Business Analytics Professional or Data Scientist?

Data scientist (DS) and business analytics (BA) job demand in the United States has continuously been on the rise. Miller and Hughes project that “by 2020 the number of positions for data and analytics talent in the United States will increase by 364,000 openings to 2,720,000. By the end of 2020, job openings for data scientists and similar advanced analytical roles should reach 61,799. This is a significant number, but it represents just 2% of the projected demand across all job roles requiring data and analytics skills” [1]. Certainly, there is a greater demand for data science and business analytics professionals than supply at this point in time. Thus, a need exists to train professionals with the skill sets required for data scientist or business analytics jobs.

The U.S. Bureau of Labor Statistics published employment by major occupational groups and predicts a 12.1% growth in employment under computer and mathematical occupations between 2019 and 2029, which is the highest growth in any sector other than the healthcare sector and community and social service occupations. Even though it predicts growth, it is intriguing that this Occupational Outlook Handbook doesn’t yet include specific occupation categories for data scientists and business analytics (they have “Business Analyst”). This shows the lack of awareness in the general public about the differences in DS or BA professions.

The fields of data science and business analytics are still evolving and are yet to realize their full potential. PwC revealed that “by 2021, 69% of employers expect candidates with data science and analytics skills to get preference for jobs in their organizations. Yet only 23% of college and university leaders say their graduates will have those skills” [2]. This shows us the existence of a talent shortfall or a skill mismatch that has left recruiters to decide how candidates meet the skill requirements in the evolving ecosystem. The study further highlights the need for “a common nomenclature to trade in DSA competencies and skills, and a closer, more collaborative relationship with higher education aimed at creating programs that will provide job candidates with the skills they need.”

Other than Radovilsky et al. [3] who compare the individual skill sets required for DA or BA professionals, a large share of the research in this field has been concentrating on a particular vertical of DS or BA (such as Seal et al. [4]) or focusing on the academic requirements of data science or business analytics courses like De Veaux et al. [5]. Therefore, we see that there is a need to analyze and “create a common nomenclature” for the different skill sets required for a data scientist and a business analytics professional and provide a career pathway for deciding whether a BA or DS career path would suit an aspiring candidate.

Literature Review

As Donoho stated, “The would-be notion takes Data Science as the science of learning from data, with all that this entails. It is matched to the most important developments in science which will arise over the coming 50 years” [6]. Most of the existing research in the field of data science and business analytics has been predicting significant advancements in the field. The research content in the fields has been constantly evolving, and more universities have started to offer big data/analytics courses. According to Mills et al., there were dramatic increases in courses offered in big data analytics (583%), visualization (300%), business data analysis (260%) and business intelligence (236%) between 2011-2016 [7].

Radovilsky et al. performed a holistic comparison among the different skill sets of business analytics (BA) professionals and data scientists (DS) that are required from an industry perspective [3]. They explored the job descriptions published in LinkedIn, Glassdoor, Monster.com, Dice.com and Indeed.com, and used the term “cloud” to visualize relative frequencies of terms in a document data matrix. Further, the study focused on identifying “groups of knowledge and skills required for business analytics and data science professions.” In addition to the three knowledge domains of business, technical and analytical skill sets that are used in Cegielski and Jones-Farmer [8], Radovilsky et al. applied a new knowledge domain, communication, which according to them is important on its own as part of the required skill set for DS or BA professionals [3]. Radovilsky et al. highlights that “BA has a major emphasis on business-related skills including general business and effective data management, and technical skills such as database design and management, as well as data warehousing. Contrary to that, the DS groups of skills focus on the technical skills related to software development, algorithms and languages, software and data management.”

Another research in the same direction was conducted by Stanton and Stanton [9]. They analyzed job postings to determine the skills needed for a career in data science, data analytics and business analytics. They categorized their findings into three categories: credentials, soft skills and software skills. According to Stanton and Stanton, the most valued credential was prior experience, and even entry-level jobs in DS, DA (data analytics) and BA desired individuals with five or more years of experience. A degree in business was rated second for business analytics jobs, and a degree in computer science was rated second for data science. In terms of hard skills, data analysis, programming and artificial intelligence were the most preferred for data scientists, while data analysis, modeling and business strategy were the ones that were most preferred for business analytics positions. In terms of soft skills, data-driven analytical skills and written communication were the most sought after in data science, while analytical and problem-solving skills, along with written communication, were prioritized for business analytics jobs. The study also stated that Python, Java and SQL were the most favored software skills.

