Receive daily articles & headlines each day in your inbox with your free ERE Daily Subscription.

Not logged in. [log in or register]

What’s Wrong With HR Metrics? Pretty Much Everything!

by Apr 16, 2012, 6:49 am ET

Or why HR metrics need to focus on helping managers to improve their people management decision-making

For at least the last decade, HR departments around the world have been pouring tons of time and money into developing HR metrics. Unfortunately, that effort has largely led to continued levels of frustration and, at best, a large number of what I call “so-what” metrics with little strategic impact. It doesn’t matter whether your HR metrics were provided as part of the software that you purchased or if they came from a major HR consulting firm; the results have been the same: dismal at best. After three plus decades of thought leadership and research in HR metrics, I’ve concluded that the current approach is an abject failure and that HR simply can’t continue on this current painful path.

The time has come to completely disregard today’s approach and to look to other functions that have had significantly better luck influencing executives with their metrics (i.e. customer service, supply chain, branding, and finance, to name a few). Even if you are currently happy with your metrics, this article should provide you with sufficient reasons as to why you should rethink your approach and to shift toward what I call “people management decision-making metrics,” a far superior approach that focuses on helping managers improve their people-management decision-making.

The Top 20 Major Faults With Most HR Metric Approaches

It is possible to avoid metrics altogether. For example, you can find out how well you’re doing in HR simply by counting your budget increases, your new headcount increases, or tracking how often you are mentioned in the annual report. However, if your goal is to improve the quality of your people management decisions and be metric driven, you need to assess your current metric approach using the following 20 faults as assessment factors.

Past vs. Forecasting Faults

Reporting what happened “yesterday” has little value – Almost all HR metrics report history, because they tell you what happened last quarter or even last year. Learning that your turnover rate was 6% last year has little value in a fast-changing world where your turnover rate could double next quarter. The best metrics cover your current time period.

We don’t provide forward-looking metrics or predictive analytics – Even today’s metrics may be of less value than future forecasts. This is because every decision-maker really wants to know what problems or opportunities will occur over the next week or months so they can work to prevent them. However, HR simply doesn’t provide these forward-looking predictors, which are known as predictive metrics or analytics. As a result, decision-makers are forced to quickly react to suddenly occurring people-management problems because they weren’t given a chance to prevent or mitigate them as a result of forecasts. HR needs to implement “predictive HR” (which follows the predictive policing model), where analytics tell HR and managers where they can most likely expect today’s and tomorrow’s people-management problems. 

Business Impact Faults

We don’t provide “dollar and goal impacts” in our metrics – Long before there was an HR function, executives and managers have been in love with dollars. At least outside of HR, everyone knows that a problem or opportunity won’t be considered by executives as critical unless it has a direct dollar impact on corporate goals (i.e. revenues, sales, or profit, all of which are measured in dollars). HR does report less-important cost metrics (i.e. cost of a hire), but cost metrics do not impact topline growth (i.e. revenue growth), the single metric that CEOs care most about.

HR needs to calculate its direct dollar impact on each major corporate goal like revenue, but also customer service, innovation, and quality. Although almost everyone in HR assumes that this can’t be done, other “soft” business functions like branding and customer service have successfully worked with the CFO’s office to translate their standard “what-happened” metrics so that they reveal their dollar impact on revenue and business goals (I call them “business case metrics”). You can, for example, report that your employee turnover has increased 12% until you’re blue in the face and no one will pay much attention. But the minute you translate it into dollars, “losing key staff reduced revenue by $17.2 million” everyone will demand a solution immediately. Continually providing business case metrics will not only increase HR’s credibility but will also increase your funding.

HR doesn’t calculate the dollar cost of doing nothing and excessive cost-cutting – Weak metrics and delayed reporting together can result in simple people-management problems exploding into major catastrophes, because no action was taken. One of the best ways to speed up decisions and action is to calculate and report the cost of slow decision-making or doing nothing. Just like a cancer in the body, delayed decision-making can turn solvable problems into catastrophes. In the same light, HR must reduce uninformed HR cost-cutting decisions by calculating and reporting the dollar impact of excessive cost-cutting in people-management programs (e.g. reducing safety training may save a few dollars in the short term but the long-term increase in accident costs and insurance make the initial savings insignificant).

