People Analytics - what's the big deal?
Technology has fundamentally changed the way we gather and analyze people data

People Analytics - what's the big deal?

Someone asked me recently why they are seeing so much more hype and publishing around the subject of People Analytics. I thought it was a fair question. It's not like we only recently discovered we could analyze data about people. Psychologists and Sociologists have been doing that already for a long time. I've personally been analyzing data about people for at least the past 15 years.

So I understood their question to mean 'What's new?' What has changed that has suddenly propelled this topic onto the radar screens of CEOs, CHROs, academics, journalists and others who just want to learn more about it? Why are organizations suddenly setting up and growing dedicated People Analytics groups out of their limited scatterings of erstwhile workforce analysts?

Some people have suggested that organizations have reached a tipping point in volume where they can now regard their people data as 'big data', and that this now explains the fuss. I disagree. As Peter Capelli points out in a recent HBR article, it's not really about 'big data'. It's erroneous to think that insight automatically follows from volume. Richness and variety of data is equally important, and structure is the key to almost all effective analytics. When analyzing any form of data, there is always a tension between, quantity, variability and dimensionality. Quantity in and of itself never wins out.

So what is it about? In my opinion the answer lies in technology - in the massive levels of digital growth and innovation that we have experienced in the past 5 years. Technology is giving us new ways of looking at the world around us and challenging accepted wisdoms like never before. As a few people have recently pointed out, 20 years ago we were told never to get in a car with a stranger and never to meet people on the internet. Now we are literally summoning strangers from the internet to get in their cars.

In the same way, technology is giving us ways to understand and analyze people that I would never have dreamed possible when I started out as a young psychometrician in the early noughties.

As is my wont, let me put some structure to this. In the past 5 years, technological development has offered up exciting opportunities to enrich the digital exhaust, obtain new people data to analyze and to analyze people data in new ways. Here are some examples of what I mean.

1. A richer digital exhaust

Many organizations are experiencing a better quality and richer digital exhaust in recent years. The quality and richness of the data imprint of members and employees is vastly superior to the situation even 5 years ago. This is due to:

  • Improved data systems with greater levels of flexibility to capture information in both structured and unstructured form.
  • A greater variety of database structures, ranging from traditional relational databases like Oracle and SQL through to graph databases allowing highly flexible data capture and query, like PoolParty and Neo4j.
  • Improved document digitization and parsing, allowing the text within documents which were previously out of scope for the everyday analyst to now become a part of their dataset.
  • More integrated data across fewer systems, making it easier to connect data related to individuals and groups and analyze a broader set of indicators.
  • More pulse data, with more organizations 'sampling' the performance and attitudes of their members on a more frequent basis. For example, Bridgewater ask their employees to complete a feedback rating on all attendees of meetings above a certain size after every meeting.

2. New technologies to capture people data

The past 5 years has seen the realization by organizations and entrepreneurs that people data is a valuable asset that can be harvested and harnessed by new technology. This has led to exciting innovations both in how to capture this data and in how to offer it to potential consumers. Here are some current examples of this.

  • Talent aggregation is the concept that an organization-independent repository of biographical data about people can be built and utilized for a number of purposes. Primarily and most obviously this can be used to identify potential recruits, but it also offers the possibility of broader insights around organizational trends, retention and attrition, culture and many other possibilities. LinkedIn was the first mover in this space, but many others and now battling it out in this arena, including Entelo, Hired and Piazza. This raises some fascinating and crucial questions about access to and ownership of people data - a debate which has already made its way into courtrooms.
  • Digital technology allows new ways to gather data on people's knowledge, abilities and characteristics. Innovative companies like Knack and Arctic Shores design game-like tasks which collect data on the decisions people make and the ways they approach problems as they play them. HireVue, among others, is digitizing the spoken word in video recorded interviews and using machine learning techniques to try to identify the 'language of success'. Although there are varying standards of scientific rigor among the many players in this space currently, the common motivator behind all this innovation is the richness of new data that can potentially be harvested in this way.

3. Analyzing people data in new ways

All this rich exhaust and new data wouldn't have been much use to us 5 years ago. The word 'data scientist' was just coined, and the majority of analysts were limited by the size of their processor or hard drive and the UI limitations of Microsoft Excel or SPSS. Now a whole host of technology has entered the fray which offer up a wealth of capability.

