Cascading OKRs: rigid vs. directional

Click on 'Follow' to listen to this Goal Diggers podcast episode on your channel of choice (ie. Apple podcasts, Spotify, etc.)

Interested in listening to other episodes? Head over to the Goal Diggers podcast.

[fs-toc-h2]Show notes

If you've read "Measure What Matters" you've probably come across the term cascading OKRs. Typically, when you think of cascading OKRs, top-down rigid cascading is what comes to mind.

In an ideal world, rigid cascading seems logical — it gives an impression that an entire organization is pulling in the same direction. In reality, organizations are slightly more complex and require multi-directional goal setting.

Top-down cascading is a reflection of traditional command and control. Therefore, there's a good reason why many softwares and consultants discourage top-down cascading of goals.

So what is cascading OKRs all about? We spoke to Peter Kappus, an OKR consultant, about:

• The definition of cascading OKRs
• Why organizations should avoid rigid top-down cascading?
• What should organizations be focusing on instead?
• Rigid vs. directional cascading

Curious to learn more? Here are some additional resources:

• Why your Key Result can't become someone else's Objective
• Two things Measure What Matters got wrong

[fs-toc-h2]Transcript

Henrik-Jan: Welcome to another episode of Goal Diggers. I'm Henrik, CEO of Perdoo an OKR software focused on strategy execution. I'll be today's host and today we have Peter Kappus with us. Peter is an international OKR coach, using OKR and agile practices to help teams maximize clarity delivery and joy. Peter, welcome back to Goal Diggers.

Peter: Thanks Henrik. It's great to be back!

Henrik-Jan: Peter, last time we spoke about individual OKRs, today we wanted to talk about cascading OKRs. Can you tell us what cascading OKRs means?

Peter: So a lot of people first encounter the idea of cascading OKRs, through either John Doerr's book Measure What Matters, where he talks about this fictional Sandhill unicorns fantasy football team.

It was also cited by Rick Klau in his now-famous or infamous Google ventures video about OKR from a few years ago. And, the main thing that people think about with cascading OKR, is what I often call rigid cascading. And this idea that every person's OKRs, are kind of descended from someone else's OKRs further up in the organization.

So in the book Measure What Matters, John Doerr talks about this idea that, you know, the general manager wants to make money for the football team. And one way of doing that is to win the Super Bowl and then the head coach, then his Objective becomes to win the Super Bowl and so his Key Results become: a passing attack amasses 300 yards per game, defense allows fewer than 17 points per game. I'm just reading these directly from Measure What Matters, but the idea is that at each level, that person's Key Results become the Objective of the person underneath. It's all very rigidly, kind of tied together.

Henrik-Jan: Yeah. We also noticed that with Perdoo. So lots of prospects that come to us expect, actually, that this is how the software works. And then often they refer to the book Measure What Matters. And unfortunately, this is not how Perdoo works, or fortunately. What's your experience with this?

Peter: Yeah, it's funny because I think it's a beautiful dream that really appeals to people who initially gravitate towards OKRs because it makes it look like everything in your organization is pulling in the same direction.

Everyone is working towards the same set of goals. And I think it kind of sticks in people's minds is this kind of ideal image, like a flawless diamond or something, but most organizations that I've worked with, they drop that approach almost immediately within their first few conversations in a multi-team kind of environment.

And the funny thing is, you know, the real world is more complex. It's more organic. It just doesn't, it doesn't work that way. And, you know, if you read Measure What Matters, you see this example and it looks great. It looks great on paper. And I think a lot of people just stop reading at that point, because it looks like such a clear way to join everything together.

But what a lot of people, myself included, the first time I read it, I missed this and you know, and that is that the next five pages he spends totally debunking this notion. You know, he says, yeah, you could do this very rigid, kind of tight cascading where everything has to ladder up to the goal above it.

But it does some things that are not great. It's it creates a loss of organizational agility, really for one thing. Cause it's slow, you know — if you're constantly having to every quarter, wait for everyone upstream from you to set their OKRs that could take days, weeks or even months. So, so it makes you less responsive and it makes you less flexible.

That's the second thing that he talks about is: it's very brittle. So, you know, humans are notoriously bad at predicting the future. And if we pick the wrong goal, anywhere in that chain, then the whole thing falls apart, like a house of cards. And you have to rethink all of the goals all the way down into every team.

So you lose a lot of flexibility. The third thing that he mentioned is, it really marginalizes contributors. So instead of thinking up what your own goals should be, you have them handed down to you from above. And that's, you know, really counterproductive. That doesn't help. So it cuts off a lot of really interesting, important conversations that people could have.

There's one other thing too, which is that the only kind of linkages you create in that environment are one-dimensional. So they only go kind of up the chain. You miss out on all these horizontal linkages that you could have. And other places where you can kind of have more interesting conversations and connect.

So all of those things make it kind of a poor choice for most organizations.

