People Analytics ‘3.0’
In my journeys as a Human Capital Consultant, I have found that there are two objectives in the delivery of new analytics capabilities in HR: the data science behind creating predictive models and the process of applying them. They are two different skill sets as I see it.
It’s great to have a working A.I. model, but it is not very useful if it does not influence action. This question is not unique to my own thinking. Many specialists in the field are asking how do we get ‘there’ from ‘here’? The book ‘Big Data @ Work’ by Tom Davenport calls the evolution ‘Analytics 3.0’, where clearly informed practitioners use ‘prescriptive’ technologies to guide optimal action. Or we can look at Josh Bersin’s HR analytics maturity model, which provides an overview of four levels of analytics capability: Reactive, Proactive, Strategic & Predictive.
Unless we are shopping on Amazon or using our GPS to navigate traffic, this reality is a long way off for the average HR specialist. To be successful, we need to bring our analytics capability down to the ground level and make reporting insights more intuitive for the end user, i.e. the manager or HR specialist. This would help to break down consultative barriers with our clients as the discussion turns from data complexity, accuracy & relevancy to how our clients can impact business performance through improved talent management.
How HR teams can evolve from prediction to prescription?
Over the years, I have had the opportunity to see how organizations are putting into practice machine learning models to guide manager decisions.. These are 5 thought on how to to rise above data management mode and add greater value to our clients in the organization.
1) Re-inforce HR & CIO collaboration to develop tooling: Talent management programs, such as high performer recognition, retention, and performance management should be integrated with other HR portals used by managers. The days of walking around with spreadsheets to track talent programs are long over! The capability to gather data at the manager level will significantly improve our visibility into actions being taken. Rather than build these tools ourselves, we should work with CIO specialists to integrate HCM functionality with user need.
2) Simplify & standardize reporting – As many organizations develop separate ‘workforce analytics’ departments, part of the task of this organization should be to develop and manage reporting and operations. These teams should develop reporting templates in collaboration with their clients to deliver the most insightful info without going too deep. ‘Real time’ information needs to be accessible on demand via a manager reporting portal. It should be simple to understand, accurate, and trusted. The ultimate purpose is to support manager actions!
3) Build relationships outside of HR – At times, HR professionals struggle to build relationships across the business due to the fact that we are caught in our own HR loop. As a result, we are not trusted with deeper access into decision making at the executive level. Integrating HR metrics and reporting with other functions with mature reporting capability, such as Finance, Marketing or Operations should be a priority.
4) Break the mold – in addition to delivering talent programs more effectively, we also seek to add greater business value through analytical insight. For instance, we are proposing research into the reasons for high performing employee attrition, and what might be done about it. Instinct tells us that the answer might not be in areas outside the traditional scope of HR, such as employee satisfaction with operational processes. These insights could help drive changes outside the HR function and build our perceived value across the employee ecosystem.
5) Build consulting capability – One of the challenges both Bersin & Davenport point out in helping organizations develop deeper analytics capability is the skill set required to build and manage new analytics methods, which may not come naturally to many HR specialists. It is not always about having the ‘right’ answer, but more about how we are delivering value to our clients. In many people analytics projects I’m seeing, the emphasis is on the A.I. and not as much on how we are building culture around data based decision making for managers.
Question: what steps are you taking to build greater analytics capability and deliver value to your clients?