Other than the above-cited research, De Veaux et al. [5] state that the key competencies for an undergraduate data science major are: computational and statistical thinking, mathematical foundations, model building and assessment, algorithms and software foundation, data curation, data preparation and management and knowledge transference – communication and responsibility. On the other hand, Parks et al. [10] found out that on an average, the courses in the data analytics programs in U.S business schools had 31% courses that focused on developing an analytical skill, 38% IT knowledge and skills and 31% business knowledge and communication skills. De Veaux et al. state that students with a background in mathematics, statistics or computer science may be viewed ideal for graduate study in data science. Parks et al. state that the graduate program in business analytics is appropriate for students in functional business units and the sciences, as well as information technology because it leverages information technology and business thinking to turn data into actionable intelligence.

The study by Mills et al. [7] provides a different dimension and analyzes how the academic landscape in information systems changed from 2011 to 2016 in response to the industry demands for data scientists. They grouped the newly added courses based on the four pillars of analytics that was suggested by Kang et al. [11], which are: data processing storage and retrieval, data exploration, data product, analytical models and algorithms. The results suggest that upon a comparison of the syllabi in 2011 and 2016, pillar 2 (data exploration) and pillar 3 (data product) experienced the largest growth. This includes courses in visualization, business data analysis and business data analytics, and supports prior claims that programs are rushing to develop curricula in this area. This gives an idea of the skill sets that graduates possess after completing formal education in data science from the selected universities that were studied.

Seal et al. [4] use text mining on the job descriptions posted between October 2018 and January 2019 containing the following terms: business analyst, business analytics, data analyst, data analytics and data analytics manager. This was compared with the core and elective course descriptions in various universities to do a supply/demand gap analysis. They point out that some of the skill sets that are overemphasized in academia are of lower priority to the employer. Prescriptive models, unsupervised learning and big data cover 71.4%, 67.3% and 53.1%, respectively, of the programs teaching competency in core courses while the percentage of job ads seeking competency in these were 2.2%, 2% and 7.8%, respectively. On the other hand, some soft skills such as communication team skills, data reporting, database tools and spreadsheets should be better addressed in academia to better equip the graduates. Even while restricting their scope to business analytics, the authors Seal et al. [4] state that “Data scientist positions indeed need depth in technical skills, whereas the other positions (business analytics) need more business-oriented skills with a broad knowledge of the technical toolsets in the business analytics area.”

Verma et al. [12] conducted a content analysis on the online job postings from December 2016 to February 2017. The job categories investigated in this study were data analyst, business analyst, data scientist and business intelligence analyst (DA, BA, DS and BIA). They grouped the skills into seven categories: decision-making, organization, communication, domain, structured data management, statistics and programming. The study concludes that decision-making is the desired skill for all jobs while communication and structured data management followed. These conclusions were in line with the observations made by Seal et al. [4], which highlighted the importance of soft skills such as communication and team skills while conducting a text mining on the job descriptions.

Delen and Ram [13] have identified that business analytics could be categorized into descriptive, predictive and prescriptive analytics and have identified skill sets that each of these categories requires. Their study postulates that descriptive analytics generally requires business reporting, dashboards, scorecards and data warehousing skill sets, while predictive analytics would involve data mining, text mining, web/media mining and machine learning. Prescriptive analytics is viewed as the top of the hierarchy with expertise in data optimization, simulation, decision modeling and network sciences required to predict the best possible business decisions and actions.

Foster [14] conducted an industry specific analysis of data scientists. He selected professionals working in supply chain management (SCM) and analyzed their responses to see what skill sets data science professionals working in the SCM industry currently possess. He collected data by posting in various SCM groups in LinkedIn. He determined 22 skills that were divided into five skill groupings – business skills, statistics skills, programming skills, math/operations research skills and machine learning/big data skills – and asked the respondents to identify the skill sets that they possess. Out of the 22 skills, business skills was the most common followed by data manipulation and visualization. As part of the survey, Foster asked the participants to self-ID themselves to understand how the DS/BA professionals in SCM viewed themselves. The results show that leader and businessperson were tied as the most self-identified followed by researcher and jack-of-all-trades.

Fayyad and Hamutcu [15] studied a detailed review of the research that has preceded them in the field of data science and created a framework of the skill sets that are required in the data science profession. Their framework consists of two domains: 1) science and math, and 2) programming and technology. Under science and math, the authors have incorporated scientific method, mathematics, computer sciences, statistics, operations research & optimization, data preparation and exploration and machine learning. Under programming and technology, the skills include general purpose computing, scientific computing, database and business intelligence and big data.

In addition to the skill set comparison required for various fields, Davenport [16] suggests that AI is one way to help bridge the talent gap experienced by organizations in a shortage of skilled analysts and data scientists by automating tasks, increasing productivity and improving outcomes. Davenport indicates that automated machine learning systems can dramatically improve the productivity of data scientists or analytics in model development and deployment.

