Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Ask HN: How do you handle underperforming remotes at your company?

news.ycombinator.com

Post reply on HN
Print view View original

21–30 of 38 posts

Re: Ask HN: How do you handle underperforming remotes at your company?

#21

Post by tharkun__ »

The_Colonel wrote: post #10
FirmwareBurner wrote: post #8

Earlier quoted context omitted.

Or check Jira and Git, it's not rocket science. If the person has a task assigned that would take a junior max 3 days of chill work to knock out but it's been 7 days and still no results and the person hasn't reported any blockers on the dailys, it's pretty obvious there's a performance issue.

It's quite easy to hide (provide plausible deniability for) low productivity if the person is smart about it. Complain about environment issues, broken CI builds, unexpected complexity etc., there was one meeting too, then put in the minimum amount of progress.

For many of these there are fixes.

I've had people try the env issue thing. I asked them to never try longer than 15 minutes to solve an env issue by themselves. Post on the public dev channel for help or ask me privately if they don't feel comfortable with it (it happens, that's OK). A lot of people just aren't good at troubleshooting env issues and I happen to be good at it even if I have never had that issue and I keep an eye on others posting and often can find a solution for people quickly. It's amazing how many people can't read error output, especially stack traces, and figure out which is the relevant message to search by.

Broken CI same thing really. And it's easy to compare if it's just them or everyone's builds are broken.

Complexity I offer help. Pairing for as long as needed. I've had a few guys where it became clear here that they really just weren't good. There are many signs here that you can collect and document. I've also done Re-implementations from scratch (not looking at their code) in some instances, proving their "hidden complexity" that took a week to not solve took me literally an hour to code down.

Re: Ask HN: How do you handle underperforming remotes at your company?

#22

Post by FirmwareBurner »

icedchai wrote: post #20
FirmwareBurner wrote: post #8

Earlier quoted context omitted.

Or check Jira and Git, it's not rocket science. If the person has a task assigned that would take a junior max 3 days of chill work to knock out but it's been 7 days and still no results and the person hasn't reported any blockers on the dailys, it's pretty obvious there's a performance issue.

I usually check Git, see nothing committed, then ask them privately "how are things going with your project? Do you need any assistance?" Another couple days goes by, some garbage that doesn't run is committed to git... I ask again. The cycle repeats. I'm just the tech lead on the project, I have no real authority, so if the situation doesn't improve I do mention it to the actual manager. They usually then do nothing…

Damn I'd like to work at your company to coast for a bit to relax. Where I'm at right now you'd be fired on the spot after a couple of weeks like that if your manager hears you're doing nothing.

Re: Ask HN: How do you handle underperforming remotes at your company?

#23

Post by FirmwareBurner »

GoToRO wrote: post #15
FirmwareBurner wrote: post #8

Earlier quoted context omitted.

Or check Jira and Git, it's not rocket science. If the person has a task assigned that would take a junior max 3 days of chill work to knock out but it's been 7 days and still no results and the person hasn't reported any blockers on the dailys, it's pretty obvious there's a performance issue.

Aaa... the mythical junior.

What's mythical about that?

Re: Ask HN: How do you handle underperforming remotes at your company?

#24

Post by 6510 »

rShergold wrote: post #17

I’ve had to deal with this a few times now. After asking around the greybeards I really respect this is my current run book. First: Talk to them about it. This is the hardest part. The aim of this conversation is to let them know you’ve noticed a change in performance and you want to know what you can do to help. It’s an open conversation but there are two possible outcomes. 1. They’re bored of their work. This is mo…

> First: Talk to them about it. This is the hardest part.

This is only hard if you do it after talking with everyone else.

Re: Ask HN: How do you handle underperforming remotes at your company?

#25

Post by FirmwareBurner »

The_Colonel wrote: post #9
TooSmugToFail wrote: post #2

Assuming you have objective performance metrics, and consistent criteria, the appropriate course of action would be to warn them, re-evaluate after a trial period, and fire if nothing changes.

"objective performance criterias" in a development role is a joke.

Then how about subjective performance criteria?

In a motivated and well performing team, everyone has a pretty good gut feeling on who's slacking off and pretending to work.

Unless ... the whole team is full of slackers who want to collectively engage in theatrics and perpetuate that there's no such thing as objective performance metrics in order to not be exposed as slackers.