Henrik-Jan: Okay. But one thing that I never understood myself, when organizations have this expectation of cascading OKRs, and Key Results becoming Objectives for someone else, is that it is a good and well-known best practice, even amongst these organizations that want to take this approach. That Key Results should contain a metric. And then it's also well known best practice that Objectives should never contain a metric. If a Key Results is a Key Result or an Objective, depending on who looks at it or wherever it appears in the hierarchy, that never holds up, right.

That never works. And then the other thing that I find surprising is that if a Key Results should contain a metric, then only the metric that is used in the Key Result can drive progress for the Key Result. And the completion of sub-goals doesn't automatically mean that the value for that metric has changed, right?

So that expectation of progress propagating up and, other people completing sub-goals, be it Objectives or Key Results. However, you want to call it in this way, that automatically that progress, drives progress of the goals, higher up the chain. Whereas of course not always your Key Results will have metrics, and sometimes it's hard to identify good metrics, but at least a good amount of your Key Results should contain a metric.

Progress cannot propagate up anymore. And the only thing that can drive progress for that particular Key Result is whatever value that metric has at any point in time. Right? Or am I...

Peter: Yes, that's exactly right. I mean, Objectives don't usually have metrics or these big inspiring. You know, I think like the book covers that kind of gets you leaping out of bed in the morning.

But usually, we don't leap out of bed to increase conversion by 2%. You know, we might, leap out of the bed to completely dominate the market or whatever. But then the Key Results are the very measurable proof, that they hit that Objective. And that's, I think what's so powerful about OKR since they have something for everybody.

So if you're not a numbers person, you can get really inspired and excited about that Objective and that can get you going. But then they also encourage you or they force you really to back that up with some measurable, outcomes that you can actually see that have numbers in them. So as soon as you start to mix those two things together by turning one person measurable Key Result with the number into another person's Objectives.

As you say, you're really limiting their kind of definition of success, because it means that they have to hit that particular number in order to be considered successful. Which again is great if you can predict the future and kind of dictate everything from the top down. But generally, you know, the real world doesn't quite work that way.

So I think that, you know, a lot of people get kind of seduced by this ideal of rigid cascading. But then they struggle to kind of move beyond that.

And the reality is always much messier and more complicated and more organic. So I think, you have a need to need to account for that.

Henrik-Jan: What is it then that organizations like so much about cascading OKRs? Like, are there particular people inside a company that do benefit from this approach?

Peter: You know, I think, it appeals maybe to folks who initially gravitate to OKR in the sense that it gives you this illusion of perfect alignment. But, you know, another thing John Doerr says in the book is that, or I think he might've been quoting someone else who said "micromanagement is mismanagement". And, you know, when you have this kind of rigid cascading, you are kind of micromanaging and telling everybody exactly what goal they need to do.

And so I think. You know, OKRs are about simultaneously creating focus and clarity on commitments, but also providing that flexibility and different options around execution. You know, I've used the term "freedom to operate" before of, of letting people do what is right in their own sphere, because they are closest to the work where they, they need to do that.

So I think for me, the people who benefit are more: if you feel like you're in that kind of top-down command and control kind of role, if you're really embracing that sort of Management By Objective (MBO) style, where everything gets pushed down from the top, then it might benefit those kinds of people. But, to me, I think that's kind of a hold over from, from a bygone days of, top down management.

So I don't think there are a lot of folks who benefit that way.

Henrik-Jan: Yeah. I also remember reading a quote from someone elsewhere I said a lot of the ways that we approach management nowadays is based on how we used to run factories. But now it's more about running creative firms and that's just very different.

I think that cascading approach where everything trickles down and progress propagates up that maybe works well in a factory. But it probably doesn't work in the innovative and creative firms that we have today. And another thing that I liked so much myself, having worked with OKRs in our own company for several years now, is that with OKRs we sort of like mutually agreeing on the outcomes, the results that are so important. And we leave it entirely up to the person leading the Objective or up to the teams to decide how they're planning to achieve those outcomes and how they're planning to achieve that OKR. Which I think also helps me and my co-founder move away from micromanagement.

And only when things are not working out, of course, that's maybe when you want to start looking like, okay, what have you been doing to achieve these outcomes? We do that by tracking Initiatives alongside our OKRs, and then we can have a discussion. Why was it not working out? What are other things that we can try?

Peter: Yeah, exactly. That's a beautiful example. And I think it's a nice example as well of how OKRs are really an alignment framework. They are not an execution framework and you do need to have those initiatives kind of alongside the OKRs, but they are different things. I also love the Bill David quote, who said, "You can tell someone to clean up a mess, but should you be telling them which broom to use?" So it's this idea that you can set the destination, but you allow people to use their own expertise to figure out what the journey looks like to get there.

Henrik-Jan: Yes. Which essentially also means making the use of all the talent that you have, right? And I'm sure that talented people don't like to be micromanaged or don't like to be told what to do on an Initiative level, but are happy to work hard and use their talent to achieve the outcomes that are so important for the organization.

Peter: Absolutely. Exactly. And I think that's the thing about, you know, in the old world it was easier to, use top-down management because things were a bit more predictable and, traditional management techniques, like that, assume that the problem space is static. It's not going to change. And with things being so dynamic, as you mentioned, you need to use a lot more creativity and a lot more intuition from people at the front line to tell you as a manager: what needs to happen next?

