4 ms·

People who are consistently conservative in their estimates do get noticed, and the estimates adjusted appropriately in the same way that people who always unde

People who are consistently conservative in their estimates do get noticed, and the estimates adjusted appropriately in the same way that people who always under estimate get their estimates adjusted. That’s fine, the worst that’s going to happen is a conversation about learning from previous estimates. People who sandbag to give themselves an easy life, or to try and empire build by getting more resources, definitely get noticed, and not in a good way.

swiftcoder 3y ago
> try and empire build by getting more resources, definitely get noticed, and not in a good way This is actively incentivised in a lot of tech companies - maybe more so for managers than ICs, but the bigger-your-fiefdom the easier it is to promote phenomenon tends to be pretty widespread.
- aardvark179 3y ago
  Empire building is often incentivised, sandbagging to try and get those resources generally isn’t.
rightbyte 3y ago
"Adjusting estimates" is just another way to force engineers to lie in their Jira tickets about how much time they spend on each ticket. At one point my agile master was complaining to me about estimations, so I just focused on reporting the correct amount of hours to make the burn down chart be straight. Decoupling reported time from actual time spent. Good lord I switched jobs.
couchand 3y ago
Humans are bad at estimation and there's no evidence that "having a conversation" will make them any better at it. Your organization should acknowledge and handle the fallibility of humans, not try vainly to excise it.
- aardvark179 3y ago
  If your colleague always estimates that a particular type of task will take a couple of days, and it always takes a couple of weeks, you wouldn't discuss it next time they estimate it is only going to take a couple of days? Yes, humans are bad at estimation, which is why it's good to reflect on how well you do and to correct if you consistently over or underestimate.

swiftcoder 3y ago

> try and empire build by getting more resources, definitely get noticed, and not in a good way This is actively incentivised in a lot of tech companies - maybe more so for managers than ICs, but the bigger-your-fiefdom the easier it is to promote phenomenon tends to be pretty widespread.

aardvark179 3y ago
Empire building is often incentivised, sandbagging to try and get those resources generally isn’t.

rightbyte 3y ago

"Adjusting estimates" is just another way to force engineers to lie in their Jira tickets about how much time they spend on each ticket. At one point my agile master was complaining to me about estimations, so I just focused on reporting the correct amount of hours to make the burn down chart be straight. Decoupling reported time from actual time spent. Good lord I switched jobs.

couchand 3y ago

Humans are bad at estimation and there's no evidence that "having a conversation" will make them any better at it. Your organization should acknowledge and handle the fallibility of humans, not try vainly to excise it.

aardvark179 3y ago
If your colleague always estimates that a particular type of task will take a couple of days, and it always takes a couple of weeks, you wouldn't discuss it next time they estimate it is only going to take a couple of days? Yes, humans are bad at estimation, which is why it's good to reflect on how well you do and to correct if you consistently over or underestimate.