Three Ways Data Scientists Can Differentiate

My post, Dear Spreadsheet Jockey, Welcome to Big Data covers what a data scientist should not be. A data scientist should not be a spreadsheet jockey with better technical hands on skills and access to better data profiling and analysis tools. The problem here is the word "scientist" which implies experimentation, developing hypothesis and presenting results. A data scientist needs to do a whole lot more if they are to successfully help companies develop actionable insight from their data sources.

Here are three ways data scientists can differentiate themselves:

Data Scientists Must Speak Business First

IBM's post on What is a Data Scientist captures this best
What sets the data scientist apart is strong business acumen, coupled with the ability to communicate findings to both business and IT leaders in a way that can influence how an organization approaches a business challenge. Good data scientists will not just address business problems, they will pick the right problems that have the most value to the organization. 
A data scientist can't just be handed the questions that need researched. The scientist needs to study the data, identify patterns, prioritize their time aligned with business priorities and report back insights that can make financial impacts. 

Data Scientists Must Inspire Technologists

A few months ago, Forbes suggested that the Data Scientist Will Be Replaced By Tools. I disagree. Automation and algorithm design are the tools of data scientists as are data visualization, profiling, cleansing, modeling, master data management and other tools. Someone needs to own the development of these algorithms and to perform analysis. Is that engineering? Well yes, there are many technologies to master, but its iterative discovery that makes this role closer to a scientific process than an engineering one.

It's also unrealistic for the data scientist to fully master all of these technologies, or for them to be efficient working across all of them. For this reason, data scientists need to inspire technologists that can be part of a team developing algorithms, dashboards, and other tools to help the business leverage its data.

Data Scientist Must Develop Data Products

Data Scientists can't work in a lab... and can't work alone. Scientists need to help define new business opportunities, or help change how businesses identify new markets and customers, or help optimize operational processes. To accomplish this, data scientists have to help Management become part of the discovery, ask questions, and see answers. They can't service all of these opportunities and stakeholders themselves. It simply doesn't scale, which is part of the reason everyone is looking for data scientists.

The answer lies when Data Scientists can deliver self discovery tools to managers and make them part of the process. These tools, or "data products" need to be simple and provide managers lenses into into the enterprise's data. They have to provide enough flexibility to enable managers develop their own insights that lead to action.

Put these three skills together, and it is no wonder it is hard to find good data scientists.

1 comment:

  1. True indeed. Document archiving have been an important factor in data management of big companies.

    David Huffman


Comments on this blog are moderated and we do not accept comments that have links to other websites.