[tensor]

I don't think managers generally think that there is no loss of productivity when you have churn. Loss of productivity due to churn is a very well known phenomenon. But it's also true that almost any employee can be replaced, and if they can't you should probably fix your org structure or processes so that they can be. Having single employee dependencies is one of the worst scenarios for organization robustness, and that includes people at the C-level. The old adage that a good manager should aim to make themselves replaceable has some merit.

All that said, I think people often incorrectly assume that having people in an office will yield productivity gains. I don't know of any research that definitively shows that to be true, and good management should be doing result based performance management not ass in seat performance management anyways. Despite this, I still routinely meeting high level people in companies who insist that assess in seats somehow give more productivity. I don't buy that hypothesis.

[vanviegen]

> Despite this, I still routinely meeting high level people in companies who insist that assess in seats somehow give more productivity. I don't buy that hypothesis.

I'm teaching a software development programme to mostly 18 to 20 year old students. One of the ideas behind the programme is that students need to work through many increasingly difficult programming assignment. And that they do this for 40 hours per week in our 'office' (a class room with fancier furnishing and less density).

Due to COVID, we experienced some periods during which we had to revert to (partially) working from home. On average, about 40% fewer assignments were completed during these periods.

Our explanation (and also the reason that we have students come in full-time in the first place) is that:

A) Students who get stuck need to be able to easily seek help at that moment. Many students just don't ask for help when working from home, regardless of the many things we tried to stimulate this.

B) When students get stuck, feel frustrated, feel tired or are satisfied with something they've accomplished, many will come up with the idea to play a quick computer game. And get sucked in for the rest of the working day. This doesn't seem to apply to the few older students (with working experience) we have though.

So what does this tell us about employee productivity when working from home? Probably little, though I wouldn't be surprised if an organisation with lots of juniors has similar experiences.

[davidbanham]

Just pointing out that your data has a pretty clear confounder. “Wfh” and “wfh during the initial stages of a global pandemic” are very different things.

[vanviegen]

We've had to go back and forth between the two situations multiple times. Same results.

[rapind]

This isn’t about “single” employee dependency though, companies are seeing significant attrition due to these policies. There’s no way to know this is safe and it’s probably hard to estimate costs.

Even in an ideal configuration where your employees are basically redundant raid drives (and yes you need to pay a lot for this redundancy which I doubt most employers do), you generally don’t want to lost too many at once, and even if you survive you still have high costs associated with replacement (training, hiring, fit, etc)

It might still be the best long term decision in some cases, but man, you better be making this decision based on more than a hunch.