Mehrabad and Brojeny [17] performed research to understand the use of intelligent and systematic methods in human resource management and personnel operations of an organization. According to the study, job analysis is one of the most appropriate domains in which expert systems can be used. As advised in the study, research is aimed at applying a structured multicriteria decision-making model, using the analytic hierarchy process (AHP) to determine whether the business analytics or data science route may be most applicable for an individual seeking career advice in these two fields.

Research Methodology

To examine the skill sets for future jobs in the business analytics and data science professions, we used research data triangulation – specifically, a literature review, sampling from the INFORMS analytics community and industry best practices. Based on the literature review and focusing on Cegielski et al. [8], we relied on three types of skills required for the BA/DS profession: 1) business and soft skills, 2) technical skills and 3) analytical skills. Using these three categories and informed data from our full literature review, INFORMS analytics community input and industry best practices, we developed a BA/DS skills comparison framework as shown in Table 1. The skills highlighted in red denote the ones required for business analytics professionals alone, the ones in green denote those required for data scientists alone, and purple denotes the skills required for both.

Table 1: Major skill sets required for business analytics professional, data scientist and both.

Analysis, Findings and Recommendations

Our analysis shows that of the 16 skills listed in the business and soft skills section, 11 are desired skills for both the business analytics professional and data scientist. Some of the business-specific skills such as defining metrics and KPIs, developing a plan of action to implement business decisions derived from the analyses, developing strategies based on data, assessing dollar value to the company of predictive models and project management were skills that were mainly desired for a business analytics professional.

Under the technical skills, data warehousing and business intelligence are the ones that are desirable for a business analytics professional, while software development and technical or computer science skills are seen as desirable for a data scientist. Programming knowledge in SQL, ETL and OLAP skills, database design and management, processing cleaning, validating structured and unstructured data and other basic IT knowledge are desired skills among both data scientists and business analytics professionals. Data reporting, product development, proactive problem solving and data storytelling are also seen as required skills for a data science or business analytics professional.

Out of the 21 analytical skills we identified, statistical analysis is one skill that is often applied to BA professionals, yet it lacks the same importance among data scientists. We saw that mastery of analytics tools, visualization and dashboard design, predictive analytics and strong computational and statistical thinking are desired for both the business analytics professional and data scientist. However, a data scientist is expected to possess strong proficiency in data mining, AI/machine learning, web/media mining, programming languages like Python, R and algorithms, model building and deep learning skills, data exploration, representation and transformation. A data scientist is expected to keep track of the latest developments in the field of big data technologies and proficiency in Spark, MapReduce, Hadoop and Snowflake, along with natural language processing (NLP), XML and JSON mastery. Overall a strong coding and computer science background are highly desirable for a data scientist.

From the above classifications, if we consider a scale with business and soft skills on the left of the scale with technical skills in the middle and analytical skills on the right (as shown in Table 1), we see that as we move starting from the center to the left, business analytics skill sets increase with a decrease in the data science skills, and when we move to the right starting from the center, the data scientist skill set increases.

A strong business analytics professional could identify her/himself as possessing a strong affinity to understand the business and interpret insights from the data to develop a strategy. She or he would have strong statistical and mathematical skills with the ability to communicate the inferences obtained through data analysis and be proficient in data visualization tools and dashboard designs.

On the other hand, a data scientist would be someone who could easily program and work with computer coding while being proficient in big data technologies. She or he would have a strong affinity toward AI and data mining techniques and would be very comfortable with data transformation tools. Model building and assessment along with software development are some of the core strengths of a data scientist. She or he should possess an understanding of the general business context and should be able to apply critical thinking and communicate his or her insights from data with the help of data visualization tools.

Based on our analysis and the framework created, we developed an Excel-based model to discover how the skills possessed by a potential candidate matches with the skill requirements for a business analytics professional or data scientist. The 51 skills that we identified were given weights based on our literature review and from our own experience, as well as that of others in the field. The user would have to input “1” if he or she is interested/currently possessing the corresponding skills or “0” if he or she is not interested in acquiring the skill. Based on the input, the model provides the calculated percentage with the skills required to be a business analytics professional and the calculated percentage with the skills required to be a data scientist.

Users can conduct an analysis of their skill sets by clicking the following test model link below. Please note that you may input 0,1 in cells B3, B18, B20, B33, B35 and B55 – “0” if you do not possess the skill or are not interested in acquiring it in the future and “1” if you possess the skill or are interested in acquiring it in the future. The result of our calculation will be shown in cells F3 and G3. F3 represents the percentage by which your skill set aligns with that of a business analytics professional and G3 represents the percentage by which your skill set aligns with that of a data scientist.

