Fractional Focus: Unveiling the role of Data Science leadership with Scott Strong
Learn how a data science executive thinks about building scalable data systems at startups.
January 17th, 2024
Introduction
In the intricate and evolving landscape of data science, understanding the nuances of leadership in this field is crucial. Sitting down with Scott Strong offers an exclusive look into the mind of a seasoned data science executive, former founder, and consultant to startups. Scott Strong's journey in machine learning and data science across various industries exemplifies the fusion of technical acumen and strategic thinking.
Below we explore Scott's unique approach to building scalable data systems, the lessons from his successes and failures, and his perspective on driving business forward with data science.
Fractional Focus with Scott Strong
What is your specialty and how did you gain mastery over it?
My specialty is in machine learning. Specifically, understanding how to apply machine learning and data science to a diverse set of industries. I initially gained exposure to machine learning through my education in aerospace engineering where I learned to leverage machine learning to aid in the design of aircraft and spacecraft. While in business school I recognized the potential machine learning had outside aerospace, which compelled me to explore how these skills could be used in other applications. Since then I've worked in 6+ different industries and have mastered the ability to speak with decision makers to understand their unique problems, products, and companies and converge on data and machine learning solutions that drive their business forward.
What was the most successful thing you built or have worked on?
During my tenure as the Head of Research at CloudZero, I steered the company in a new direction over the first year through extensive discussions, rigorous research, and prototyping. This shift has been sustained, and the products I contributed to have flourished. The company has since doubled its commitment to this direction, securing millions in financing and achieving record-breaking revenue.
This shift in direction was not easy, especially because I was not a founder at CloudZero. The original vision for the company was a chaos engineering company, helping their customers detect issues in their cloud deployments. After struggling to get the data and modeling necessary to achieve this vision, I felt there was something that we were missing, and so I went looking.
This missing piece was a treasure trove of data hiding right in front of us. As most of our clients at the time used Amazon Web Services, I noticed that AWS was particularly ruthless with the data that they tracked and sent out to each customer when it came time to pay the monthly bill. This clue led me to discover an entire ecosystem of cost data that could be used as a proxy for the health of any of our customers' cloud deployments.
I continued to pull on this thread until I was able to prototype some pretty compelling demonstrations of how cost data could drive developer decisions, and notify our customers when things looked off.
After weighing the options and testing these prototypes out with a few of our closest customers, we were convinced that this was the way forward, and the original chaos engineering vision fell into the background.
What was a mistake you have made along the way?
As a former startup founder, I learned many things in my tenure as CEO of my financial planning company. While my startup wasn't successful in the way that our society would define, I found the lessons and mistakes made during this time to be extremely valuable.
One lesson that stands out is how decision making, company vision, day to day work, and market need can all exist at once (and be true), but still be disconnected with each other.
When we founded the company, our vision was to be a one stop shop -- the complete package, for financial advice, planning. There was, and still is, a significant market need for a product like this. Something your average person can go to, to figure out what to do with their hard earned money. While we worked diligently towards this goal, and had success in doing so, we ignored the trap that some startups get caught in when throwing around words like: all-in-one, complete, holistic, etc. If I had to do it again, I would be much more specific about what we as a company were setting out to do, and who we were serving. In the end our vision of a holistic financial planning solution for the average person was our undoing. We had the fire-power to put out the first iteration of this platform, but weren't able to iterate quickly due to the breadth of problems we were tackling. Specificity, iteration, and customer feedback are the lifeblood of any startup, and this is something extremely valuable that I will take with me.
What tactical advice do you have for someone in your field?
Machine learning and data science is like running a marathon. There is so much preparation that needs to be done before the average person or company even gets to the start line. I have a whole blog post just on this topic where I address:
- Data Science Misconceptions
- Challenges in Implementing Machine Learning
- Realities of Data Science Work
- Data Scientists as Change Agents and
- The Broader Impact of Data Science
Hiring a data team is a nightmare, especially when the founding team or decision makers are non-technical. Having goals of what you want to do as a company can help, but seeking out professional help from folks who can do a deep dive into your company, what drives business, how your products work, how your users use your products is the only way to prevent hiring mistakes that can take years to undo. I've seen this several times, and it's not pretty.
What advice would you give to someone who wants to build out a product?
Machine learning alone doesn't solve problems. It takes human intuition, knowledgeable experts in your domain, an understanding of the business/market you are entering, and someone who can put this all together to leverage machine learning in the ways it excels to drive a business forward.
Tactical Takeaways from Scott Strong’s Insights
- Identify Market Needs: Recognize the importance of aligning company vision with market demand
- Prototype and Pivot: Understand the value of prototyping and pivoting business direction based on data insights.
- Strategic Decision-Making: Balance decision-making with company vision and daily operations to ensure alignment with market needs.
- Hiring Expertise for Data Teams: Acknowledge the complexity of hiring data teams, especially for non-technical decision-makers, and consider seeking fractional expertise for an in-depth analysis.
- Leveraging Machine Learning Effectively: Combine human intuition, domain expertise, business understanding, and machine learning to drive business success.
Scott Strong's insights provide a roadmap for anyone navigating the complex world of data science leadership. His experiences highlight the critical balance between technical expertise and strategic thinking. From his innovative work at CloudZero to the valuable lessons learned as a startup founder, Scott's story resonates with the challenges and triumphs in the field of data science. For those looking to harness the power of data science in their business ventures, Scott's tactical takeaways offer guidance and inspiration. In partnership with GoFractional, Scott Strong is ready to assist businesses in leveraging data science for growth and innovation.
Ready to elevate your business with expert data science leadership? Connect with Scott Strong through Go Fractional to explore how data science can transform your business strategy and operations. Whether it's building a new data product, assembling a stellar data team, or developing a comprehensive data strategy, our experts are here to guide you. Book an introduction with Scott and begin your journey to data-driven success!