0

The many issues of HR metrics

HR metrics

We live in a supposed era of big data — sorry, Big Data — right now, and as such, every team/silo/organization has to “prove it” with metrics. As a result, there’s been a bunch of discussion about HR metrics of late, typically in the form of won’t-happen-for-a-while concepts like “people analytics.”

Logically it would make sense for HR metrics to be often-considered, as HR touches the biggest spend of any company (hiring people) and theoretically has data on the performance of those people (and their managers). But the whole deal with HR metrics is extremely fraught, for a number of reasons.

HR Metrics Problem No. 1: No decision-maker cares

This will gradually change, but in general most executives could give 0.2 shits about Human Resources. To many of these guys, who view themselves as world-builders, HR serves these functions:

  • Processes
  • Fire drills
  • Get out the people I don’t like
  • Hopefully staffed with a few hot 20-something blond girls fresh out of school

Is this normative at all places? No. But almost everywhere I’ve worked or talked to my friends about, this is how executives view HR. It’s not “seat at the table” material. This problem underscores everything else. If you don’t care about the department, well, you won’t care about HR metrics either.

HR Metrics Problem No. 2: Our relationship with data

This is fraught in all departments right now. Many top decision-makers don’t really understand data, and many organizational processes are set up so that people can off-load the responsibility for the analytics. No bueno. Companies also tend to “throw money at the problem” of data, hiring $$$ data scientists instead of doing the more logical thing, as Peter Cappelli points out:

In short, most companies — and that includes a lot of big ones — don’t need fancy data scientists. They need database managers to clean up the data.  And they need simple software —  sometimes even Excel spreadsheets can do the analyses that most HR departments need.

Yep. Simplicity matters a lot here.

And No. 3: the HR metrics we capture and how we do it

Usually this is going to be about turnover, cost per hire, and maybe some employee morale evaluations. The problem is that a lot of this comes from performance reviews, which are awful, or employee engagement surveys, which typically occur once a year and that’s it. (And then no one thinks about it until the process begins the next year.)

Admittedly there are more real-time HR metrics solutions now — Waggl, TinyPulse, etc. — but I wouldn’t say they’re “at scale” in terms of companies using them a lot. It still very much feels like we half-ass HR data; we collect a bunch of stuff once a year, maybe do a few slides on it, and that’s it.




 

Kind of amazing if you think about it, since HR “owns” the people aspect of the business — which should be a really big deal. It isn’t, though. I think that’s largely because no one cares, somewhat because “we’ve always done it that way,” and analyzing supply chain or operations numbers seems more “business-like.” Guys want to feel “business-like” because it’s fun to them.

How can we improve HR metrics?

Couple of ideas:

  • Make sure the databases where info resides “talk” to each other (as noted above)
  • Tie everything to cost — how much $$$ is being lost on turnover?
  • Connect turnover rates back to specific managers, so that they can be ID’d and improved
  • Calculate cost per hire, but don’t live by that number; cost-cutting measures shouldn’t be the norm when getting good people
  • Analyze the metrics you have more frequently
  • Use quick, “pulse” surveys as opposed to once a year stuff
  • Have actionable returns on the bad parts of the employee satisfaction surveys
  • Care

Those are just some quick ones off the top of my head. I’m sure there are a million LinkedIn thought leaders right now meowing about People Analytics and how it’s going to change everything, but you know what? It won’t. First off, execs still won’t care. Second off, if you design a “prototype perfect hire” and then try to get 1,000 of those, it just means your company will have tons of homophily. Your ass will get disrupted faster than you say “Johnny from Seattle just designed a new app.”

If you want HR metrics to improve, then, start by caring — then move to tying everything back to the money and being consistent with your analyses.

Anything else you’d add on HR metrics?

Ted Bauer

Reply If You'd Like