Re: Ask HN: How do you handle underperforming remotes at your company?

#26

Post by tkiolp4 »

FirmwareBurner wrote: post #8
carlosjobim wrote: post #6

Depends entirely if you're his colleague or boss. To get good answers you have to ask good questions.

Or check Jira and Git, it's not rocket science. If the person has a task assigned that would take a junior max 3 days of chill work to knock out but it's been 7 days and still no results and the person hasn't reported any blockers on the dailys, it's pretty obvious there's a performance issue.

Or perhaps: the task is really not that important and nobody cares. If it should take “3 days of chill work” and nobody complains on day 5, well, perhaps it’s better to not solve the task at all.

Re: Ask HN: How do you handle underperforming remotes at your company?

#27

Post by FirmwareBurner »

tkiolp4 wrote: post #26
FirmwareBurner wrote: post #8

Earlier quoted context omitted.

Or check Jira and Git, it's not rocket science. If the person has a task assigned that would take a junior max 3 days of chill work to knock out but it's been 7 days and still no results and the person hasn't reported any blockers on the dailys, it's pretty obvious there's a performance issue.

Or perhaps: the task is really not that important and nobody cares. If it should take “3 days of chill work” and nobody complains on day 5, well, perhaps it’s better to not solve the task at all.

Found the Googler.

Clearly from OPs complaints it's clear that people are checking in on the work, which is missing, that's why he opened this topic.

Re: Ask HN: How do you handle underperforming remotes at your company?

#28

Post by icedchai »

FirmwareBurner wrote: post #22
icedchai wrote: post #20

Earlier quoted context omitted.

I usually check Git, see nothing committed, then ask them privately "how are things going with your project? Do you need any assistance?" Another couple days goes by, some garbage that doesn't run is committed to git... I ask again. The cycle repeats. I'm just the tech lead on the project, I have no real authority, so if the situation doesn't improve I do mention it to the actual manager. They usually then do nothing…

Damn I'd like to work at your company to coast for a bit to relax. Where I'm at right now you'd be fired on the spot after a couple of weeks like that if your manager hears you're doing nothing.

We had someone who got fired for performance issues. It took a good 6 months.

Re: Ask HN: How do you handle underperforming remotes at your company?

#29

Post by The_Colonel »

FirmwareBurner wrote: post #25
The_Colonel wrote: post #9

Earlier quoted context omitted.

"objective performance criterias" in a development role is a joke.

Then how about subjective performance criteria? In a motivated and well performing team, everyone has a pretty good gut feeling on who's slacking off and pretending to work. Unless ... the whole team is full of slackers who want to collectively engage in theatrics and perpetuate that there's no such thing as objective performance metrics in order to not be exposed as slackers.

My experience in remote teams was that people have smaller awareness of other team members. There's less communication, which means you have less direct one-to-one communication, but you also learn less information indirectly about others from such conversations.

> everyone has a pretty good gut feeling on who's slacking off and pretending to work

I remember one such case where I had a gut feeling, but I didn't undertake any action. Writing to a manager felt like snitching, and I could have been completely wrong (maybe they had valid reasons, unknown to me, to have low productivity), which would cause me embarrassment. I think it often takes a team consensus, sort of validation with somebody else. But such sensitive topics are often communicated subtly, indirectly, with body language or voice intonation. But I won't be writing my colleague on Slack if they share my suspicion...

Re: Ask HN: How do you handle underperforming remotes at your company?

#30

Post by The_Colonel »

tharkun__ wrote: post #21
The_Colonel wrote: post #10

Earlier quoted context omitted.

It's quite easy to hide (provide plausible deniability for) low productivity if the person is smart about it. Complain about environment issues, broken CI builds, unexpected complexity etc., there was one meeting too, then put in the minimum amount of progress.

For many of these there are fixes. I've had people try the env issue thing. I asked them to never try longer than 15 minutes to solve an env issue by themselves. Post on the public dev channel for help or ask me privately if they don't feel comfortable with it (it happens, that's OK). A lot of people just aren't good at troubleshooting env issues and I happen to be good at it even if I have never had that issue and I…

For sure, if you like to be a detective and expose the slacker, it is possible to do it. But more often I see people rather giving the benefit of the doubt.
Post reply on HN