
Michael: Do you feel Maestro should be integrated into Tableau Desktop or be a stand-alone tool?
Filippo: It isn’t clear at the moment. Maestro is still in beta and we have no idea what the price will be. There isn’t an official position at the moment so I can’t make a prediction. I can say that I feel that Maestro will be a separate tool to install, but I don’t know if it will come with Tableau Desktop with an extra charge or for free.
Michael: How would you define data-driven journalism?
Filippo: First of all I want to say that data journalism is, for all intents and purposes, journalism and not something separate from it. We have been analyzing data since journalism was born. Maybe it is better to say that the skills required are changing and “write well” is no longer enough.
Data journalism means the use of data and numbers in journalism to uncover, better explain and provide context to a news story. It often involves the use of statistics, charts, graphs or infographics. Data journalism isn’t something totally new, it has been around as long as there’s been data. But today we have spreadsheets and files formatted for computers and we can use open data and new technology to find stories, insights in data, communicate our intuitions.
This is the age of data.
This is the age of data. We have a lot of information and a lot of stories out there, hiding in plain sight. We (as journalists) only need new technical skills to find these stories. Looking around I can see that media organizations increasingly use data analysis and data visualization for their analysis. At the moment I also see designers, data journalist, statisticians, “nerds” a little bit out from the world of the mainstream media, especially in Italy. But I think that this condition will change. At the Wall Street Journal and other big media organizations, designers, programmer and journalists work together as a team. Also, the newsrooms will change, and the USA usually lead the way to new methods.
Michael: How do you feel we as authors of data visualizations can best assure someone who consumes our dataviz that the data is accurate and unbiased?
Filippo: Obviously we need to be careful with our sources. But this is not enough. Nowadays it is absolutely important to know how to use the tools and how to process data, to be sure we aren’t missing something.
Third, but not less important, I noticed that there are higher requirements, employers are always looking for a lot of technical and analytics skills. And that’s fine. But it is particularly important that we do not neglect that there is much more. It happens that many people assumed those people with advanced economic, mathematical or statistical skills have difficulty communicating their intuitions to others in an effective way and to use the right graphics.
A knowledge of data visualization best practices is as important as it is to know how to use the tools. Every image, every visual representation, brings a message, sometimes a story. We can create accurate dataviz from the data point of view, but we need more to communicate what the data is telling us. Especially regarding data journalism, where we need to talk to unskilled people, to as many people as possible. I absolutely think that the design of a dashboard can make a difference, even if we need to be sure that our dashboard never leads to incorrect interpretations of the data.
Michael: Thank you, Filippo. I really appreciate your insights and look forward to seeing you again at the Tableau Customer Conference 2018 in New Orleans.