performance management

the fast and furious way to organizational shrapnel

Kris Dunn over at HR Capitali$t (you should be reading his posts daily) recently posted Is Giving Employees a Yes/No Vote in Firings a Better Way to Go? It’s an interesting piece about software company Valve’s unusual practice of terminating through employee vote.

It got me thinking about the danger of copying innovative pieces from other companies without also using all their supporting systems. Removing the key leadership responsibility of performance management from leaders and putting it into the hands of peers is a very good plan for disaster. It’s not too hard to imagine the workplace devolving into the Lord of the Flies anarchy of a 6th grade popularity contest.

It’s also pretty easy to make the mistake of off handedly dismissing it as “it’ll never work”. Remember Puttnam’s Law: It is more acceptable to fail in conventional ways than in unconventional ways. The reward for succeeding in unconventional ways is less than the risk of failing in unconventional ways. Us humans like to downplay and ignore innovative success despite evidence that it seems to be working.


My only knowledge of Valve is from reading their Employee Handbook and their organization is completely unlike  99.99% of the companies out there. It is as flat of organization as you will find so the management structure as most of us know it simply doesn’t exist. In their structure, the employee vote doesn’t undermine performance management, it supports it. And it works because all the supporting systems work together. Analyzing or adopting this one component in isolation of the rest of the system is futile.

Consider it this way. Highly modified cars often have giant turbochargers, use nitrous oxide, run on exotic fuels (not available at your corner mini-mart) and can put out 4,5, or even 10 times the original power. But, installing a huge turbo or filling the trunk with nitrous bottles in your economy car after a marathon weekend of The Fast and the Furious is a quick shortcut to turning your engine into very expensive shrapnel. Yes, turbos, etc. can provide big-time power, but all the supporting systems (engine block and internals, transmission, differential, axles, etc.) must also but upgraded. Radically changing one component of interrelated systems rarely works.

What thinks you?


customer service: if you want a 10, do 10 work

You want your employees to do well, right? Of course, and you know that you can’t manage what you can’t measure so you set up some way of measuring their performance. And then you discover that you truly get what you measure, regardless of whether that’s what you actually wanted or not.

A salesperson recently revealed the disconnect between measured and desired outcomes when a co-worker purchased a car a couple of weeks ago. The salesperson’s parent company has a big focus on customer service and providing an outstanding experience. Each customer is surveyed after the sale and any rating lower than a 10 (the highest possible) drew negative attention for both the sales person and the dealer.

On the surface this sounds great. You can just imagine the company saying that they want every customer to have a 10 experience so that’s what they will measure and reward for. The problem – for this salesperson, at least – is that the focus shifted from providing an 10 experience to getting a 10. This is key: the focus shifted from the customer’s experience to the salesperson’s ratings.

This sounds similar, but it is very different. Because the salesperson was so worried about his ratings, he never bothered to provide service worth rating. I need to mention that she is one of the kindest, non-confrontational, charitable people one could hope to meet. She does not gripe or complain maliciously, yet had little good to say about the salesperson. Some highlights of my co-worker’s experience:

The car was being shipped because what she wanted wasn’t in stock the day it was purchased. Rather than keeping her posted, she had to constantly hunt down and badger the salesperson to find out the status of her order.

Whenever she pointed out his poor efforts, he blamed other people. In fact, it sounded like he spent the entire time saying, “You are going to give me a 10, aren’t you? It was never my fault things went wrong. You need to give me a 10.”

She was told that if she gave the salesperson and the dealer a 10 on the survey they would give her a free oil change.

The salesperson said that if she wasn’t going to give him a 10 it would probably be best if she didn’t do the survey at all.

She had to endure a bunch of whining about just how hard his life is and why she really needed to give him a 10.

He was so obsessed with getting a 10 that she hesitates to give less out of a mild concern of some type of retribution.

The whole thing sounded boring, repetitive, insulting, and possibly immoral. If you want a 10, do 10 work. If you can’t do 10 work and your career hinges on it, find another career.

I wonder if the company knows how much their dealers and salespeople are aggressively gaming the system? Punishing for anything less than a 10 seems counterproductive because if forces people to be short-sighted and silly and ultimately creates an experience that discourages repeat sales. People make mistakes, things go wrong, customers can be unreasonable, some people will never give a top rating except at gunpoint, etc. Insisting on continuous very, very high performance is fine, but it places much greater emphasis on outstanding hiring and very thorough training.

A top performer will still stand out and rise above when things go bad and try to make it right because they are focused on delivering a great experience, regardless of the circumstances. Marginal performers will retreat into fear and self-preservation. Their well-meaning system forces these extremes.

How would you set it up differently if you were the parent company?