Can we face the truth? HR doesn’t have Big Data. Now what?

Every time I hear a presentation from a Silicon Valley based human resources organization about its Big Data analysis of the workforce, I remember Peter Cappelli’s article in the Harvard Business Review, “There’s No Such Thing as Big Data in HR.”

Most organizations, Cappelli observes, have thousands of employees, not millions. And the documented performance for many of those employees occurs only annually. Google has the size to conduct “Big Data” on different sub groups of its workforce. Companies of a thousand employees are too small to generate reliable empirical research on sub-groups of their employees.

Even if your company has the size to produce empirical analysis of the workforce, the problem for most human resources departments is that the needed data is stored in separate data warehouses that can’t combine data.  For example, the databases associated with recruiting is housed separately from the databases associated with performance management, which is housed separately from the databases associated with company performance. That was always my struggle when trying to gather data to study how to identify which qualities make the best biomedical engineer or salesperson.

Unless your databases can be bridged and made compatible, it becomes difficult to conduct research. (However, you can always do the old-fashioned approach: become a spreadsheet jockey.) More than data scientists, human resources departments need data managers to clean up the data.  Better yet, they need human resources information systems that can retrieve and makes sense of all of the information they generate.

Cappelli’s most compelling point, perhaps, is that the questions many of the Big Data studies are trying to uncover have already been investigated by academics for decades. As an example, he refers to Google’s Project Oxygen, a multi-year research project designed to identify the qualities of a good manager. Most of the conclusions of Google’s research are the same as decades-old academic studies.

Here is another example. Today, there is a lot of interest in “unconscious bias,” which is the unconscious discrimination and incorrect judgments that occur during stereotyping. Neuroscience has uncovered that it is not just the bigot and sexist who has biases and discriminates. It is all of us. We all have “unconscious biases” that affect our judgment. We all, or at least most of us, unconsciously want to stay in that safe zone where we work, socialize with and with whom we make decisions.

Research on unconscious bias in performance management goes back almost 100 years to Edward Thorndike, who in 1920 published a paper called, “The Constant Error in Psychological Ratings.”  Thorndike discovered that military leaders generalized a soldier’s physique to other qualities such as intelligence. If he looks strong, he must be smart.

A few weeks ago, I listened to a member of Google’s human resources team share the company’s Big Data analysis on successful work team.  The leading quality of successful Google teams is that they provide psychological safety. That is, in simpler terms, they have empathy and build trust. My compliments to Google for doing this research and for publicly sharing the findings. Sitting in the audience, however, I smiled and thought of Peter Cappelli’s article: this question has already been answered by academic research.

We may receive more insights on the best qualities of engineers, managers and salespeople, and where to recruit them, from Artificial Intelligence. AI can search many large external databases—but this work is just beginning in HR.

Do you have a major question about how to improve job candidate selection, reduce turnover, or figure out the qualities of a good manager? Check out the academic research—and watch for discoveries from AI.

Victor Assad is the CEO of Victor Assad Strategic Human Resources Consulting and is a Managing Partner of InnovationOne. He consults on innovation, global talent strategies, developing agile leaders and teams, and other strategic initiatives. Questions? Please email Victor at Visit for valuable free reports. For innovation visit

1 comment

  1. Nice blog and reinforces what I mentioned when we got together at lunch…Google’s research didn’t come up with anything new. But since Google did the research, it is now a “bright, shiny object” has gotten lots of kudos. 😊

Leave a Reply

%d bloggers like this: