In 2002, when Lee Gelb advocated a rigorous productivity analysis of Starbucks employees, her belief in the value of metrics was regarded as quixotic or worse, by many people in human resources. True, F.W. Taylor had pioneered what came to be known as "scientific management" through detailed labor studies nearly 100 years before, but within HR the idea that productivity could be assessed as if workers were robots flew in the face of the department's ostensible people-first mission.
And top management wasn't interested, anyway. Gelb, senior vice president of human resources at Starbucks, soon found that her goal of using sophisticated metrics to track, rank, and analyze the ROI on recruitment and development of every employee garnered little support anywhere in the company. "We really had to put up a good fight," recalls Gelb, who left the stimulant supplier seven years ago. "Management agreed that it was important, but they wouldn't give us any money for it. So we set out to prove to them that such programs would save money and reduce turnover."
Remember, this was 2002 — with $3.3 billion in revenues, Starbucks had already risen to become the world's fastest-growing brand, according to Interbrand's rankings. Between 2000 and 2003, the number of stores nearly tripled, from about 2,300 to 6,600. "We were spending an enormous amount of money on executive recruiters, with not very good results," recalls Gelb. That capital outlay, she argued, would be better invested in additional training for homegrown talent.
But while corporate took great pride in touting its higher level of "employee satisfaction" — which, the theory went, gave profits an extra shot by reducing turnover and thus the costs of entry-level training — it was far less certain about programs that would help staffers move up the ladder. "It's wonderful to keep people," says Gelb, "but everybody needs training. We shouldn't have been putting people in positions where they were going to fail because they didn't have the experience or expertise." Fortunately, even without the budget she sought, Gelb was able to tap into expert advice because "outsiders were anxious to work with us. They thought they would gain a lot by having an association with Starbucks."
Most companies aren't in that position. In fact, just figuring out what to measure and how to measure it has often proven to be more than most companies could tackle. Even a basic metric like turnover wasn't useful unless HR could figure out, using surveys perhaps, why employees were leaving. "HR didn't know what to measure, or what to do with the measurements they had," says Richard Beatty, a professor of human resources at Rutgers University.
HR, in short, settled into a comfort zone, delivering a handful of rudimentary metrics that usually failed to connect employee performance to corporate performance. Certainly none of those measures helped senior executives respond to the economic collapse. The relationship between finance and HR, often strained to begin with, got worse, since the failure to deliver meaningful metrics reinforced the suspicion that HR was far more interested in lavishing empathy on workers than applying brainpower to the business. Executives were more apt than ever to tune out HR suggestions about improving "work/life balance" or bettering "employee engagement" scores, since they had no idea how such efforts would rescue the company's earnings per share.
"HR people have the responsibility for a very expensive resource — the workforce," says Beatty. "But in the end, what does HR deliver?"
Pressured by the downturn, senior management now wants to know why HR can't behave more like its functional siblings — IT, say, or purchasing — and identify moves that will maximize the company's human-capital investment. Where are the inefficiencies in the company's talent supply chain? If employees are "the company's most important asset," as HR executives oft proclaim, why isn't that asset being coolly assessed like any other?
"Company executives are all over HR saying, 'Give me something that will impact quarterly results now,'" says Drew West, product marketer at Kronos, which makes workplace-measurement software. More than that, they are also asking HR for help with understanding the future. "This recession came out of the blue for many organizations," says Kevin Martin, vice president and group director at Aberdeen Group. "Now they want to plan for the unexpected. They want to look more than 18 months out and assess how the people they have will map to what they need."
The fact that CFOs suddenly want more out of HR is a complete reversal of how they've usually viewed it: as a locus for cost-cuts. Is now really the time to spend money on recruiting? Do programs that help employees develop additional skills matter when there are so few jobs for them to go to? Is there a CFO anywhere who wants to hear about the long-term value of staffing up a given area at a time when revenue-per-employee is under scrutiny?
Thus HR must confront a major change in expectations as it seeks to demonstrate exactly how it boosts productivity, fattens margins, or indisputably sharpens the company's competitive edge. But that kind of change may have to come from without as well as within. C-suite executives may have to provide a clear mandate, and funding, to propel HR to the next level.
HR has always been a backward-looking function, serving up measurements of long-established processes and practices, or interviewing people as they leave rather than as they advance. If anybody expected much more of the function, they certainly didn't say so at budgeting time.
"Within HR, the sense was that we always had potential that was going unrealized," says Cynthia Maltbie, founder of HR firm People Logic and former executive vice president of human resources at Fidelity Investments. "HR has not been an incredibly well-funded area, and we needed to have funding to pay people to do the deeper analytical work. So rather than take a leadership role, we tended to be more oriented toward fulfilling the agenda an executive would set. We tended not to step out boldly."
