Big Data is a hot topic. Everyone is talking about it or working with it. It’s still too early to tell how Big Data will change the world, but the signs are there: it is changing. The data scientist is the person in the organization who has to bring about this change.
The data scientist is a relatively novel, popular profession, both with students and clients. Harvard Business Review classified it as “the sexiest job of the 21st century,” and with good reason.
Data scientist: "the sexiest job of the 21st century"
Data scientists are jacks of all trades
The data scientist is the modern jack of all trades. They combine the roles of hacker, analyst, communicator, and trusted advisor. They also have to be able to understand the business and solve complex problems with simple solutions.
Important factors to consider concerning data scientists
The most important factors concerning data scientists are:
- What skills should a data scientist possess?
- What steps should you take to find a data scientist?
- How do you bind data scientists to your organization?
- What developments are happening in the data science field?
- Which tasks and responsibilities are part of this profile? How do you implement them in existing organizations?
These questions aren’t easy to answer, but we’ll provide a starting point on this page.
What are the skills of a good data scientist?
The skills of a Big Data scientist are very diverse. They have to be a jack of all trades. Their most important skills are:
- A thorough university education in the direction of mathematics, econometrics, or statistics. Using these skills, performance management can be deployed effectively.
- Technical analytical skills, like being able to work with econometrics and statistics. The data scientist also has to be able to work with packages like SAS Visual Analytics, Tableau, Power BI, and Watson analytics. Finally: experience with R is important.
- Technical computer skills. They have to be able to work with various languages and tools. They have to be able to achieve quick results using them.
- Having an inquisitive mind and daring to think outside the box.
- Thorough knowledge of the business, and knowing which problems are important to solve.
- Being a good communicator and being able to communicate technical matters to non-technical people, like the marketing department.
These skills are difficult to unite in one person. And if you can find this jack of all trades, the question is how long you can bind them to your organization.
What are the most important trends for this function?
Developments are happening rapidly, both on the technical level (BI software) and on the business level. More and more organizations are making the difference using Big Data Science. The most important trends in this field are:
- Free birds: data scientists have to be free to look for inventive solutions to business problems in Big Data.
- Own tools first: let the data scientist use their own tools that they’re comfortable with first. There are so many different software tools available that it’s impossible to define standards, let alone force them. And the market hasn’t stopped growing yet, so leave the choice to them.
- Storage is key: ensure plenty of storage for large volumes of data processing. The cloud is an ideal solution, and in-memory data processing is gaining popularity. The disadvantage is that more and more data is stored outside of your “house”.
- Pressure is increasing: start-ups especially have to be able to use data to show that they’re viable. That puts a lot of pressure on the data scientist to be able to deliver the necessary data quickly and correctly.
- Over-heating labor market: the demand for data scientists exceeds the amount entering the market. Organizations run the risk of missing the boat. They can’t recruit a data scientist.
- From individual to team: it’s very difficult to find this jack of all trades, the data scientist. For that reason, some organizations form teams who fulfill this role collectively.
These aren’t all the relevant trends. There are many developments in the field of software development. More and more, Business Intelligence vendors are integrating Big Data in their software. In doing so they attempt to simplify and automate the data scientist’s process.
Do you want to hire or recruit a data scientist?
Data scientists are rare and hard to find. Passionned Group has been working in this field for over 14 years. This has allowed us to build a huge network. We can find you this jack of all trades, or form a team with the required competencies. If you’re a data scientist, feel free to contact us. Together, we’ll look for the right candidate for your organization.
The advantages of a Passionned Group data scientist
Data scientists mostly operate in an environment more traditionally suited to solving Business Intelligence issues. We often see a large gap between both worlds. Because of that, well-intentioned initiatives often prove unsuccessful. A good roadmap can help you bridge this gap.
All our experience and tools, which the data scientist will bring to bear, can help you avoid a large number of pitfalls, which greatly increases your chance of success. Passionned Group is completely independent, and not bound to any vendors. This gives us a unique position in the market.
How can you become a (better) data scientist?
Obviously you can’t just take a quick training course and get to work as an experienced, fully-fledged data scientist. A good data scientist does a lot of research, accrues experience, and learns things in practice every single day. The data analyst of 15 years ago isn’t the data scientist of today. A different set of skills is needed. It’s not just about making a correct analysis anymore. It’s all about using data to solve business problems in the here and now.
