Talent science is the business capability of using advanced data analysis techniques and predictive models to drive HCM decision making. It calls for a logical connection between decisions about people, HR program investments and strategic business outcomes. This is true along all levels of Talent Science, from predictions at the individual level in operational reports (e.g., flight risk per employee) to process outcomes (e.g., operational workforce planning forecasts) to dashboards (e.g., prediction of business results based on projected attrition or employee engagement levels) or the definition and validation of new predictive models in qualitative analyses. The advanced techniques leveraged in talent science have the highest potential business impact.
Vendors also use the terms “workforce analytics,” “talent analytics” or “talent management” for a wide array of reporting and analytical capabilities leveraging HR data. By leveraging Gartner’s model defining Talent Science, HR and IT professionals can cut through the hype and select the most meaningful technologies per their own system landscape and analytical requirements.
In the case of talent science, we refer to technologies that enable business leaders and HR professionals to view and analyze statistical insights about their workforce. Talent Science span a range of analytical capability:
Typical operational reports are lists with, in some cases, very elementary summary statistics such as counts of heads, sums of full-time equivalent (FTE) numbers or average durations. HR professionals tend to be primary users and leverage standard reports as delivered in HR applications, ranging from core HR and payroll systems to specialized systems on key talent processes. These reports are used to follow up and act on individual cases. These reports or individual-level analytics can range from descriptive analytics (FTE) to predictive (individual flight risk, chance of being high potential, fit to role, etc.).
Many HR technology providers propose process metrics and dashboards as part of their basic offering. These include visualizations on volumes, completion rates, and process outcomes. These processes can range from purely administrative payroll or HR inquiry processes to more strategic talent processes such as leadership assessment and employee engagement measurement. As with operational reports, process analytics can range from being descriptive (completion rate) to being prescriptive (e.g., resource requirements at a given point in time per anticipated activity levels).
Many organizations put in place scorecards displaying results on key performance indicators selected to track the progress and impact of strategic initiatives that are part of the ongoing operating plan. This requires the ability to combine the output of multiple metrics on a single screen or page. Some organizations will include HR metrics along with business metrics sourced from other systems in the same dashboards. Metrics can be basic descriptive measures such as the recruiting metric “average days to fill” or measures like “annualized attrition rates,” which project anticipated future results.
Advanced analysis is generally conducted by specialist Talent Science professionals to evaluate and suggest strategic options based on workforce data. They leverage tools that enable complex analysis, segmentation and statistical assessments of data points across end-to-end HR and business processes to conduct diagnostic analysis, or set up and test predictive models. Qualitative analysis starts with the formulation of a workforce-related business question, followed by a hypothesis and the testing of that hypothesis. The impact of action plans may be simulated through statistical modeling to aid in final workforce-related investment decisions.
Aligning data models and calculation methods across layers of analysis, ranging from strategic analysis to operational individual or team level analysis, greatly increases the ability to cascade and track HR program execution. Basic workforce reporting and simple dashboards are now widely leveraged, but many organizations struggle to make the most of their investments and move forward with more strategic usage of Talent Science. Talent Science functionality is improving with the introduction of new technologies around big data and predictive analytics, especially when it comes to embedding analytics within human capital management (HCM) applications.
Talent Science Is Bridging the Gap
Talent science is simple. It tackles the historical challenges inherent in investing in the human side of business by extracting an ROI using medium data.
Data doesn’t always have to be used in a ‘big’ way to be effective. The new realm of talent science leverages cloud-based infrastructure and increases in processing power to gobble metrics and provide real-time talent management metrics to large organizations.
With talent science you can extract and apply macro and micro metrics to manage every part of your employee lifecycle including pre-hire, development, succession planning and employee engagement.
Along with a plethora of startups looking to match candidates to the right company, Google has taken medium data to the next level to debunk the efficacy of their own hiring practices. Their longitudinal study revealed no correlation between GPA and success on the job.
A recent report in the Harvard Business Review by researchers at the University of Minnesota has shown that using even the simplest of algorithms in the hiring process can improve the rate of success on the job by 25 percent.
Medium data will often trump our instincts. This is not to say that we shouldn’t use our guts in the hiring process, but we must be prepared to make more informed decisions. The alternative is to fall victim to repeating the same mistakes we’ve always made and that we are blind to correct.
Humans are awfully complex, but understanding everyone on an individual level allows managers to tailor their management style to best fit their direct report.
Historically, managers were provided some broad training and left to their own devices. Talent science and the rise of medium data can provide them with actionable insights into who may be at risk of leaving and what the manager can and should do about it.
One of the best studies they’ve done looked at what makes their managers successful. While the results were not terribly earth shattering, it provided an irrefutable framework against which to look at promotions and succession planning. The best and the brightest contributors were no longer the only ones being taken away from what they do best in order to do something they’ve never done, manage people.
The traditional “measure and hope” model of engagement is broken. Being able to overlay disparate pieces of data from the psychological to the demographic allows organizations to go well beyond simply understanding which pockets of the company need attention.
The broader and more intelligent use of data allows organizations to customize action plans based on the inner workings of groups and the insights gleaned from what drives engagement elsewhere.
It’s Time for Change
Even the smallest pieces of data are completely changing the game on the talent side. No longer are we solely reliant on the subjectivity and whims of our middle managers to determine success within the organization.
Ultimately, the most powerful component of data is that it strips away the subjective. And, if there is one thing that is consistently over-used, it’s the subjective process when it comes to hiring, promoting and engaging people.
The irony is that data is used to inform nearly every decision in Corporate America these days with the exception of those that, arguably, matter most.
We are living in the Dark Ages when it comes to people decisions. Those who start this process now will be dancing on the graves of their competitors by the time the data Renaissance hits in full force.