From Strategy to Action: How OKRs Transform Fortune 100 Organizations
The Meridian Point Podcast
Host: Kumar Dattatreyan
Guest: Sevak Markarian
Kumar Dattatreyan: Good afternoon. It's Kumar Dattatreyan with the Meridian Point. And today we're joined by Sevak Markarian, a transformation leader who's cracked the code on one of business's toughest challenges: turning strategy into action. With a rare combination of a doctorate in organizational leadership, an MBA, and a master of science in engineering physics, Sevak brings both analytical rigor and human insight into his work. He's known for operationalizing OKRs in ways that actually drive performance, not just create more meetings. Working with Fortune 100 companies like HP, AT&T, Chevron, and Kaiser Permanente, Sevak bridges the gap between executive vision and frontline execution, helping organizations move from strategic planning sessions to measurable cultural transformation. So without further ado, let me invite Sevak to the stage. Hi, Sevak. It's great to have you on.
Sevak Markarian: Great. Thanks for having me.
Kumar: Of course. So Sevak, most organizations have brilliant strategies that just die in implementation, right? I mean, I'm sure you've had that experience. I certainly have. You've made a career out of closing that gap. What's the most common breakdown that you see between when leadership sets strategy and when teams actually execute on that strategy?
Sevak: Kumar, I really appreciate the question and the opportunity to be here, especially to talk about leadership in chaos and how well-crafted OKRs can help this organization in VUCA environments, which stands for volatility, uncertainty, complexity, and ambiguity environments. So as you mentioned in my introduction, my career has taken a few unexpected turns. Each of those has deepened my understanding of today's leaders' struggles and how to translate strategy into real and measurable outcomes. So to help our audience feel the gravity of this challenge, I usually use a very dramatic analogy because it really sticks.
Imagine yourself you're in a hospital and a surgeon walks out to a family and says, "The operation was successful, but unfortunately the patient didn't survive." How would you conclude this surgery? What would you say to the surgeon? Would you give him kudos? Would you say good job, or what would you say? He thinks his surgery was...
Kumar: Yeah, that's a pretty stark visual metaphor. Yeah, that would be a total failure.
Sevak: And unfortunately, that happens today in many organizations. Our leaders start celebrating their activities, their efforts, the success of execution of the plan. What they miss is that actually the outcome expectation for business needs. So strategy looks great on paper, but it dies in implementation because there is not much alignment around the real impact of it.
Kumar: You mentioned alignment. So I mean, I think that's a key word. And I wonder how important that is and how you achieve that alignment between the people that create the strategy and the people that actually execute the strategy.
Sevak: Great question again. So that's where the OKR, a good OKR, can bridge that gap. So OKR is not just another framework. It actually is a discipline that helps us to refocus back to teams and what truly matters. So a good OKR can help to clarify the outcomes, create transparency across the teams, and give leaders a continuous feedback loop to adapt and continue to shift the project or the outcome.
Kumar: So, yes, please. You know, I don't know how many of the viewers know what OKRs are, so maybe it would be good to just give a brief primer. What does it stand for? Where did it come from? You know, you say it's more than a framework. So can you sort of elaborate on that a little bit?
Sevak: Great. Great. So OKR stands for objectives and key results, and it became very popular when Google started to bring... They were doing a lot of great work in the nineties, but they were not really moving the needle. So they brought a person who wrote the book What Matters Most, John Doerr, to help them to make their goals more tangible and more impactful. So when he came over, he started to look at a lot of work that was focused on the goals but not on objectives.
So to give our audience an understanding about the difference of the goals and objectives, I usually use this example. Companies usually put their goals such as "build a fuel pump prototype," right? Sometimes they even go further. They say "reduce fuel pump weight by ten percent by creating that prototype." John Doerr just translated the goals to something meaningful and purposeful where it touches the customer. For example, it says "reduce fuel cost concentration by developing a light fuel pump prototype." Now the increasing or decreasing is the purpose, and another analogy I use for changing the purpose to the need... That's a great conversation I had the other day with one of my teams. They said, "Sevak, we have goals. Why do we need OKRs?"
And I asked them, "What is the goal?" He says, "Finish the job. That's the goal. Finish all the user stories or tasks, right?" And my question was, "Why do we need to do that? Is that a need?" He goes, "Yeah, if we do not finish those, we will not get our promotions. We will not get our bonuses. We will lose our work." Now I ask this question: is that a need? And the answer was, "Yes, that's the need." And I said there's a difference between purpose and need. And he asked me to elaborate on that one.
