3 ms·

Used to think OKRs were total bullshit, until I started working on a project spanning multiple teams with cross-org dependencies. OKRs are effectively a quarte

Used to think OKRs were total bullshit, until I started working on a project spanning multiple teams with cross-org dependencies. OKRs are effectively a quarterly negotiation to get alignment across teams so that large, complex, projects get done on time. They create accountability and the ability to escalate up the chain when a team you have a dependency on isn't prioritizing what you need to unblock your project. Generally, if you think OKRs are bullshit, it means they're being applied at a too granular level, or misapplied in an organization that doesn't require strong collaboration across team or org boundaries. It "seems" easy and obvious when you're a leaf node how a team, org, company should be run and coordinated, but when you actually try and do it, and lead, it's nowhere near as easy as it seems. OKRs, misapplied, suck. But there has to be some way to get large groups of people to align on specific targets. Effectively something like them have to exist.

smugglerFlynn 3y ago
This. There is a heavy lifting issue hiding behind using OKRs as a tool: you need to build an organization that is able to held teams accountable, and has a working escalation chain. The prerequisite for OKR is that your organization is driven by objectives which cascade from C-level down to teams, and then all the way back via feedback loops, instead of VPs and directors micromanaging every link of the chain on the go. Drive-by style of objective setting, when things get randomly thrown at random teams, is another problem that needs to be resolved first. Your OKRs will fail without building this kind of management culture. And you’ll be surprised how many orgs are not working that way, and insane amount of resistance people in the org have when you try to implement these (logically efficient) practices.
bhawks 3y ago
OKRs are primarily a communication tool, then a prioritization tool, then an execution tool. If you don't need to improve comms or prioritization then OKRs are of limited value.
drdrek 3y ago
Bad org culture will be bad no matter the mechanisms. You can use OKRs to be toxic and you can use Agile to be toxic and you can use your own bespoke bullshit to be toxic. There is no management framework that can fix bad culture because its only a part of it. Incentives structure, existing culture, hiring practices, company life cycle, market conditions, etc all contribute to it and no amount of data can fix those. Managing a company is like parenting, if you think that you can replace attention, care and hard work with a framework you are doing it wrong.

smugglerFlynn 3y ago

This. There is a heavy lifting issue hiding behind using OKRs as a tool: you need to build an organization that is able to held teams accountable, and has a working escalation chain. The prerequisite for OKR is that your organization is driven by objectives which cascade from C-level down to teams, and then all the way back via feedback loops, instead of VPs and directors micromanaging every link of the chain on the go. Drive-by style of objective setting, when things get randomly thrown at random teams, is another problem that needs to be resolved first. Your OKRs will fail without building this kind of management culture. And you’ll be surprised how many orgs are not working that way, and insane amount of resistance people in the org have when you try to implement these (logically efficient) practices.

bhawks 3y ago

OKRs are primarily a communication tool, then a prioritization tool, then an execution tool. If you don't need to improve comms or prioritization then OKRs are of limited value.

drdrek 3y ago

Bad org culture will be bad no matter the mechanisms. You can use OKRs to be toxic and you can use Agile to be toxic and you can use your own bespoke bullshit to be toxic. There is no management framework that can fix bad culture because its only a part of it. Incentives structure, existing culture, hiring practices, company life cycle, market conditions, etc all contribute to it and no amount of data can fix those. Managing a company is like parenting, if you think that you can replace attention, care and hard work with a framework you are doing it wrong.