And, really, what would they have said anyway? In recent years it has become fashionable for top management to mimic GE's iconic CEO, Jack Welch, who preached a staunch up-or-out approach. Welch praised HR as a critical function, arguing that its job was to rank employee productivity and fire the bottom 10%. Otherwise, the presence of those laggards would drive top performers away, planting the seeds of the company's implosion. Welch's fundamental rank-and-yank philosophy caught on (though the target percentage varied), in part because of its simplicity.
But the concept has its flaws. For one, rank-and-yank depends on skilled supervisors rendering rational judgments, yet Michelle Malay Carter, vice president at consulting firm PeopleFit, estimates that only about 40% of organizations have the proper structure in place to ensure that the most appropriate person evaluates employees. And since the company would always have a bottom 10%, this strategy inevitably breeds fear.
Then there is the question as to just how dead your deadwood really is: Without broader benchmarking, how can you possibly know that there are better replacements out there? Fortunately, this particular pearl of managerial wisdom "seems to have run its course," says Carter. But it may provide a useful example of how HR policies do, in fact, matter, and affect the entire organization.
Power to the People
That's worth bearing in mind as companies grapple with the concept of more-meaningful HR metrics. Over the past decade, as more of HR's transactional functions have been outsourced and automated, there has been keen interest in developing some sort of ultimate HR tool. A sophisticated system that pulled data from all over the company could enable company analysts — perhaps embedded in the HR department — to assess human capital in much the same way it ranks customers: by how much value they add to the company.
The technology could determine "what head count you'd need to support a new initiative," says Jason Averbook, CEO of Knowledge Infusion, a consulting firm specializing in HR analytics. The main contribution of HR would no longer be in filling an open IT position as effectively as possible today, but in determining "what kinds of coding skills you'll need in three years," Averbook says.
Metrics, however, can go only so far, as Jeff Mattiuz is learning. The vice president of finance at Oberg Industries, a tool-and-die maker that has seen its orders drop 45% in the last year, has turned to metrics in hopes of making some wise head-count decisions. He measures direct-labor utilization, overtime hours paid, company wages against industry benchmarks, and similar measures. The analysis "makes it easy for a numbers person to just cut people, reduce benefits, and restructure," he says.
Theoretically, that is. Walking the floor of his nonunion shop, Mattiuz can see that some of his recommendations have not been heeded. In this case, don't blame HR: they are firmly behind him, and acknowledge that some employees don't have enough work to justify their hours. But the operations managers disagree. "They tell me the statistics are just wrong," he says, "and I do recognize that numbers don't tell the whole story."
That's part of the story, anyway. "It's tough when you get down to personalities," he says. "There are certain employees that managers want to protect, so they shelter them and keep their hours up. I enjoy a healthy debate, but we can't sustain the business if we don't make the tough decisions."
It's one thing to measure; it's another thing to act. One possible solution, Averbook says, is to staff senior HR positions with people who have come from finance or other areas where ROI analysis and multivariable correlations have long been studied. Other companies may choose to restructure so that the head of HR reports to the CFO or to a chief performance officer.
"We are moving into a world where we have to be much more concerned with competing on the basis of our brainware," says Rutgers's Beatty. "The way we manage talent will make a profound difference in a company's success." A more sophisticated use of metrics would seem to be essential in establishing a link between popular HR initiatives, such as training, and their impact on corporate performance.
"The language of organizations is numbers," says Beatty. "HR needs to have the numbers to win arguments with management about where investments need to be made." And management needs to make sure there is a large enough number in the budget to allow HR to do that.
Josh Hyatt is a contributing editor of CFO.
Open to Misinterpretation
The numbers tell a story — but if you aren't careful, that story could be a fairy tale.
Like box scores, human-resources metrics offer a few critical data points but virtually no context. So before you start stamping bar codes on workers' foreheads, you need to be clear about your company's larger strategic goals — and how the metrics you choose will help you achieve them. There is little point, after all, in devising metrics that help you decide whether an expensive training program is truly effective when the more critical issue may be whether rework is eroding the profitability of specific kinds of projects.
In short, HR metrics can provide definitive answers, but to which questions? Before you order up equations from HR, therefore, ask yourself a few questions:
Are we measuring the right dimension? If you want to calculate the productivity of employees at your call center, don't study the length of their average call. These aren't "411" operators, after all; you want them to take the time to satisfy customers, so customers don't keep calling back. Instead, consider collecting customer feedback and studying the call-back rate. And do it over time.
Are we too narrowly focused? If you use a simple metric to assess salespeople — how many customers they contact daily, say — you will soon find that your salespeople contact more customers every day than QVC does. But the reps may alienate them by being too aggressive or not providing adequate service. They may even become skillful in passing customers on to another department at frightening speed. "You want to look at one thing — but from different angles," says HR veteran Cynthia Maltbie.
Will the data tell us anything? This sounds obvious, but plenty of companies benchmark salaries and benefits against others in their industry — some of which are likely not relevant. Being in the same vertical does not mean that you and your rivals have goals, organizational structures, or corporate cultures that are in any way alike. — J.H.