Blending skills with backgrounds
Ryan Swann, Director of Data Analytics at OGP, believes that different backgrounds are the key to a data science team’s success.
1. What is your current job title and what are your main responsibilities?
My current job title is Director of Data Analytics for the Office of Government-wide Policy, OGP for short. We are responsible for using data and evidence to accomplish three key areas: evaluate policy, define cost savings, and efficiencies. By looking across the government with regard to different metrics and measures, we’re able to identify agencies that appear to be doing better than other agencies and then learn from them to establish the best practices and share information with other agencies that may be able to benefit from that information.
2. How do you use data science in your job? How has data transformed your agency?
We do what I call the three layers of moving from reporting to actual analytics. For example, the first aspect is looking at how many widgets we have, or this is how many buildings we have, or this is the square footage, this is how much it cost us. The next piece is metrics and measures, where we determine whether those numbers are high or low: Is that what we’re paying, what does that mean? Should I be worried or should I be happy? The third piece, which is the analytics piece, is how can I impact it? So we do everything from ETL (Extract, Transform, Load) to database work, to SQL coding in R and Python, to predictive analytics where we’re building models. Typically, we’re trying to answer a particular question. For example, what do we expect our costs to be in 5 years? So those types of business questions allow us to determine what type of analysis to do.
3. How many other data scientists do you work with?
My staff is 9 data scientists, and we typically have about anywhere from 11 to 15 contractors that are also data scientists. We’re in the process of redoing our contracts with those people. In a typical day, since I was a data scientist before I became a director, I’m very hands-on with the team whenever they have an issue solving a problem.
“If you are looking to come to a place and do one thing, this isn’t the place for you”
4. How have you seen data science improve outcomes in your department or team?
We have some data scientists who studied high-level math, we have some statisticians, we have some economists, we have an English major, and we have a couple of business majors. When I set the team up from scratch, one of the things I said was that if you are looking to come to a place and do one thing, this isn’t the place for you because I have no idea what technology we’re going to get into. The team is constantly training, constantly learning new things, and that’s been my approach. Where I’ve seen data scientists flourish is where we have acquired ambassadors for “data as an asset,” even if those ambassadors are not themselves data scientists. This highlights that when you unlock the value, a lot of people benefit. In data, people need to understand that even though you may not be a data scientist, if you’re the data entry person, for example, you are just as important as the data scientist. So it’s creating this culture of shared responsibility from the entry level person all the way to the executive level.
5. What was your first job out of college?
I was a computer engineering major at the University of Maryland and I wanted to switch to business but I had too many credits and ended up graduating with a degree in communications with a minor in computer science. And right out of college, I got a job at the Department of Treasury as a “performance measure of metrics specialist.” We pulled data from a call vault system. I learned a majority of my data skills outside of school. I did learn object orientated computer languages in school — for example, C, C++, C sharp, but I learned SQL, R, outside of school. A majority of the people working with me at the Treasury came from a variety of backgrounds — such as Physics, Psychology, or Criminal Justice. I think anybody can learn data science if they want to and it’s not something that has to be studied in school. I’m a living testament to that and a lot of people around me are as well.
“It’s bigger than just ones and zeros, bigger than just terabytes of data”
6. What were some key moments/jobs that lead you to your current role?
At the Department of Treasury, I was working for the assistant secretary for the CFO, and we were trying to solve problems. This was my first experience into the strategy side of data. It was getting the pieces, getting the people to agree, the stakeholders to agree, who all have different agendas, but you have to get them to agree so you can get the job done. It was that experience that really made me see that it’s bigger than just ones and zeros, bigger than just terabytes of data, that you had to be able to connect what you were doing to the business side or to the value proposition. This is similar to a startup. You could have an awesome idea or an awesome product but if you can’t convey the value of the product or solve somebody’s problem, your product will never go anywhere. You have to connect it and that was the first time that happened for me.
7. What are 3 traits that you would consider to be the most important traits for a data scientist to possess?
1) Communication and negotiation skills — The ability to take something technical like regression or Bayesian analysis and be able to explain it to a business owner, a deputy secretary, a president, in a way that is meaningful to them. That skill is one that you have to develop. After graduating from the University of Maryland, I sat on a couple of University boards and one of things I was a huge advocate for was making a communications class a core requirement for all students at the University because in all majors, communication is fundamental.
2) Technical skills — You need technical skills to be a data scientist. The good thing about it is you can always learn. There is so much information available through the Internet, YouTube, and other organizations that can help people learn this craft. People from all walks of life have found careers in data science who never started out that way. They realized they could turn subjective decision-making into objective decision-making by using data and when that light bulb goes off, people are then able to learn this technical skill set.
3) Flexibility — Being flexible about technology, computer languages, and methodology. What we do know from our day and age is that technology is constantly changing. The amount of data that we create and consume is doubling almost every year. So if you learn SQL or if you learn R, what we’re going to be doing 2 or 3 years from now might be totally different and you’ve got to be flexible enough to understand that it’s not about the language but about the value. You’ve got to have the flexibility to stay on the cutting edge of what’s required for your industry for whatever value you’re trying to extract from the data.
“Build a foundation, look for on the job training opportunities, become a part of a team or shadow a data scientist”
8. How would you recommend someone get into the field of data science?
The first step I would say is to assess your current capability. You have to understand the foundation of information about data science, how data is structured and how some data is unstructured. I think there is a foundation that all data scientists have to understand before they can do the more advanced things. My recommendation would be to utilize resources, whether it’s YouTube or online classes or even in-person classes to build that foundation. Take a Data Science 101 course and you’ll be surprised how simple it is once you go through it a couple times. It’s not rocket science, it’s actually pretty simple, but getting over that fear of something you don’t know can be difficult. Just get started, and if you’re in an organization that has a lot of data, volunteer some of your time in the data science field so that you can learn on the job and create value where your boss may not be expecting you to. Build a foundation, look for on-the-job training opportunities, become a part of a team, or shadow a data scientist. We’re always looking for people that can give us a unique perspective.
9. What do you think is the future of open data?
I think that we are moving towards a society where there is going to be so much data that the only way to extract a decent amount of value from it is to crowdsource it. By having open data standards, we’ll be able to unlock the brainpower of institutions, companies, non-profit organizations, and educational institutions, to create value from our data. You can take agriculture, global warming, technology, telecommunications — you name it, there is opportunity there. I think what open data really allows us to do is find a way to share the right data and then ultimately scale it. We already consolidated our data centers and expanded our hard drives. But the amount of data that we create will soon surpass how fast we improve our data storage, especially as the population approaches 9 billion people. You need something like open data in order to solve some of these larger issues around the globe, and I think that is what open data really allows us to do.
“Everybody has that opportunity to make data an asset”
10. Do you have any additional thoughts or comments that you would like to share with our readers?
Even if you’re not a data scientist, even if data is not your thing, you can be an ambassador for data. I think that what it comes down to is creating a culture in your jobs, organizations, and your schools that embraces data. Either from using it and helping to move it forward or actually being the data professional. As a community and as a society, we’re not there yet and I think we have a ways to go, but I think that everybody has that opportunity to make data an asset.
This interview was facilitated by Data Society and conducted as part of the Data in Action series, which aims to highlight the many paths of data science in the government. It was done in partnership with the Department of Commerce and the Data Cabinet chartered through GSA.
