3 ms·
I think most companies do not know. Productivity of knowledge workers is pretty hard to measure at a level more granular than an entire department. In my anecd
by mattnewton 3y ago
I think most companies do not know. Productivity of knowledge workers is pretty hard to measure at a level more granular than an entire department.
In my anecdotal experience the leadership of organizations is mostly flying by feel on this issue. And, by nature of their position, they skew heavily towards extroverts who value lots of work interactions. So their gut feels more face to face contact is important and that’s what they mandate.
- brigadier132 3y agoIt is not that hard to measure, at the end of the day code needs to be written. If the work you are doing is design heavy, then documents and charts need to be created. And then code needs to be written. Then finally, a product or feature needs to actually be launched.
- sodality2 3y agoNot everyone works in software. It can be hard to measure deliverables in a position that isn't deliverable-oriented. Maybe you're crucial to the department but your output isn't measurable directly.
- nine_zeros 3y ago> It is not that hard to measure, at the end of the day code needs to be written. If the work you are doing is design heavy, then documents and charts need to be created. And then code needs to be written. Then finally, a product or feature needs to actually be launched. Very naive. If this was my manager, I would immediately know that my manager is unskilled.
- brigadier132 3y agoWell I'm not a manager so it is naive but I've been working at these companies for close to a decade now and they produce nothing and there is rampant waste everywhere. So the current management philosophy just sucks. Also I'm very good at what I do (developing software), the common thing among the people that produce actual value to companies is that they are high output. Whether it's code or designs or whatever. And I'm talking about quality output obviously.
- mattnewton 3y agoWas that feature the right feature though? How do you measure the org’s ability to ship another feature after that at an engineering level (ie, is it easy to maintain? Did they do good documentation or mentor others?) You very quickly end up with giant spreadsheets of tons of variables trying to model this without “customers gave us more money” which rarely works at the individual or even team level. I think about stories like this: https://www.folklore.org/StoryView.py?story=Negative_2000_Lines_Of_Code.txt https://www.folklore.org/StoryView.py?story=Negative_2000_Li...