Make Way for the Business Scientist

Data permeates today’s business environment and has sparked important changes in how decisions are made. Historically, an effective VP of sales or service may have tried to innovate by developing hypotheses about how to improve business processes, but access to data and analytics tools has made it possible to test, compare, and refine ideas faster. Where once the only option was to test an idea using live experimentation in the market, today’s decision-makers can compare and contrast ideas using sophisticated analytical models developed by their firm’s data scientists.

There’s a well-documented gotcha inherent in this view: while business folks have tons of interesting ideas worth exploring, the data scientists needed to explore the relevant data with advanced mathematics and programming skills are a precious commodity. It would be great if everyone in business had a data scientist riding shotgun all the time, but that’s not going to happen. 

I know what you’re thinking. The tale of the data science unicorn is not a new one, nor is the notion of becoming a "citizen data scientist". But I disagree with the narrative that has developed around these ideas. Citizen data scientists do what data scientists do with simplified tools and at a less sophisticated level. This paradigm gives primacy to data science and undervalues the business expert. Furthermore, I don’t know anyone who thinks of themselves as a citizen data scientist.

I’d like to introduce a new, more nimble alternative – the business scientist – individuals defined by their business acumen and not their comfort with higher math. They are the kind of people who look at business problems and seek to answer the question, “What if?” I’d argue that if you’re a business expert, you are probably already well on your way to being a business scientist. 

The approach of the business scientist reflects what a high school teacher might describe as the scientific method. They start with a hypothesis – a story that describes a business challenge, oddity, or opportunity – as well as the beliefs that support current thinking or expectations. 

The laboratory bench for the business scientist is self-service AI powered analytics. This is the critical game-changer for the business scientists because self-service AI says that you don’t have to be a mathematics or programming wizard to undertake rigorous data exploration. Instead, automated discovery augments the business expert’s capabilities and enables them to ask challenging questions and test theories using natural language and visual tools that are familiar to them, rather than writing code. 

When the self-service AI process comes up with something that requires further investigation, what is needed is someone who knows the business and is capable of making sense of why that anomaly happened or what to do about the correlation being presented. Beer and diapers is the famous unexpected purchase combination. Expect AI to hit you with these types of odd ideas all the time. It can’t understand the implications, but the business scientist can. 

The business scientist is the user of self-service AI powered Analytics and organizations of all kinds, across many industries, will need hundreds of thousands of them.

We are at a great time and place for your role as a business scientist to emerge. The availability of self-service analytics tools powered by AI (Augmented Analytics) means domain experts can undertake meaningful research toward valuable business insight.

This does not replace the contributions of data scientists or suggest that business scientists don’t need support from data science experts. It lays the foundation for getting the best of both of these roles.

I particularly appreciate MIT’s Eric von Hippel ideas about user-driven innovation. He suggests that when consumers don’t have requirements dictated to them but instead freely solve for their needs and desires, innovation accelerates. He explained how mobile banking emerged not from banks but from consumers who used telephone calling cards to transfer funds to friends and relatives around the world. (Intrigued? Check it out here.) 

Similarly, when a businessperson, who is the end-user or consumer in the context of business, is able to directly leverage their extensive domain knowledge using the mechanisms of data scientists, there is new potential for establishing a cycle of innovation. AI Augmented Analytics enables business scientists to build solutions that previously were gated by the availability of data scientists.

Ultimately I am advocating not for the creation of a new job title, but a new way of thinking about the intersection of data science and business. One is not better than the other. They are complementary. Therefore a more effective model is one that enables a business to do more business science in general, with data science being just one of the pathways available to them.