And I said, so another analogy: there is a seventy-four percent chance a deer runs away from a lion. Kumar, do you know why that is?
Kumar: Why? I'm surprised it's only seventy-four percent.
Sevak: Yeah, there is a seventy-four percent... The reason is the lion hunts for the need, but the deer runs for the purpose. So where I was going with this is if you make the work purposeful for your employee, there is more statistically a higher chance you will succeed.
Kumar: Yeah, I like the way you put that. And also it helps me frame the word alignment a little better in context of how the doers, if you will, the people that execute the strategy, if they're aligned to the purpose, they can operate more or less autonomously because they know where they're going. They know the destination. Do you agree with that?
Sevak: Absolutely. So that's why a good OKR is called bidirectional OKRs. Usually goals are top-down. The executives put the goals for you from the top and you're supposed to deliver those. Sometimes you do your personal goals, but not necessarily the goals for the organization or your team. So bidirectional OKRs is while executives put the OKRs to you, the teams create their own OKRs. For example, by doing this prototype, I think we could achieve ten percent, or maybe nine percent of maybe not the weight, right? Maybe we do some optimization in regards to the process time.
Now, where this alignment comes in, Kumar, is the executives thought with their knowledge they thought they could reduce the weight and achieve the goal. But the team thinks we could improve the time of the process to achieve that goal. Now, when you bring those two OKRs together, a good discussion happens here. Is this really the weight or is it the process time? And that's where alignment comes in.
Kumar: Yeah, makes a lot of sense. And so this is... I like the way you put that: a bidirectional system, right? So it's not just a top-down set of goals like "go do this" with maybe a limited understanding of the why behind it. There are conversations that happen when you create OKRs, the objectives with key results, conversations at every level of the organization. And ideally these OKRs cascade, but they're not tightly coupled. They're more loosely coupled. So the people that are doing the work have autonomy in how they do that work in order to meet the goal. Do you see that? I'm sure you coach to that in some way, right?
Sevak: Yes. And guess what autonomy brings in? Innovation. Let the team talk about how they could achieve those OKRs in their understanding. And guess what? The team knows much better than executives about the product. So let the team innovate to bring the results and bring that OKR at the end. And one thing that you mentioned, Kumar, you said OKRs are loosely coupled. I would say they're highly aligned. So in order to make them loosely coupled but highly aligned, I use three alignments: strategic alignment, tactical alignment, and cultural alignment.
Strategic alignment is that OKR that makes the executives to think about the purpose of the organization, the vision of the company, and to align the OKR to purpose and vision, not to just make the arbitrary goals such as "we should do ten percent increase this year." No, that's not a strategic alignment. That's just an arbitrary goal. So strategic alignment, I create that in the OKR.
Then tactical alignment. Tactical alignment means once we agree on the OKR, then we say, "Okay, what are the actions we should do on day-to-day basis to help us to achieve that OKR?" So the tactical alignment is how do we align actions to OKR? So actions are very, very small things. For example, if I wanted to lose weight, I would say I reduce my calorie intake. Those are little, little actions, right? Or I increase my exercise. Those are little, little actions that will bring that key result I want, I would say, "lose ten pounds," let's say, right? So those are tactical alignment.
Then there is cultural alignment. And cultural alignment is bringing all of those together to share and collaborate across the organization. If we don't have that transparency, everybody could come up with their own OKR. They will not agree on anything. So we want to have cultural alignment that talks about, "What is good for all of us?" Because at the end of the day, organization is just like a football team or any kind of a team. If we don't bring success to team, we can bring success to ourself only, right?
Kumar: Yeah, I love it. So let me dig in a little bit, if that's okay. So the strategic alignment... I've worked with many, many clients in my own career, and this notion of strategy is something that sometimes gets confused. And you had mentioned something about "increase revenue by ten percent." That's more of a goal, that's more of an outcome metric maybe, but not truly a strategy. So what is strategy in the context of this strategic alignment?
Sevak: So strategy usually is the big picture. I use it as a horizon. When you are planning, you have a three-year plan. So your three-year plan is you say, "Okay, our vision is to do this." So your strategy should align to your purpose and vision, and that's where OKR comes in. So OKR is good for three months to one year. So it helps you to focus on the actions that required to bring you to strategy. So what makes a good strategy? A good strategy, based on a good book, the book's name is Good Strategy, Bad Strategy. I highly recommend that one. He talks about each strategy is like a hypothesis. When you create a strategy, your hypothesis is, "If I do A, B, C, D, then I will achieve X."
