« What's In It For Me? | Main | New Research on Cutting-Edge Performance Management: What Are We Learning? »



Feed You can follow this conversation by subscribing to the comment feed for this post.

Its an interesting comparison between the quality of a new car and a new hire. Many times, we'll spend more considerably more on the latter (and they'll stay with us longer), yet we don't often know if we're getting the quality we want (or need).

I can't help but think, though, that your hiring managers use some version of 'quality' in their assessment process. After all, they take a pile of ostensibly-qualified candidates and winnow the list to something manageable on the basis of quality, right? And from that list of Most-Highly Qualified (a moniker which really isn't about qualifications, but about a hypothesis of quality), the managers then discern (ala Car-nac, pardon the pun!) the highest quality candidate from the lot.

But, perhaps what you're suggesting is that what those hiring managers use as their proxies for quality are perhaps proxies for something else.

If only we had a Consumer Reports Guide to Candidates or a Kelly Blue Book to let us know how much each candidate was worth and what other buyers in my area had recently paid for theirs!

Interesting analogue to the KBB, since I recall using something like that with my new (used) car purchase recently. Of course, that accounted for condition, accessories/options, and even the mileage - which were all highly-quantifiable.

The only real certainty with people, is that you can be fairly confident that the odometer hasn't been rolled back (or at least fairly confident . . . at the present time). If only we could roll back those personal odometers quite so easily.

Best to "kick the tires" as they say.


Self-fulfilling prophecies complicate the metrics. Hiring is heuristic, with the least objectionable candidate inevitably garnering the most approvals from minions and managers and others in any CYA chain. "Taking a chance" on a problematic recruit with known "issues" is a death sentence in all bureaucracies where no error is ever forgotten or forgiven. Especially in The Puzzle Palace, I'm sure, where Doing Wrong is far more noticeable than Doing Right (which is assumed, overlooked or hidden).

Only two suggestions: (1) The best predictors of future behaviors are past behaviors. (2) Track those previewed as Super, OK or Lousy for predictive accuracy.

Kicking the tires today will generally yield a broken toe. Our best motor vehicle got great preview ratings when initially launched. The entire class then received terrible post-purchase reports as used cars, but ours has been absolutely perfect for almost a decade. Anything can happen.

I’m convinced that the traditional methods of defining (not defining) quality are pretty worthless. We need to quit focusing on degree type, name of university, college courses on a transcript and grades. And predicting future performance by looking at past performance is questionable at best.

Why not step back and think about the type of actual, practical skills that are needed in a job — skills that actually show a person’s ability to perform? And then figure out how to get an applicant to demonstrate those skills.

Google, among others, has done that. They don’t even care if a candidate has a degree or not --- much less one from MIT or Stanford. They are more interested in what a candidate can do --- not what he/she knows or has done. So they put them through their paces before they hire them by giving them mini-projects that test their skill. It’s not just results they look at but how the candidate achieves those results. In today’s world the how is as important as the what.

Michael Schrage on HBR’s blog calls these projects “projectlications” (project applications) or “applijects” (application projects). The purpose is to see if a candidate can actually produce and work collaboratively with a project team (if teamwork is a critical part of the job).

Yes, yes I know that pulling this off is not easy but it’s better than the current method of selecting “quality” employees. And after all --- since we are unquestionably high quality Compensation pros, we can certainly figure out a way to make this happen.

It’s easy to look at a resume and “assume” quality from where a person has graduated --- but you know what happens when you “assume” . . .

I like the idea of the projectications, although you already intimated how the idea may be impractical. I kind of wonder whether a more sophisticated battery of job-related testing and assessments might be an answer also.

Alternately, as the line from a movie goes, and the question asked following a series of bad ideas proposed to solve a particular situation, "Do you have any other bad ideas"? The response to the question of determining employee quality may just be, "This IS the best bad idea that we have".

No amount of testing will give you what you want. Testing is just another form of a "grade". I don't think a anything can take the place of watching someone actually apply what he/she has learned either at university or through job experience.

Using mini-projects takes some ingenuity but the results are worth it in my humble opinion.

The comments to this entry are closed.