TEST MODEL.XLSX

Future Research

We have tried to limit our work to identifying the skill sets that are required for a business analytics professional and a data scientist. We hope that this analysis and resulting spreadsheet model will help those individuals in determining the right BA or DS career path for them to take. This should also help students in determining which graduate program to pursue, if any, in either the BA or DS discipline. This research can be extrapolated to develop an AI-based system for HR professionals to assist in identifying people with the right talents to meet their BA or DS requirements. Further careful analysis of the skill sets can be done to identify which among these have a high priority for a particular role. A careful ethnographic analysis on the daily activities of a BA professional or data scientist could provide insights as well.

References

1. Miller, S. and Hughes, D., 2017, “The Quant Crunch: How the Demand for Data Science Skills Is Disrupting the Job Market,” Burning Glass Technologies.
2. PWC, 2017, “Investing in America’s Data Science and Analytics Talent –The Case for Action,” PwC and BHEF, https://www.bhef.com/sites/default/files/bhef_2017_investing_in_dsa.pdf.
3. Radovilsky, Z., Hegde, V., Acharya, A., Uma, U., 2018, “Skills Requirements of Business Data Analytics and Data Science Jobs: A Comparative Analysis,” Journal of Supply Chain and Operations Management, Vol. 16, No. 1.
4. Seal, K.A, Leon, L.A., Przasnyski, Z.H., Lontok, G., 2020, “Delivering Business Analytics Competencies and Skills: A Supply-Side Assessment,” INFORMS Journal on Applied Analytics, Vol. 50, No. 4, pp. 239-254, https://doi.org/10.1287/inte.2020.1043.
5. De Veaux, R.D., Agarwal, M., Averet, M. Baumer, B.S., et al., 2017, “Curriculum Guidelines for Undergraduate Programs in Data Science, The Annual Review of Statistics and Its Application,” https://www.annualreviews.org/doi/abs/10.1146/annurev-statistics-060116-053930.
6. Donoho, D., 2015, “50 Years of Data-Science,” based on a presentation at the Tukey Centennial workshop, Princeton, N.J., Sept 18.
7. Mills, R., Chudoba, K., Olsen, D., 2016, “IS Programs Responding to Industry Demands for Data Scientists: A Comparison Between 2011-2016,” Journal of Information Systems Education, Vol. 27, No. 2 (Spring).
8. Cegielski, C.G. and Jones-Farmer, L.A., 2016, “Knowledge, Skills and Abilities for Entry-Level Business Analytics Positions: A Multi-Method Study,” Decision Science Journal of Innovative Education, Vol. 14, No. 11, pp. 91-118.
9. Stanton, W.W., Stanton, A.D’A., 2020, “Helping Business Students Acquire the Skills Needed for a Career in Analytics: A Comprehensive Industry Assessment of Entry-Level Requirements,” Decision Sciences Journal of Innovative Education, Vol. 18, No. 1.
10. Parks, R., Ceccucci, W., McCarthy, R., 2018, “Harnessing Business Analytics: Analyzing Data Analytics Programs in U.S. Business Schools,” Information Systems Education Journal, Vol 16, No. 3.
11. Kang, J.W., Holden, E.P., Yu, Q., 2015, “Pillars of Analytics Applied in MS Degree in Information Sciences and Technologies,” Association for Computing Machinery’s Special Interest Group for Information Technology Education, Chicago.
12. Verma, A., Yuruva, Y.V., 2019, “An Investigation of Skill Requirements for Business and Data Analytics Positions: A Content Analysis of Job Advertisements,” The Journal of Education for Business, Vol. 94, No. 4, pp. 243-250.
13. Delen, D. and Ram, S., 2018, “Research Challenges and Opportunities in Business Analytics,” Journal of Business Analytics, Vol. 1, No. 1, pp. 2-12.
14. Foster, S., 2019, “Data Science Within Supply Chain Management: An Analysis of Skillset Relevance,” Doctoral dissertation, Capella University.
15. Fayyad, U. and Hamutcu, H., 2020, “Toward Foundations for Data Science and Analytics: A Knowledge Framework for Professional Standards,” Harvard Data Science Review, June 30.
16. Davenport, T.H., 2018, “From Analytics to Artificial Intelligence,” Journal of Business Analytics, Vol. 1, No. 2, pp. 73-80.
17. Mehrabad, M.S. and Brojeny, M.F., 2007, “The Development of an Expert System for Effective Selection and Appointment of the Job Applicants in Human Resource Management,” Computers & Industrial Engineering, Vol. 53, pp. 306-312.