In a modern business context, there’s no situation in the world where it’s okay to say “As the leader, I don’t understand what’s going on, but my team does, and that’s good enough.” Yet, business leaders frequently claim ignorance of the most basic principles of data science.
Leaders don’t have to know the intimate details of data science processes – they’d have no need for a data scientist if that were the case – but as the line between big data and business operations disappears, it’s more important than ever for business leaders to speak a little “data science.” As such, here are a few important things business leaders should know about data science.
Why It’s Important To Understand The Basics
Data science is good storytelling, but it is still science. Telling a story can often obscure the facts or make links where there isn’t any. Proficiency in the fundamentals can help you avoid
- getting taken – manipulating the data, not telling the whole story, targeted information gaps, all these things could make it easier to coerce you into a bad decision.
- asking the wrong questions – data pulls are only as good as the question you’re asking. Data must be evaluated regularly, and that requires the right starting question.
- replicating bias – data is neutral, but its aggregation and results are often the product of our preconceived ideas. Understanding the basics of data science helps you sort out the messiness of data in the real world.
What Business Leaders Should Know About Data Science
You don’t have to understand the ins and outs of building a system or the intimate details of statistical analysis, but you do need to speak a bit of the language. Here are XX fundamentals you can’t get away with not understanding.
Data can tell a story and data can also tell a story (if you know what we mean). Understanding how things like color, baselines, reference points, and choice of presentation affect our psychology and can either highlight or obscure what the data actually says is a critical part of understanding data science.
Some data science teams may feel immense pressure to find the story in the data, so understanding the how and why of visualization could help mitigate the tendency we have to find patterns or “torture” the data. Leaders that understand the basics of visualization can effectively manage reporting and accept the data as it stands.
We think of data as inherently neutral, but data analytics relies on quality generation. If you don’t know how data is generated, you can’t judge the quality of the analytics. Deferring to your data science team for complicated analytics could lead you down the wrong road.
For example, if a data set included a large group of people who previously expressed an interest in your company’s product, it would be difficult to judge the true effectiveness of the ad. You’d never know if the ad was effective or if the customer was already considering a purchase.
You don’t have to know how to rebuild an engine from scratch just to get in the car and drive it, but you do need to know a few basics to drive responsibly. Like the engine, data tools don’t need to be fully understood, but you should understand a bit about the tools in the pipeline to lead your teams.
A basic understanding of the tools your team is using to generate data, store data, and visualize data can help you give your team more operating freedom (because everyone is on the same page) while alerting you to potential difficulties if there’s a breakdown. You’ll also understand how to budget resources to a data ecosystem instead of taking a “good enough” approach with the simplest infrastructure.
Domains Outside of the Data
Your team is full of data experts, but they may not all be business experts. We aren’t saying to return to “gut feelings” (quite the opposite), but your experience in business, healthcare, finance, or any other field outside of pure data science is a crucial part of understanding that data.
Say you run a targeted ad campaign for a particular product, and that product seems to do wildly well in the first quarter. While you may be tempted to take the data as is, understanding data science teaches you that if something is too good to be true, it probably is. A broad field knowledge can help interpret the data more correctly.
Data Science Roles
Data Scientist. Data analyst. Data engineer. It’s all interchangeable, right? Not exactly. Data science may be the sexiest career of 2019, but there are important distinctions between each title.
Knowing your data engineers from your business analysts can help you find the right person for your team and give you an idea of realistic expectations. The roles do intersect, but expecting your data analyst to maintain overall pipeline architecture and create systems to extract and clean unstructured data could be asking a lot if you aren’t aware of the difference.
A New Era of Business Leader
As companies prepare for big data integration, business leaders need to adapt to their roles as team leaders for their data science employees. Your data science team should have the expertise to process data with a lot of freedom, but you still need to understand the basic structures of what’s happening to create value from that data. Make sure you’re ready for big data’s emergence.
Editor’s note: Want to learn more about how to get your AI initiatives off the ground? Attend the Accelerate AI conference this April 30-May 3 in Boston and learn all about connecting AI to your business today!