Yeah. Are there scenarios where cascading OKRs does make sense? In the way we just described.

Yeah, exactly. Like I was just saying: unless you're in an environment where you can and you must perfectly predict what everyone has to do, and where the environment is not changing, and the cost of change is extremely high. So for example, if you're building a motorway or a nuclear power plant or something, then I think it makes sense to kind of rigidly cascade everything down from the top.

But in reality, those kinds of situations don't usually need OKRs because you're not doing it necessarily innovative new kind of stuff. You're following a blueprint, right? You're following a very necessarily rigid plan of what has to happen when. And everyone just needs to kind of show up and get on with what they need to do. So I often say if you have a project plan and you know, you need to execute certain things, you don't need OKRs , you need to get on with executing the project. But in reality, it's a bit more nuanced. It's a bit more subtle. And you do end up having to do some cascading, but it's usually not that kind of rigid tight sort of thing. I sometimes call it directional cascading, which is more useful, kind of in the real world and John Doerr talks about this a little bit in his book, you know, right after he starts to debunk the idea of this rigid cascading.

Henrik-Jan: Okay, so one thing that resonated with me is that you mentioned that, when you do want that top-down dictating of, of all these OKRs, you mentioned the word project management.

That's essentially what it is then, right. You're not really using OKR to identify the outcomes and the results that are so important for your business. You're using OKRs to track what needs to be done. So essentially you're using OKRs for project management then.

Peter: That's exactly right. Yeah. It's just a sort of glorified project management frame.

Henrik-Jan: Interesting. Let's talk about directional cascading. That sounds like a very promising approach. Could you tell us some more about how that works exactly? Like how do organizations then, for example, set OKRs on different levels? How does alignment work in that way?

Peter: Yeah, so again, you're not going to get away from aligning OKRs, especially in a large organization. And that's again where the power of OKRs comes from. And when I talk about directional cascading, what I really mean is looking both vertically and horizontally through the organization and really just being about transparency and having difficult conversations about where you want to go, how you're going to know that you've arrived there and what needs to happen in order to, to reach that point.

And for me, it's less about looking down at this thing as a kind of network graph where everything is tightly coupled and connected, and it's a bit more, looking at it kind of organically.

It's like you might look at a flock of birds or a school of fish and kind of, if you are at the top or anywhere in the organization, you should have a sense of where people are moving. And if anyone is going in a different direction.

That's another thing. You know, Laszlo Bock, former head of people, operations at Google, he wrote a great book called Work Rules, and he pointed out that, he says "having goals improves performance. Spending hours, cascading goals, up and down the company, however, does not.

And he said, he goes on to say, we have a market-based approach where over time our goals all converged because the top OKRs are known and everyone else's OKRs are visible. So for me, you know, you can kind of visualize this, a network graph with little lines, connecting all of the Objectives and Key Results.

But instead of those lines being kind of rigid, things that are enforced, I think of each one of those is a conversation. So really, if you're in an environment where everyone is publishing their OKRs — whether that's in some kind of a tool or a spreadsheet, or even just up on the wall. It's really about having those conversations. Starting early and really actively debating, you know, it's a little bit like if you think of the, like Kim Scott's Radical Candor approach.

You can apply that to OKRs and really go and approach people and say, Hey, you know, I disagree with this, or I think you're missing something or I'm confused by that. And it's all of those rich conversations that you want to have that really gives you the benefit of aligning those OKRs is in a very organic, nuanced kind of way.

I tend to be very cautious when I come in and I see a tool or an approach that forces you to kind of do that rigid cascading. Cause I think it kind of shuts down a lot of those conversations. And you know, if you think about Google's 20% time, that's almost the antithesis of rigidly cascading OKRs where you know, you're letting people to kind of go off into a different direction and try things and then kind of bring them back to what works.

And, you know, I guess at the end of the day, what it comes down to. Leaving people, uh, some wiggle room to, to use their own expertise, to use their own experience. I'm a big believer in bringing kind of the emotional side into OKRs as well, because that just taps into a much richer, more nuanced set of conversations that you can start to have.

So I think for me, it comes down to being very transparent leaving yourself enough time as an organization to have all of those conversations and to debate with each other and do that in a way that is respectful and constructive. But, You know, encouraging people to descent a little bit into, to push back and say, well, actually I don't think that this is how we're gonna do things.

And I did mention, I think it was Andy Grove who said that the frontline employees or staff members who I think he said were in the trenches, you know, those are the best indicators of what needs to happen next. So it's important to make them part of that conversation.

Henrik-Jan: Yeah. Couldn't agree. More, Peter, this was so interesting. Again, it was great to have you back on the, on the Goal Diggers podcast and I'm already looking forward to the next episode.

Peter: Thanks again, Henrik. Great to be here. Good chat as always. Yeah. Hope we can talk again soon. Thanks.

Henrik-Jan: Thanks.