[rosseloh]

>if you're constantly hard at work on something important, when something else comes up (someone has a question, there's a bug or an outage, whatever), you either have to delay the thing you're already working on, or delay the thing that came up. This tends to have a cascade effect on most kinds of work, locking up all your people resources.

Working in a retail store/break-fix repair/MSP environment, for a small business in a small city, this is absolutely the case. There is nothing more frustrating than having three customer projects on your plate, all of which are important (think "the email server is down"), and then the doorbell or phone rings and you end up spending half an hour walking an old lady through resetting her facebook password. It's an absolutely massive productivity killer, as well as making the day feel longer.

More employees would be the normal solution, but that's not possible here (we've had more in the past, it wasn't financially viable, apparently). Unless of course they started paying commission based on what people actually got done instead of a regular wage, which I'm not a fan of. (Though to be fair, if we did switch to that, the one employee who barely does anything would either get his ass in gear, or leave, so win/win maybe?)

[salixrosa]

Sadly, from what I've read the commission-based approach often leads to worse long-term results, especially in software engineering. It depends on the kind of work, of course. The metric I use (and in this case I have no idea how others look at the problem) is the number of decisions the person has to make, especially having long-term effects or effects on other parts of the company. It's hard to make the right choice for the org when you stand to make a bigger chunk of money right now from the other option.