For example, Walmart's strategy was, "I bring, I reduce my inventory, and I create an atmosphere that people could come and buy everything in one place." Because the stores were very selective, they would say, "Okay, go here for meat, go there for grocery, go there for clothes." They bring everything in one, and that reduced their inventory cost a lot. So Walmart came out of strategy saying, "If I do this inventory and everything in one place, people will buy from me because it's easier." And they tested their strategy. They tested their hypothesis. Right. And if they succeeded, they go forward. If they didn't, they would have to shift. So each strategy is based on the outcomes of small wins on you. Those are OKRs. Those are OKRs to tell you that your strategy is working or not working.
Kumar: So a strategy is sort of like... If I were to extrapolate a little bit from what you just said, it's like a hypothesis backed by some assumptions that we believe to be true. And maybe also some data and facts that we think we know to be true. And we're saying that if we execute this plan, if we do these actions, we believe we'll get a certain set of results. And those results will take us towards our vision, towards the objectives of the company. And it's an experiment, right? It's a hypothesis. We believe it to be true, but we may be wrong.
Sevak: Correct. Correct. And you said something very important. You said small action. You see, most of the strategies fail, they fail from here. We usually use big bets, one big bet, that they hope they succeed. Instead, if you use small bets, you have multiple chances. So you say, "Okay, I can test this and see if this is working," instead of "I will put all my eggs in one basket." So in strategy, you should have multiple OKRs that tests the hypothesis. All of them...
I think Albert Einstein, he has an equation. He says three to the power of two. So which means in order to bring a successful strategy, you need to have three OKRs. Each OKR should have three key results. Each key results should have two actions. So multiple actions bring you to see if your hypothesis is valid. If you put everything on one and it didn't work, well, guess what? You just spent a whole year on something that didn't work. So multiple smaller bets.
Kumar: Yeah, I love that. And what I heard you say also is that it's important to have metrics associated with your OKRs, your objectives and key results so that you can assess if it's working. Because if you have a hypothesis and you don't know whether it's working or not, it can be really difficult. So I really like that aspect of it.
There's a quote I heard from somewhere. I can't remember the source, but they said they'd rather be directionally accurate than precisely incorrect. So sometimes in organizations we see them measuring things very, very precisely, but they're so far off from what really matters. So I think you're sort of speaking to that a little bit here.
Sevak: Absolutely. You see, the problem with most of the OKRs that people are doing right now, I heard one client in the manufacturing, they said, "Yeah, we had OKR and it didn't work. So we went to Agile, and Agile didn't work. And then we went to this and that, and nothing worked." And I started to talk to them, I said, "Tell me about your OKRs." So then they started to pull up their objectives. I'm like, "Okay, those are goals." They said, "What do you mean?" I said, "Yeah, you are basically creating the goals. What is the outcome of this?" And they don't have outcomes associated with their goals. They don't have outcome. So I said, "Here, where is the key result? Where is the objective?" They said, "Yeah, the objective was building this and building that."
And then we have to translate that to the outcome of it. So if we do build something, what is happening? Is customer going to get this? Are we going to reduce something? Those are the key results. Key results are basically saying, "If we do that, this is what we expect to happen." So if you do not know what to expect, how you can measure yourself and see if the actions that you do are productive or not?
Kumar: Yeah, absolutely. So you've worked, right, with many Fortune 100, Fortune 500 companies over the years. You've been in this business for quite some time. What are some of the common mistakes that you see folks making when they're trying to execute on something like OKRs or they're trying to get this strategic alignment and tactical alignment and cultural alignment that you talked about? What are some of the pitfalls?
Sevak: Very good question. So I see many mistakes. I think we can make a podcast just on the mistakes, but I will name a few big ones. So one of them is they try to put OKRs bottom-up. You know, the top executive says, "Oh, we want to have OKR culture," and they hand it down to the team and saying, "Hey, go and create OKRs." And nobody from the top is participating in this. So that dies immediately because a good OKR culture is bidirectional. It's bottom-up and top-down. So a good OKR should start with the board and CEO, right at the top. So that's the one thing.
