Data Science

Data Science comes of age. So what?

07 Aug 2015 | Share

Back in my school days, my parents always said that I should challenge the status quo and go after my goals rather than just daydream.

My mom’s favorite sentence back then was: “If you want to win an Oscar, you should star in the movie.” Nowadays, I can’t help but smile whenever I hear this saying and my thoughts drift off to the world we live in today, where businesses’ main objective is to put their data center-stage and turn it into insights in order to solve business challenges. Only then, are they escorted to the ROI stage to claim their Oscar.

Numerous organizations continue to suffer from data siloes or legacy systems. Their IT teams constantly adhere to the ever-changing requirements of business and technology, which in turn results in complex patchwork landscapes made up of proliferated systems, rather than driving strategic decisions using agile and modern solutions. KPMG’s latest study reveals a staggering result: 85% of those surveyed state that one of their overriding problems with analytics is determining the right solutions to use in order to examine and interpret data.

Bon voyage Big Data

Some 2.5 (1018) quintillion bytes of data is produced every single day. But, here comes my first “so what?”

I believe that the majority of us have already realized by now that data is haunting us everywhere we go. This has given rise to one of the overused terms I have witnessed in a lifetime: “Big Data”. Yes, Big Data is getting bigger every millisecond, but most businesses are just trying to leverage a slice of it, rather than delve into unnecessary analyses of an unexploited ocean of data.

Drum rolls please: here come the data scientists

The role of data scientists is said to have emerged in order to ride the Big Data wave and make sense out of it. They are individuals that are hired to answer the business questions and take the right decisions. The biggest hope in those individuals is that they unearth innovative insights that will accelerate new revenue streams in order to grow the business.

But first, who are those people?

“Let’s call them data scientists, it sounds more exciting,” the industry said, and so it became their title. If we could travel back in time, data science 10-15 years ago was inherently part of data mining and business intelligence, terms that now make up the term “data science” today. We seem to have gone full circle.

And yet, I am an avid supporter of simplicity. To me, the job of a data scientist is to understand the raw data extracted from systems and sources, analyze it, and make it more valuable, accessible and useful for a company.

A normal working day would not only include diving into data, deriving SQL queries and creating reports. An average day would also combine translating technical issues and data discoveries and explaining them to stakeholders in the firm, and also working in the opposite direction, where the needs and concerns of management are conveyed to the technical side of the business. In simple terms, data scientists act as the interface between the technical and business worlds.

What skills does their fancy title embrace?

There are several lists that detail the skills and attitudes required for those aspiring to become data scientists. As an employer, let’s take a look at 4 crucial knowledge areas that data scientist applicants should cover:

The technical tool kit: a solid understanding of mathematics and algorithms as well as several programming languages including R, Python and SQL is key. But I assume you already knew that. The art is to improve a business’s logic and enable it to work across different platforms. Data scientists are not only math geeks; not understanding how their algorithms impact the overall business can be devastating so a strong mathematics skillset alone is definitely not sufficient.

This leads me to my second point: owning a commercial mindset and industry knowledge. A data scientist should be capable of staring at data and spotting industry trends, as well as comprehending how the business functions and what the business hopes to achieve from data analytics in terms of profit. You might be wondering why a data scientist should have such a mindset. Well, the answer lies in having one vision across all business units. It is fundamental to make sure the corporate goals are understood while appreciating the business’s technical requirements, too.

Having an understanding of human behavior and communicational skills is the third element that shapes a data scientist in my opinion. The ability to present and convey the results of data analyses, and their impact on the business, to stakeholders is key when hiring a data scientist. In turn, this will lead not only to the successful and effective delivery of information, but it will also avoid any misunderstanding between the business and technical parts of an organization.

Finally, personal motivation and intellectual curiosity define a perfect candidate. Determined to optimize business processes and unearth value by heading in new and profitable business directions is without a doubt the fuel I need for a passionate company like ours. Staying up-to-date by reading books, news, blogs and MOOCs is crucial.

My favorite statement that describes a data scientist comes from William Vorhies: “This is a profession that requires life-long-learning and intense curiosity.”

Skills that sound great, but do such people actually exist?

Finding good data scientists continues to be a challenge. Reasons include finding people who have the aforementioned set of skills which are needed to complement the massive shift to data-driven businesses, while today’s technology allows large amounts of data to be processed and analyzed. Here are some statistics to prove my point: McKinsey reports that by 2018, between 140,000 and 190,000 data scientist job openings will remain unfilled. Despite salaries skyrocketing for data scientists due to the immense demand, finding such individuals to bridge the gap remains difficult. Burtch Works issued a study just 3 months ago that indicated that the compensation of data scientists is on the rise.

Therefore, a well thought-out strategy should embrace the creation of Data Science teams which combine a mix of people with the aforementioned skillsets. Even if the perfect data scientist is nowhere to be found, organizations should be able to blend the know-how of a great technical person with an experienced project manager in that field.

Lifting Data Science to a strategic level

The director of Burtch Works, Linda Burtch, states that “within 10 years, if you’re not a data geek, you can forget about being in the C-suite.” I prefer to take a step back and reflect on this strong statement before I agree with Burtch. I am a firm believer that Data Science should definitively be a key element of a company’s strategy. That doesn’t meant that every executive director has to be a data geek, but rather be responsible for putting data scientist teams together and ensuring there is synergy among them. Hence, creating a data-driven organization that is backed up by the observations and decisions of data scientist experts is indispensable. It is not by chance that more and more companies continue to introduce the role of a chief data officer (CDO).

Parting thoughts

There can be little doubt that businesses want to get the best value out of their offerings and maintain the best position in their industry. It is also clear that we are not in the world of doing each other a favor, but rather uniting to offer the best experience for organizations, customers and consumers. This is why I have a simple call to call for those businesses: just make sure you reflect for a moment and ask yourself how you would achieve those goals.

A big part of that answer lies in Data Science. While many have called it the sexiest job of the 21st century, I do believe it is the most important job for sustaining a competitive advantage among businesses for at least for half a century to come. This highlights how essential it is to hire data scientists who can and do think on their feet, ensure that all stakeholders speak the same language, and integrate quality, agility, and vision in everything they do. Only then it is possible to remain in the results-driven game while maintaining a holistic strategy.