OKR

Anything to do with the traditional world of get a degree, get a job as well as its alternatives
Post Reply
jacob
Site Admin
Posts: 15907
Joined: Fri Jun 28, 2013 8:38 pm
Location: USA, Zone 5b, Koppen Dfa, Elev. 620ft, Walkscore 77
Contact:

OKR

Post by jacob »

Does anyone have any experience with the OKR system of management? OKR stands for "objective/key-results". As far as I know, it's popular in Silicon Valley. How does it feel in practice in terms of stress? creativity? busywork? meetings?

bostonimproper
Posts: 581
Joined: Sun Jul 01, 2018 11:45 am

Re: OKR

Post by bostonimproper »

I’m a product manager in big tech— pretty much every company I’ve worked at does some combination of OKR setting and creating a prioritized project list to do annual, half, and/or quarterly planning.

The canonical book on OKRs is Measure What Matters by John Doerr. A lot of management is forced to read it at some point, I guess, so you should start there. In practice, though, every organization (or even team within an organization) has a different philosophy and practice in doing planning. Really, it’s like any other management tool— applied well by good management, it can be great; applied poorly by bad management, and it can be terrible.
How does it feel in practice in terms of stress? creativity? busywork? meetings?
Not sure what you mean by this? In broad strokes, this is the general process I’ve seen for planning:
- Upper level management has a long-term vision. They create a primary set of OKRs (north star) for the year.
- Teams come up with ways that they may be able to help the company within their own remit. Sometimes upper management doles out individual OKRs to individual teams. Sometimes teams propose their own OKRs which ladder up to company OKRs, based on the metrics that make the most sense for that team to own. This is usually led by team managers, though often brainstorming with their team. At the end of this brainstorming, you end up with a team plan that says: Here’s what we’ll do (projects) and here’s how we think it’ll ladder up to company goals (team OKRs).
- Team management “pitches” their team level plans to upper management, including team level OKRs and projects that would move the needle on those OKRs.
- This is followed by a lot of back and forth— Are we really going to hit our org-level OKRs of +5% on this metric if the team says they can only do +3%? Can the team be more “ambitious” here? Is XYZ that doesn’t roll up to one of our objectives really that important? What if we increased head count, are there underfunded projects that get us to our goal? Yada yada yada.
- At some point management has gone back and forth and edited the plan enough that everyone is “satisfied” and teams move on to executing rather than developing plans.

It’s a lot of busywork and meetings for managers (team and upper), but that’s kind of the job? In general having a planning process (of which OKRs are a part) does help give visibility and maintain alignment with upper management on what teams are and should be working on. I would say “planning season” can be, like, a 4-6 week endeavor in a bigger company, so in general doing it once every half-year is doable and mostly useful but more frequent than that just prevents you as a manager from ever reaching flow state in execution mode.

In terms of OKRs specifically, (rather than obsess about the vagaries of a planning process) the idea is that by being objectives/results-first, then you can better orient teams to do what matters. So the projects they come up with should better align with company goals. I think that’s somewhat true. Even with OKRs, though, teams can often get into a failure mode where they already have in their head what “needs” to be done and use OKRs to demonstrate to management that their preferred projects kinda sorta move the metrics they want so please just let us do this thing already. Another common failure mode I’ve seen is that management picks really bad KRs (metric) that doesn’t align well with the objective they’ve set out, and teams end up optimizing for the (bad) metric instead of the underlying qualitative goal (objective).

So, yeah, not sure I answered your question. Happy to clarify anything. Obviously, this is limited to my own experience as a tech peon.

macg
Posts: 174
Joined: Tue Mar 31, 2020 1:48 pm
Location: USA-FL

Re: OKR

Post by macg »

They have implemented OKRs here at my work over the past couple of years, and so far it's worse than useless. My management chain doesn't have a particular vision of the implementation, so the small team I'm on has no real guidance, we just end up filling things out at the end of the year purely to have documentation for the year-end review. So it's "worse than useless" because we have to spend time making things up and creating documentation purely for the sake of having the documentation, as opposed to actually helping us achieve our goals.

I personally don't understand why it's popular or how it's useful, even after researching its usage out there. Perhaps that's because of the failure of the implementation where I am. But even the concept of it is unclear to me ...

Some things that jump out that I don't get:

- it's not supposed to be used to track everything or track real details, it's higher level. If you look at it from a GTD perspective, that means it's maybe at a Project level, or even at a higher level like a Goal/area of focus. So it's more like a high level goals summary maybe? And it's not necessarily clear how to translate day to day activities up to OKRs, or even if you're supposed to...