Another thing is people create too many OKRs and they overwhelm themselves. As I said, the three to the power of two equation is really good. Say three OKRs, three key results, two actions. So you end up with around, let's say, around fifteen actions approximately. Eighteen actions. This is good number. This is within the number that you could handle. So if you have fifty OKRs, twenty OKRs, how you can handle everything? Just forget about them. You only can focus on a few of them.
Kumar: Yeah, I love that. It's like that whole notion of focus, right? The sun doesn't burn paper. A magnifying glass doesn't burn paper on its own. It's when you put the two together, you focus that light through the magnifying glass on a single point, and then you get intensity.
Sevak: That's a good analogy.
Kumar: Yeah. So I guess the question is, maybe to bring it down a layer or two, what are the key results, right? So if you were to be an organization and you want to implement, let's say, OKRs across the organization, how do you start measuring for success? Like, what are the metrics? We're in the age of AI now. You had mentioned before we started the podcast about how AI can help us with OKRs, maybe help us generate some of the metrics. But I'm curious what your thoughts are on the metrics side specifically.
Sevak: Very good question. So there are two type of OKRs actually. There is a committed OKR, there is an aspirational OKR. The aspirational OKR is more strategic and the committed OKR is more tactical. So the one that we focused mostly on the tactical one, the committed OKR. There are metrics you could measure based on the outcome you want. So for example, I would recommend to use something such as DORA metrics or outcome-based metrics. What is DORA metrics? DORA metrics is for technology mostly, but you could use in manufacturing and so on. It says how fast you do your deployment. So if you have a good system, you should be able to deploy faster. That's how you know you are doing right. Not how many user stories you are completing or how many story points you do, because story points could be game-able, right? We could make a story point more and say we are doing more.
But at the end, the outcome-based is what matters. So if you are not delivering it to the customer, if the customer experience is not changing, if the revenue is not changing, those are the outcome. So come up with the outcome-based metrics. Another one, for example, if you build something and you deliver to production, there is a feedback mechanism. How much of your production, when you release something, introduces defects? Or how much of your build actually succeed to go to the production without issues? Those are the ones. Or the other things such as mean time to restore and such things. Because if you introduce something that broke and you immediately can recover fast, that's much better value and that's good efficiency and adaptability. So these are the metrics I measure myself with.
Kumar: Yeah, wonderful. And for those who don't know, Adam, I'm sorry, DORA, what does DORA stand for?
Sevak: DORA stands for DevOps Research and Assessment. So it's basically DORA metrics. They came up with these to measure how healthy your engineering is or how healthy your systems are. And it's not only for software. People mostly use it for software, but you could use it for other businesses too.
Kumar: Right, right. So one of the things you had mentioned before we started the show, which I thought was really interesting, is how AI can be leveraged in some way. Can you talk about that a little bit?
Sevak: Yeah, this is actually very exciting and very new, what I'm working on right now. So nowadays the AI agents are a big thing. So I worked with one of my friends to do the agents and we thought about this OKR and strategic agents. So we develop some agents that actually help you to see the connections between your OKRs through AI by automatically scrapping information from your ALM tools. And then they go through and look at what kind of teams are working on what kind of OKR or actions and such, meaning like if I finish these epics, I will be able to achieve actions that are required for my OKRs. And by doing that, the status can automatically come in, and you could see the status of your actions for OKR through AI.
Kumar: Yeah, that's interesting. Yeah, there's a slight lag in the connection, just so apologies to people that are watching live. Hopefully it doesn't affect... I mean, I understood everything you were trying to say, so that's a good thing. So we're getting towards the top of the end of our time together, but I wanted to end the show with a few questions, lightning round questions, just to get your quick responses to some of these, maybe not so quick. We'll see.
So here's the first one. So if you could wave a magic wand and change one thing that would make organizational transformation happen faster, what would it be?
Sevak: Having leaders more engaged into transformation and being not only output-driven but also outcome-driven, because transformation takes time and requires a lot of attention.
Kumar: Wonderful. So for a mid-level manager caught between executive strategy and the team's reality, what's the one thing they should do tomorrow to start bridging that gap?
Sevak: Middle managers are usually the ones that are stuck in their transformation because they're not doers and they're not askers. So what I would say as a middle manager, try to get in there, understand what is happening. Don't be successful in your role. Be valuable in your role. Now, is it to management, to whatever you do, but just don't be a person in the middle. Try to add value to transformation.
