Sith Sigma

Measuring quantitative vs. qualitative people metrics December 1, 2009

Regarding measuring the staff metrics, it’s relatively easy to measure quantitative performance indicators. I can measure how many rebel command ships a Captain of a Star Destroyer blows up, or how many rebels a Storm Trooper captures, and how many staff are fed to the panna monster for incompetence by my Sith Lords.

But measuring quality is hard. If someone puts in 50hrs/week and another puts in 80hrs/week, by sheer numbers the second person is putting in an enormous amount of extra effort. However, it’s really about results, so if the person at 50hrs/week gets more done in less time compared to the other person, should be rewarded more.

Technical work is a little bit easier to measure quality – if you can link back the amount of defects to a person, that’s one way to do it. However defects aren’t the only indicator of quality, if someone implements a “better” solution that allows for easier maintenance going forward… that’s a higher degree of quality.

Or for the Star Destroyer Captain, if he can take out a rebel command ship with the least amount of damage taken to the ship, that’s a statement of quality to his strategy.

But how do you quantify it in measurable terms… Sometimes you need to just start somewhere – in this case dollars spent on repairs would be a start.

The Danger of Entitlement and Incentives March 10, 2009

One of the problems with salary increases, bonuses, and other incentives (ESPP, stock options, etc…) is that they quickly become viewed as entitlements. The first or second time you try it, you get a boost of appreciation as it’s unexpected, but from then on it’s just merely expected in return for existing levels of performance.

So what’s a Sith Lord to do? Well for starters be careful how often you use it – reduced and unexpected frequencies help, and the moment it becomes expected stop before employees become dependent on it (I’ve seen people base their personal finances on anticipated bonuses, ESPP, and stock options).

Also, try to rotate between incentives so that it doesn’t become the expected, routine, and periodic quarterly bonus check. But rather incentives targetted towards rewarding something specific (e.g. a thanks for completing Death Star #23 ahead of time and under budget).

You can be successful at periodic incentives, just make sure they reward very specific achievements (preferrably goals and expectations that were communicated early on). You don’t want employees to feel they’re getting it out of entitlement (e.g. for merely existing for a period of time), but because they completed and outstanding job on a certain task (and if they want another one, they’ll need to repeat the outstanding performance – because it’s not going to happen with status quo).

Causing a Crisis to alter behavior February 11, 2009

Back in 2007 we posted research that showed that not even the possiblity of death was enough to change people’s behaviors. Most humanoids are weak minded, they hate change, and love staying within their comfort zone. But the reality is that you can never just coast along for long, change is inevitable, so you either get on board with it or you die.

As a manager you don’t want your business to die, and you don’t want people to lose their jobs. The key to managing people is to recognize that there is no one size fits all solution, you have to personalize your approach with each individual – what works for one person may not work for another.

So in your manager’s toolkit you may have tried various reward mechanisms, recognition, compensation, ownership, etc.. You may have tried to subtly point out that jobs are at stake – but people may have assumed that their particular position is safe because they’re working hard, so it’ll probably be someone else.

Here’s something different to try – create a crisis on purpose to see if it causes people to change. It takes 20-45 days to modify behavior, so with a bit of luck, perhaps a good jolt will cause people to adapt. It’s a little bit risky because you’re taking a calculated risk by exposing the business to the consequence of instability, so you’ll want to plan it out and identify what indicators you’re going to measure to know when to pull the plug on your crisis experiment.

If people aren’t filling the need, and things are functioning, how is it that the crisis doesn’t occur normally? Probably because you’re having to jump in and constantly handle the burning ambers yourself before they blaze out of control. But you can’t grow an organization if you’re having to burn up your time dealing with these little things while the team itself isn’t growing.

So what kind of crisis’s can you incur? If you run a support team, how about you stop jumping in to save the day whenever the team isn’t picking up the slack? Let them get overloaded, throw out suggestions, but see if they’re able to adapt. Or if a deadline is coming up on a project and normally you would jump in to save the day… don’t save it. Or if an automated alert indicates an important system is out of disc space and no one responds because you’ve conditioned them to subconsciously knowing that it magically gets taken care of… let it go unresolved.

You need to be creative with this one, but this is just one of many things you can try to alter team or individual behavior. Good luck!