  • Open source statistical programming software has made he jump from academia to the world of enterprise. R and Python, for example, are being adopted more and more to analyze people data. This kind of software opens up the possibilities for what can be done. It automatically understands different data types, from numbers to text strings. It can process data quickly and more efficiently, reducing time and screen freezes. It offers up multitudes of pre-programmed packages for working on people data. And, as open source software, its easy to gain access to learning or advice through resources like Datacamp and Stackexchange.
  • Cloud based computing is very accessible, allowing tasks that have a high processing demand, such as text analytics, to be done effectively and quickly. So much people data is formed of unstructured text, and the ability to process this text allows possibilities that simply were not there previously, such as sentiment and emotion analysis, topic modelling, predictive analytics and artificial intelligence.
  • Visualization possibilities are vastly greater today. Facilitated by the adoption of open source technology, we can view people data in multiple dimensions, in time series, using network graphing and other diagrams that were not so easily accessible previously. We can create dynamic charts that change and update on the fly. All of this is critical in facilitating a stronger understanding of complex people phenomena.

We are in the midst of a genuine paradigm shift in how we gather and analyze people data, and it's the technology that has taken us here. Even though many are still trying to wrap their heads around such a rapidly moving space, there is no doubt that it's a big deal, and it will continue to be for some time.

I lead McKinsey's internal People Analytics and Measurement function. Originally I was a Pure Mathematician, then I became a Psychometrician. I am passionate about applying the rigor of both those disciplines to complex people questions. I'm also a coding geek and a massive fan of Japanese RPGs.

You can message me on LinkedIn or engage with our People Analytics group within McKinsey's Organization Practice.

All opinions expressed are my own and not to be associated with my employer or any other organization I am associated with.


Vikki Williams

Leadership | Transforming Customer and Colleague Experience | People, Process and Technology | Transformational Change | Digital Transformation | Continuous Improvement |Trustee

7 年

Really interesting Keith - thank you for sharing.

回复
Yudong Cao

Senior Manager, Data Science at Capital One

7 年

Keith, I'm really amazed by this article. I appreciate the opportunity to be enlightened! Coming from a Mathematics (Bachelor's) and Psychometrics/QuantPsych (Master's) background, I have always found People Analytics the most practical application combining the theoretical and technical aspects of both realms. While working as a data analyst trying to find paths towards becoming a data scientist, I wonder if you could offer some career advice for fresh newgrads with similar background like you wishing to pursue a career in people analytics? What I find frustrating in my job search is that most DA & DS are emphasizing too much on all sorts of machine learning modeling techniques, while the theoretical Psychometrical modelings I learned from my master's program, like Factor Analysis, Structural Equation Modeling, IRT, etc. are rarely mentioned and seemingly not even cared by the mainstream analytics professionals. I'd love to hear your thoughts and insights on how Psyshometricians are uniquely different from Statisticians/Data Scientists, and in terms of People Analytics, what psychometrical methods are the most vital and useful, just so professionals who had relevant academic training could stand out from others. I greatly appreciate your attention and reply!

回复
Sam Stone

VP of Product & Design, EvenUp

7 年

Great post. One other factor I would add that's fueling the growth of this field - predictive analytics has become a competitive necessity in other functional areas (sales, marketing, etc) and I think that's generating C-level "pull" for predictive workforce analytics, in addition to all the spot-on "push" reasons you mention.

Greg Newman

People Analytics Nerd & AI for HR @ Deloitte Australia

7 年

Hey Keith great article as always. I would add one extra vector that I think is really important to the growth of people analytics and that is an increasing maturity from the business and from HR people and an appetite to make data-driven decisions. I think a new generation of leaders are evolving who want to make people decisions like they make financial decisions based on sound numbers and data and it's that changing attitude which is fueling the demand for people analytics. but I also agree the growth in technology to answer those questions is also very important.

Joakim Lorentz

Tech lead for hire

7 年

Great summary! I'd like to give even more credit to visualization tools that actually allows non-techies/statisticians to analyze and understand the figures in a much clearer way (not to mention the mobile dashboards that even allows us to do this on the go between meetings, where a lot of managers spend their time)!

回复

要查看或添加评论,请登录

社区洞察