We don’t calculate risk metrics – One of the hottest areas in business is calculating metrics for major business risks. Unfortunately, most HR departments simply don’t have a capability in the risk area. I’m not referring to potential legal risks. Instead, we need to calculate the major risks involved in more impactful problems like key employee turnover, weak hiring, poor employee development, a lack of leaders, and the tremendous cost associated with retaining bad-managers.

The most common business metric, ROI, isn’t even calculated -- The most important and commonly calculated metric in any business function or program is its ROI, which is simply the ratio of its cost compared to the dollar of return from that expenditure. Unfortunately, most HR functions don’t even calculate their ROI. The formula for this primary HR metric (a.k.a. workforce productivity) is the ratio of corporate profits compared to total labor and HR costs combined. Not only should this powerful ratio improve each year but it should be superior to the workforce productivity results produced by your competitors. 

Decision-making and Action-related Faults

HR metrics do not drive action – One of the primary reasons for developing metrics is to drive actions that result in continuous improvement. In fact, I classify HR metrics into two types, 1) “so-what metrics” (i.e. that’s interesting) and 2) “action metrics,” which result in someone taking an action after reading them. If HR metrics don’t actually change behavior, why do we have them? It’s quite easy to determine if your metrics drive action. Simply distribute your metrics in the normal way and then interview your users a month later and ask them simply “what major actions did you take as a result of these metrics?” If they didn’t act or change their decision-making (which is the normal answer), you have failed.

HR metrics are not designed for decision-making – To further complicate the metric problem, when HR reports its metrics, they are not even provided in a decision-making format. The metrics from HR are not accompanied by information that can drive decision-making (i.e. “what decision must be made?; what is the cost of doing nothing?; who must make it?; by when?; What is the recommended course of action?). This fundamental omission is almost unexplainable. If you expect professionals to make decisions, you must provide decision guidance tools that facilitate immediate and accurate decision-making.

The absence of “why” in our metrics makes taking action difficult – Almost all HR metrics focus on telling you “what happened” (i.e. turnover increased 2%, time to fill decreased 1%). And although the raw number revealing what happened is obviously important, in order to fix problems, metrics must also reveal the root causes of the problem (what are the causes? why did this happen?). As result, even if executives or managers are alarmed by your “what happened metrics,” they can’t take accurate action immediately because they have to wait to learn precisely why these changes have occurred. And in many cases, HR doesn’t gather metrics covering why things happen.

HR metrics are not provided in real time – Even forecasted problems and opportunities can arrive almost overnight, without much warning. Rather than being part of the steady predictable trend, these problems and opportunities seem to come out of nowhere. In a chaotic world, trend lines and historical data by themselves have a limited value. Instead, what is needed is real-time or virtual monitoring data on what is happening now. Once again, HR fails because it seldom can even accurately report a simple metric, like headcount, in real time. If it takes a week or month to gather data, in many cases, the wait will cause the decision to be “made for you” as a result of the passage of time. Real-time metrics might include today’s workforce productivity, this week’s turnover, this week’s engagement scores, or the relative availability of top talent in the market today.

Our metrics are not directly available to line managers – Let’s face it: most HR metrics are only designed for the use of HR professionals, and these metrics are also only provided to HR. HR professionals do need these metrics because we provide advice, but the individuals who actually make the daily people-management decisions (line managers) also need direct access to HR metrics. Although there are few exceptions, line managers often must consult with their HR professional in order to get the metrics they need for decision-making. Instead, what we must do is to provide real-time people management metrics and data, in a decision-making format, directly to line managers. The data must be provided to them in a user-friendly format and on their smart phone or tablet computer, so that they can make most routine people-management decisions independently but accurately.

HR metrics are not embedded in standard business reports — Even valuable metrics that are placed in a standalone HR report are liable to be missed by executives, simply because they don’t get around to reading them. The metric reports that executives do read and pay attention to are weekly or monthly standard business and financial reports. But if your metrics are not reported in these standard reports, they will not drive action. And because many HR metrics (i.e. employee turnover and weak hiring) do have a significant business impact, it only makes sense that they should be provided alongside the business-impact metrics from finance and marketing.