Opening Big Data scientists
Passionned Group is on the lookout for data scientists to help service its clients. If you want to broaden your horizons, have a new experience, or if you’re “shopping around”, reach out to us. We’d love to have an introductory conversation.
Data Science: the world of big data
Data science has become an integral part of our daily lives. Many organizations are working with it and using data science to make the difference. Big Data, and its analysis, is quickly taking over the world. Consider, for example, smartphones that can recognize objects in photos, or data analyses that can detect diseases. Or organizations like Netflix, who develop movies and shows based on the Big Data provided by their users.
Data scientists are the ones who have to accomplish all of this. They’re not the people from 15 years ago who deliver statistical reports. The data scientist of today has to independently look for clever solutions to real business problems. They make use of all kinds of structured information and very complex data sources. However, this is only possible if you have a solid architecture.
Shortage of qualified data scientists
Technology and data sources are developing so rapidly that there’s a great shortage of qualified data scientists. It’s a very attractive position. Many (young) people are studying this field now. But it’s going take some time before the scarcity problem is solved.
Until then, there’s a lot of work to do. That work doesn’t start with the question “where can we get the right people?” It starts with the question “what are the data scientists in your organization going to do, and especially, what can they add?”
How important is your data quality?
If you don’t quite know how to answer this question yet, our advice is to get started with different matters first. Business intelligence and data-driven working have to be fully developed and mature in order for data scientists to have a solid foundation.
Too often we see that organizations dive headlong into Big Data without really being concerned with their own data quality and management. You have to work on both to successfully catapult your organization into the data-driven future.
Lack of analytical and management talent
The demand for data scientists is rapidly growing. This function is becoming increasingly important within organizations and its salary is growing to match its importance. Research by the McKinsey Global Institute shows that the lack of analytical and management talent in successfully implementing Big Data is one of the biggest challenges the USA is facing. The McKinsey Global Institute estimates that there are four to five million openings for data scientists in 2018 alone. The hunt for data scientists has been opened. This jack of all trades in your organization should possess many talents in order to help shape and direct the explosive amount of new possibilities provided by big data. But is this realistic?
The Data Scientist: the superhero that doesn’t exist
The requirements for data scientists regularly include things like a valued university education in a field like math, statistics, or econometrics. Or technical analytical skills, like being proficient with data mining and the associated software. Also, the data scientist should be able to use several programming languages and tools and be able to achieve quick results using them. The data scientist should be a researcher who dares to think outside the box and has inside knowledge of business processes and problems. Finally, they should be an excellent communicator, so they can communicate and explain all the problems and possibilities that come with the data. They should be a champion of data-driven working and innovation. But this desired superman (or woman) clearly doesn’t exist!
A jack of all trades is a master of none
Is it possible to recruit this fabled data scientist? And even if you were to find one, a jack of all trades is a master of none. Two types of thinking combine to create this unrealistic laundry list of requirements and characteristics:
- The old way of organizing where people bet all their chips on a specialist who comes in like a knight in shining armor to make their organization innovate and work data-driven. This way of thinking is a vestige of the second industrial revolution, where radical division of labor resulted in enormous increases in productivity. Nowadays, this conflicts with the type of agile organization where teams make all the difference. In addition, an increasing number of organizations is on the way back to working with roles instead of functions. More generalization in tasks and roles is crucial in order to facilitate dialogue and self-organization, which is required to embrace agility and survive as an organization.
- The second way of thinking is about the desire to break up complex building blocks into easily digestible chunks. For the position of Data Scientist it is common to see an overlap of many different competencies, skills, and characteristics that the company is apparently unable to unite, so they are all combined into this one mythical person. Just check the job openings! This creates a demand for a super specialist that doesn’t actually exist. Besides that, getting such a specialist carries the risk of missing the true dialogue with each other and instead getting stuck in an “ask and we deliver” mindset, instead of having a valuable discussion about the why, how, and what of solutions and improvements. This often leaves the best ideas and the required support to go in a new direction unused.
Are you still thinking like a Fayolist?
Henri Fayol was a French mining director who made an important contribution to organizational science in the early 20th Century. He formulated a number of important principles of organizing that are still common in behavior and vision today. These ideas are fantastic for creating scalable organizations but are less productive for agile, self-steering teams.
