Does HR Need Real Data Science?
Data Science is the defined as:
.... an inter-disciplinary field that uses scientific methods, processes, algorithms and systems to extract knowledge and insights from many structural and unstructured data. Data science is related to data mining and big data.
Does HR really need data science or data scientists to uncover insights in people data? The answer doesn't need to be complicated.
Strictly speaking many companies are too small to use their people data for serious research. Getting insights, yes. Testing some theories about people and business performance, yes. But serious Data Science, Artificial Intelligence, Machine Learning, big data sets and research? Probably not.
That's not to say that there are concepts of data science in HR Analytics that drive the need for research. That is most certainly true of most small and medium size employers. Whether testing a new communications strategy, a benefits program change or researching the best source of candidates. The science part of the method of research and testing are most certainly applicable. The parts that I challenge is the use of large structured AND unstructured datasets, machine learning and organic AI.
Most of the data that these organizations will use are smaller, structured datasets that come from self-generated systems: ATS, Benefits, Payroll, HRIT, and Performance Management. Their intersections are normally stored in one database so there is no need for unnecessary nor complex transformations. Also, the tools used are most typically spreadsheets with pivot tables and some simple visualizations.
Many of these organizations are struggling to get their heads wrapped around the concept of Analytics. Their version of using People Analytics are relegated to report-generation and statistical manipulation in spreadsheets. That is typically satisfactory, or at least for the short term. Machine learning and artificial intelligence seem like overkill, unless part of a comprehensive strategy that allows the building of connections to the other business systems that can provide important facts to their development.
In order to break into the small and mid-size market for a People Analytics data science practice, the business would need to be very people centric, rather than capital centric and require intimate insight into worker behaviors, skills, and expected outcomes. Standard Operating Procedures would have to be well studied, documented and measured consistently.
Insights You Need to Get It Right