HR provides no external metrics — Almost without exception, 100% of the metrics that we provide to HR staff and line managers are internal metrics, covering things that happened inside the organization. But unfortunately, no organization exists in a vacuum. For example, the success of recruiting and retention are both dramatically impacted by the area’s unemployment rate (an external metric). Actions by talent competitors (i.e. hiring freezes and spurts, layoffs, and mergers) can also dramatically impact our recruiting and retention results. Compensation’s impact can be dramatically changed by monthly changes in the prevailing wage rates.

Other relevant external metrics might include interest rates, healthcare costs, commuting costs, and the cost of relocating. What is needed is a mixed approach, where decision-makers are provided with “integrated metrics” (including both internal and external data) that provides the broad range of relevant information that is required for accurate decision-making.

HR makes comparisons difficult – Almost without exception, HR metrics include a single number (i.e. our turnover was 6%). But a single number doesn’t tell you relatively where you stand. In order to make it clear where you stand compared to a standard, every reported metric must also include benchmark comparison numbers (i.e. our turnover is 14% but the industry average is 5%). This will allow managers and HR functions to compare their performance, not just to last year but also to the best, the average, and the worst within your firm and also within the industry 100%. HR does some benchmarking but it doesn’t integrate and share those numbers simultaneously with the metrics that it provides decision-makers. HR also frequently avoids preparing and widely distributing “ranked” lists of performance metrics (where the performance results of individual managers are ranked). Distributing ranked results can increase internal competition and improve best-practice sharing.

We don’t track decisions over time – If one of your primary goals is to increase the speed and quality of people-management decision-making, you need to track major decisions over time in order to determine if they improved and by how much. Unfortunately, only one firm that I am aware of has ever tracked their decisions to determine whether they improved and whether HR had a major impact on that improvement.

HR doesn’t provide just-in-time alerts — Even the best forward-looking predictive analytics will fail to forecast some major problems or opportunities. In these cases, HR metrics must have the capability of alerting decision-makers so they can act immediately. In the case of opportunities, this might include notifying managers immediately after the recruiting function learns that a key employee at a competitor is leaving (so that they can be recruited even before a requisition is opened) or when a competitor is about to undergo a major layoff, making recruiting away their talent much easier.

Metric-selection-related Faults

We fail to provide algorithms that make HR more precise – Algorithms are mathematical formulas or equations that are designed to accurately identify the precise actions required in order to improve HR results. Google, for example, has developed extremely accurate algorithms for hiring, retentio,n and leadership. Unfortunately, the HR functions at few other firms have even explored the value of algorithms in directing HR actions. Perhaps because all too often, HR professionals seem to be enamored with intuition and speculation rather than relying on data supported decision making.

The number of metrics HR provides is too large – Because in many cases, HR doesn’t really know what metrics decision-makers want, they provide a huge array of metrics. Often these metrics are not prioritized so the sheer number causes confusion and frustration on the part of users. A superior approach is to focus on the metrics that decision-makers care most about and that have the most impact on business results. The ideal number of strategic metrics may be five or less.

We provide mostly tactical metrics — Most of the metrics that HR provides to executives are tactical metrics, which means that they focus on transactions, efficiency, and cost savings. Instead, we should only report strategic metrics to senior managers and executives. These strategic metrics might include the productivity of the workforce, the dollar impact of losing key employees, the dollar impact of superior hiring, and the dollar impact resulting from the development of our employees.

Our metrics are selected in isolation – Literally every HR metric selection process that I have encountered makes a selection based on HR administrative ease, and the decision is made exclusively by HR professionals. The problem with that approach is that the end users (managers) are not involved in the final decision on the metrics slate to be used. Instead, at the very least, the CFO (the undisputed king of metrics) should be involved throughout the process both because of their expertise but also because they provide your metrics slate with credibility.

Final Thoughts

I understand the reluctance of HR leaders to abandon their current approach but it’s time to realize that when you compare our metrics to those of other business functions, even the most optimistic cannot put us in the same league with them. For example, the production function measures and strives for 6 sigma quality results (approximately three errors per million iterations) but no one even dares to calculate the error rates in HR programs like performance management, recruiting, or retention.

If the above critique and analysis are not credible to you, I suggest you take your HR metrics process over to supply chain or quality control and ask them to conduct a direct side-by-side comparison. It’s time to face reality: decision-making in the business world is becoming more and more numbers and data-driven and our current historical report based metrics approach is … so last century!

This article is provided for informational purposes only and is not intended to offer specific legal advice. You should consult your legal counsel regarding any threatened or pending litigation.