Fayol posited fourteen principles of organizing, the following five of which were the most important: specialization, unity of command, formal chain of command, unity of direction, and authority and responsibility. These principles clash with those of agile teams who need to be flexible and sometimes even disruptive. The Fayolistic mindset has often led to management patterns that keeps basing its decisions based on it, sometimes consciously and sometimes unconsciously.
Cheap or expensive teams?
These days, change takes place in the roots of an organization, so only teams can anticipate, analyze, and solve problems quickly and effectively. This requires agile employees, methods that support an agile workflow, and a combination and overlap of knowledge and competencies, including those of a data scientist, which a team or project team should contain.
It’s crucial to determine which competencies require overlap and which do not, in order to prevent making an expensive team comprising nothing but the jack-of-all-trades super specialists. Teammates with different degrees of analytical skills can easily create a rich dialogue about the meaning of data and possible solutions.
At the core, we only know a select few groups of competencies. These are task or role-focused, intellectual, emotional, organizational, social/communicative, and finally, developmental competencies. In this agile world, team competencies can be seen as an extra ingredient, although these largely overlap with the social/communicative competencies. As soon as the goal of a project or a team becomes apparent, it can quickly be determined which competencies are required in order to achieve the set goals. It is important to check the importance of dialogue, cooperation, separation, and autonomy in steering and decision-making. Based on these principles it will become apparent which core competencies should be distinct and which should overlap in order to quickly make indicators.
The most important data science skills are relational
Google’s Hal Varian confirms this when talking about data science skills: “The ability to take data – to be able to understand it, to process it, to extract value from it, to visualize it, to communicate it – that’s going to be a hugely important skill in the next decades.” Most of these skills only make sense in relation to colleagues. Data only takes on meaning in the context of a group of people or between individuals. Data advances and feeds relational processes.
Agile data science: get started quickly!
Do you want to work with data science quickly and effectively? The following suggestions help your organization to quickly reap the rewards of data science:
- Many organizations still lack true agility. Companies and managers often follow the trends of the day and have a tendency to react to everything without an overarching plan for agile working and effectively applying data science. What is the overarching plan and what are your goals?
- Data scientists often operate in organizations with a traditional structure or with managers who think in traditional ways. We often see a large gap between the world of Business Analytics and Data Science on the one hand and the world of business and commerce on the other hand. Because of this, initiatives with good intentions turn out to be failures. A good BI road map and cooperation in an interdisciplinary team can help to shorten this gap. What’s your Data Science road map?
- Data Science is a team effort. Focus on the required knowledge and skills and the overlap within your team instead of pouring all your money into hiring expensive and unrealistic superheroes. You also run the risk of crucial processes stagnating if and when this person leaves your company for the higher salary and better benefits offered by your competitor. Do you have a team with the right skills? And how do these skills overlap?
- Teams that offer the safe space to introduce differing opinions in discourse provide the best opportunities for improvement and innovation. Are you fostering cooperative discourse?
- Knowledge of team processes and innovation processes are equally as important as knowledge of the content. Invest in these core competencies, as they are a crucial part of agile working and innovation. These competencies improve the motivation and the productivity of the team and its members. How strong is your knowledge of team and innovation processes?
- Continuously reflect on the process, cooperation, and everyone’s contribution. This speeds up personal growth and everyone’s learning curve, as well as that of the team – another advantage in relation to the Data Science super specialist. Only feedback within the team makes short, cyclical, iterative learning possible. This reflection process can often be supported by both competencies as well as methods of improvement, such as PDCA. How does your organization foster reflection and cooperation?
- Personal growth, growth of competencies, team growth and organizational goals should often be centralized. How can I, the team, and the organization advance? And who or what is needed for this?
See also: What the future of work will mean for jobs, skills, and wages.
Interested, or want to know more?
If you want to know more about data scientists or you want to hire one, reach out to us. You can freely contact us for advice or a conversation about your organization.
About Passionned Group
Passionned Group is a leading firm in designing and implementing Business Analytics. Our senior Data Scientists and consultants assist both large and small companies towards a smarter enterprise. Every two years we organize the election of the smartest company.
Our Data Scientists
DAAN VAN BEEK MSc
Data Science consultant & CEO
HERMAN VAN DELLEN MSc
Data Science Consultant