Most large organisations now have considerably more HR data than they can use effectively. They have engagement survey data, performance review data, learning and development completion data, recruitment funnel data, turnover data, pay equity data, and a growing accumulation of data from the digital traces that employees leave in their use of collaboration tools, communication platforms, and HR information systems. The challenge is not primarily one of data availability. It is one of analytical capability and analytical purpose: the capability to transform the data into insights that genuinely inform decisions, and the clarity about what decisions the analytics are designed to inform.
The gap between data availability and useful insight is significant in most HR functions, and its causes are worth understanding precisely because they are not primarily technical. They are organisational and conceptual: the absence of clear questions that the analytics are designed to answer, the disconnection between the analytical outputs and the decision processes that should be informed by them, and the insufficient analytical capability in most HR functions to move from descriptive reporting to the predictive and prescriptive analytics that produce genuine strategic value.
The organisations that are extracting genuine strategic value from HR analytics share several features that distinguish them from the majority that are primarily producing descriptive dashboards of historical HR activity. Understanding those features is the most direct available path to improving the return on HR analytics investment.
The three levels of HR analytics and why most organisations are stuck at the first
HR analytics operates at three distinct levels that produce qualitatively different outputs and require qualitatively different capabilities to execute well. The first level is descriptive: what happened, expressed in terms of the historical patterns in HR data. Turnover rates by function, tenure, and demographic. Engagement scores by manager, team, and business unit. Recruitment conversion rates by source and role type. Time to fill by function and level. These descriptive analytics are the dominant output of most HR analytics functions and they have genuine value as operational monitoring tools. They tell the organisation what has happened. They do not tell it why it happened, what is likely to happen next, or what it should do differently.
The second level is diagnostic: why did it happen, expressed in terms of the causal analysis of the patterns identified in descriptive analytics. Why is turnover elevated in a specific function? What specifically distinguishes the high-engagement teams from the low-engagement ones? What is the relationship between specific management practices and specific performance outcomes? Diagnostic analytics require the analytical capability to move from correlation to causation, which requires statistical sophistication and the ability to design the analyses that can distinguish between explanatory variables that are causally related to the outcomes of interest and those that are merely associated with them. Most HR functions have insufficient analytical capability to produce genuinely diagnostic insights, which is one of the primary reasons that the investment in descriptive data collection rarely translates into the organisational improvements that the data would support.
The third level is predictive and prescriptive: what is likely to happen and what should we do about it. Which employees are most likely to leave in the next six months, and what specifically can the organisation do to retain them? Which candidates are most likely to succeed in specific roles, based on the evidence from the organisation’s own performance data rather than on generic selection criteria? Which leadership development investments are most likely to produce the specific capability improvements that the strategy requires? These analytics require not only strong analytical capability but also the integration of HR data with business performance data in ways that most HR functions have not yet achieved, and the partnership with business leaders that ensures the predictive insights are connected to the decisions they are designed to inform.
The three questions HR analytics should be designed to answer
The most common failure in HR analytics strategy is the absence of clear questions that the analytics are designed to answer. HR functions that invest in analytics infrastructure without first defining the specific decisions they want to improve are producing capability without purpose, which generates dashboards that are viewed without being acted on and reports that are received without producing decisions.
The three questions that most consistently produce genuine strategic value from HR analytics are: first, where are the human capital conditions that are most significantly constraining business performance, and what specifically is producing them? Second, where are the human capital risks that are most material to the organisation’s strategy, and what is their current trajectory? Third, what specific HR investments would produce the most significant improvement in the business outcomes that matter most, and what is the evidence base for that assessment?
These questions are more demanding than the questions that most HR analytics functions are currently designed to answer, but they are the questions that most directly connect HR analytical capability to the strategic value that justifies the investment in developing it. They require the integration of HR data with business performance data, the statistical capability to conduct genuine causal analysis rather than only descriptive reporting, and the business acumen to translate analytical findings into the language of strategic impact that senior business leaders can evaluate and act on.
Building the capability and the culture for genuine insight
The development of genuine HR analytics capability requires investment in both technical capability and analytical culture that most HR functions are currently underinvesting in. The technical capability investment includes the development of statistical analysis skills in the HR team, the integration of HR data systems with business performance data in ways that allow genuine multi-variable analysis, and the development of data visualisation capabilities that allow complex analytical findings to be communicated with the clarity and accessibility that senior business audiences require.
The analytical culture investment is less commonly discussed and equally important. Analytics capability produces strategic value only when the findings it generates are genuinely integrated into the decision processes of the organisation. This integration requires business leaders who are willing to use analytical evidence in their decisions about people and who have developed the literacy to evaluate the quality of the evidence being presented to them. It requires HR practitioners who have the consultative skills to connect analytical findings to the specific business questions that leaders are trying to answer. And it requires the organisational norm that decisions about people strategy are made with evidence rather than with intuition and experience alone, which is a cultural shift that requires sustained modelling from the most senior levels of the organisation to produce.
The partnership model that makes analytics strategically valuable
The single most important organisational condition for HR analytics to produce genuine strategic value is the quality of the partnership between the analytics function and the business leaders whose decisions the analytics are designed to improve. Analytics produced without a clear business question behind them will not be used in the ways that produce strategic value regardless of their technical quality. The analytics that change decisions are those built from a genuine joint definition of the specific decision being made, the specific evidence that would improve it, and the specific format in which that evidence would be most useful to the decision-maker. The HR analytics practitioner who can move between rigorous quantitative analysis and the kind of strategic business conversation in which that analysis is most valuable is providing a service that most HR analytics functions are not yet configured to deliver. Developing this capability is the most important single investment available to HR functions that are serious about analytics as a source of genuine strategic value rather than as a reporting function with better visualisation tools.
The organisations that have made the transition from descriptive to predictive analytics in HR are consistent in identifying one condition as prerequisite for the transition: the presence of a senior HR leader who both understands analytics and is willing to hold the function to the higher standard of evidence that predictive and prescriptive analytics require. Without this leadership, the pressure to produce the comfortable certainty of descriptive dashboards consistently defeats the investment in the more demanding and more valuable work of genuine analytical insight. With it, the HR analytics function develops progressively toward the level of strategic usefulness that the organisation’s investment in it was designed to produce.
HR analytics that produces dashboards of historical HR activity is not analytics. It is reporting. The distinction matters because the investment required to produce genuine analytical insight is substantially greater than the investment required to automate historical reporting, and the return on that greater investment is only realised when the insights produced are genuinely connected to the decisions they are designed to improve.