For a client, a data science project usually begins in one of two ways. One, the client says, “We have this business situation, and we need help resolving it.” Or two, a client says, “We have this data set, and we think we could use it to benefit our business.” Think of it as clients who want to know an answer and clients who want to know what’s possible. Both are endlessly fascinating and worthwhile.
For a data scientist, however, a data science project always begins the same way: Questioning.
The first step in the data science is “questioning,” which is pretty much what it sounds like. Before we do anything else, we have to understand the business problem, and to do that, we have to ask lots of questions. Our data science team, our executives, our sales team, and our technical solutions owners are all asking questions, getting to know and understand your business.
We need to understand the background, the audience, what business units are involved, and what data will be needed. Or, let’s say you already have a data set. If that’s the case, we ask questions to better define the opportunities. We might ask to do some initial data analysis and figure out what is available.
As much as data scientists must begin working with a set of basic questions and requirements, we must also keep open minds. As a January 2020 article in CDOTrends, 4 Skills Data Scientist Should Have, states, “the onus is on data scientists to identify real and present issues that may be unidentified or missed out even by stakeholders.”Our clients rely on us — not just for our education and expertise, but for our guidance (both common sense and insightful). So we brainstorm. We ask more questions.
- What do you hope to gain?
- Is this really the problem to solve?
- What type of inputs do you want me to use?
- What type of outputs do you need?
- Who else is involved?
- What else do we need to consider?
As you might imagine, it’s hard for data scientists to succeed if we don’t understand what we’re hoping to accomplish, so we keep the conversation going – and the lines of communication open.
One of the things I enjoy most about the “Questioning” step is that it’s hugely collaborative. This is not a “go back to your desks and ponder” step. “Questioning” actively involves lots of participants, not just on the TKXS side, but on our prospective client’s side as well. By talking — and equally important, listening — we acquire a genuine and common understanding of a client’s business problem, which is, of course, key to solving it. Only after we’ve properly framed the problem — to everyone’s satisfaction — can we move on to the second step of the data science process: Gathering.
Read more: The Seven Essential Steps of Data Science
Watch more: Data Science: A calculated approach to enhance business performance.