- you're not supposed to actually complete the OKRs, if that makes sense (it doesn't to me). An OKR is supposed to be over-reaching, in the end you're supposed to accomplish ~80% of the objective

- it's very number based. You shouldn't have an OKR that says "complete all support tickets within established SLA", it should have a quantifiable number to it, so something like " complete 100 support tickets ". This might be fine for sales and other objectives, but not all in the line of work I do.

Personally, something like GTD works far better for me, because it has the flexibility to work with day-to-day tasks (next actions), all the way up to higher level planning, should you desire it.

Again, please understand that I am forced to use OKRs in a poor, unguided implementation. There are success stories out there for OKRs, I just don't personally understand how they would be useful to a person in my position.

Smashter
Posts: 541
Joined: Sat Nov 12, 2016 8:05 am
Location: Midwest USA

Re: OKR

Post by Smashter »

I work as an individual contributor at a venture backed tech startup that uses OKRs. I haven’t enjoyed my experience with them. I think this is partly because I was an early employee, and I prefer the more Wild West vibe, where people were just doing what needed to be done without as much structure. I understand that can't last forever, but the emergence of OKRs has made everything feel just a bit more rigid, corporate, and boring. And I can't shake this feeling that we'd all be better off if the managers spent a little less time planning a little more time helping the team execute.

This could very well be specific to my company, but my specific gripe is that OKRs are liable to fall victim to Goodhart’s law. (“When the measure becomes a target, it ceases to be a good measure”) So if we want to improve our outbound email performance, leadership might say something like, “we need to improve our open rates by 10% this quarter.” Then the team will send less emails, or find some other hack that gets them to their 10% number, overall impact be damned. 

One time, after setting an arbitrary OKR and then watching the number rise, my manager said, “amazing, as soon as we started focusing on this the number went in the right direction!" Which was true. But how we got that number to go up was not discussed.
bostonimproper wrote:
Thu Dec 29, 2022 9:56 am
I would say “planning season” can be, like, a 4-6 week endeavor in a bigger company, so in general doing it once every half-year is doable and mostly useful but more frequent than that just prevents you as a manager from ever reaching flow state in execution mode.
That's a good point. We are not a big company and we set new OKRs every 4 months. It's a long and involved process that takes a lot out of the management team. I get wanting to be nimble and agile, but I wish they would do it less frequently.

User avatar
Sclass
Posts: 2791
Joined: Tue Jul 10, 2012 5:15 pm
Location: Orange County, CA

Re: OKR

Post by Sclass »

Agree with @smashter. This kind of thing works well if the management is technical enough to set the goal markers. If they understand the constraints they can often make an efficient plan to a successful outcome.

However if they are coming up with the desired results out of their imaginations there can be real trouble. Sometimes these plans can take a ship deeper into the rocks and further aground than a lighter approach if the management isn’t aware of the rocks. Been there done that.

The common mantra at my old place was “we have to aim for the stars because even if we don’t hit we will be so far ahead.” What happened is we got so far off course and so committed to a particular path we couldn’t really turn back efficiently. The product would end up a freak show of patches on patches.

Objectives and results are only good if you know what the right ones are.

User avatar
C40
Posts: 2748
Joined: Thu Feb 17, 2011 4:30 am

Re: OKR

Post by C40 »

I had extensive experience with this stuff for the last 5-6 years of my career. I managed the use of it in a factory where I worked, and then later advised many leadership teams and sometimes the executives. We used different terminology, but from what I see here, it's the same stuff. The context that I was involved in was manufacturing of a consumer good, but the methodology is applicable in nearly any situation. I applied a lot of what I learned at work about this and related methodology to my own life.

Perhaps something that we understood and did better was not being too hung up on a measurement number but rather understanding the actual business needs and focusing on improvements truly related to them.

IMO the methodology is great. In practice, it only works as well as people understand and us it. In my company there were years of people not understanding or using it well, which caused a lot of time and effort that, while not entirely wasted, was not very effective and in some cases even went against the general intent

Scott 2
Posts: 2825
Joined: Sun Feb 12, 2012 10:34 pm

Re: OKR

Post by Scott 2 »

My employer started rolling these out the month I retired. I am glad I got away.

It looked like a priority tree, with specific definitions for each level of the tree. General vs. specific. Time frame. Scale. Measurability. Etc.

From what I saw - all the same management problems persisted, with new jargon. People weren't following the process as defined, just trying to cram their personal agenda into the new framework. Too many #1 priorities. Blending of general vs. specific tiers. Scoping entirely dependent on who wrote them. Some old world stuff didn't fit, so they just pinned it on. People weren't prepared to commit to the check-ins, so were trying to short cut the process. Etc.

As defined, the canonical program looks pretty smart. But an organization who can follow it as defined, maybe doesn't need it? I think the primary benefit would be for standardized training of the management team, provided there is commitment from all levels of the org.

Post Reply