Kumar: Yeah, I love that. When you work with a Fortune 100, Fortune 500 client, doesn't matter, what's the first metric or signal that tells you the transformation is actually working?
Sevak: So I usually use the concept of DORA metrics, which is mostly outcome-driven. Find out what the end product is supposed to help you, and you measure that one. Don't be lost in the small data such as am I getting burndown of user stories, right? Because as Adam said, you could easily game on. So make sure that those measurements are good for the team, but at the end you measure yourself to objectives. Is your customer really improving the experience? Is your fuel tank really being efficient with fifteen percent, regardless if it's weight or process? So focus on the outcome and not the output.
Kumar: Okay, wonderful. All right, one last question. Maybe the last question. With AI and rapid technological change, how do you see the role of organizational development evolving? What skills are most important for leaders in the next, say, three to five years?
Sevak: We already see that, right? So a lot of processes are changing. Repeatable work is moving more to AI. Now you have to think about how do I create a more innovative organization? The companies and organizations who are going to win in the next coming five years are not the ones that are going to keep doing the same thing. It is the ones that are going to be innovating. I use this: AI is not asking you to do different things. AI is asking you to do things differently. So as a leader, product manager, whatever from organization, and even organization design working with humans, you have to think about what am I going to do differently than our competitors? That's what's key.
Kumar: Love it. All right, one last one, and then I'm going to turn it to you. I love this question. So I enjoyed it, and I learned that also, so please go ahead.
Kumar: That's awesome. So what's harder to solve: a complex engineering physics problem or getting a leadership team to actually collaborate? And why?
Sevak: Oh, definitely. So here I did a lot of... I also work in reactors with the nucleus, and I did a lot of virology and such in my past. I have two publications on similar viruses to COVID. Guess what? Those were much, much easier, Kumar, because there is an answer at the end. You know, the virus looks like what the virus is. Yes, it's hard to get there, but the virus is a virus. Now, the thing is with humans, everything changes. The mindset changes, the state of mind changes. So it's dynamic work.
And then if you do the collaboration and if you do all the impacts... And that's what I said: VUCA environment. The leaders tomorrow, I want to go back to the last thing, the leaders tomorrow are not the leaders who are good, they know their stuff. The leaders tomorrow, the successful leaders tomorrow, are the leaders who are adaptive leaders. They can adapt quickly. And the same thing with the employees. AI is going to change. If you're not an adaptive employee, I think you're going to have a hard time finding a job in ten years from now.
Kumar: Totally agree. Do you have any questions for me before we end our show?
Sevak: Well, I would like to... Let's see. So you have been working with various leadership challenges, and you have been working with also several high-tech transformation companies, right? So tell me, how do you deal with a leader who doesn't want to transform? They are happy with where they are because what they did, they're going to retire in the next coming few years.
Kumar: Yeah, that's a tough one for sure. I certainly have been there. I think what you have to do, at least in my experience, you have to find what their triggers are, you know, what they may be willing to change, right, rather than unwilling to change. So there's resistance. And when you sense resistance, maybe there is a lever you just haven't explored yet, a question you haven't asked yet to understand where they are in their work, in their profession, and in the things that they like about their work or don't like about their work.
So I take a very Socratic approach to understanding and developing a relationship with leaders and companies that I work with, just to understand where they are. What is it that they're resisting? And what is it that they fear about the change that they see happening around them? Doesn't always work because sometimes people are just who they are, right? They are further away from that willingness to change. They're close to retirement, whatever it might be. But that's what I do.
Sevak: Very good. Thank you.
Kumar: Anything that I didn't ask you that you'd like to share?
Sevak: I think this was a great conversation. I really appreciate the opportunity to share my thoughts. You know, honestly, these topics that you have, Kumar, are required hours and hours of conversations. There's nothing that you could squeeze in a short time. Each of us has many, many years of experience dealing with different situations. But I hope that the audience and yourself got what is most important and why OKR is so crucial for any organization to change from need to purpose, to go from mindset to state of mind, and to do something good in this world, not just keeping yourself busy.
Kumar: Yeah, I agree. And for those of you who are still watching, thanks for sticking around to the end of the show. And for those of you that asked some questions, thank you for those. I hope you enjoyed the show. And you're right, there's so much more we can explore, and I'd love to have you back maybe on some of the nuances that we didn't cover because I had a whole bunch of questions that I didn't ask. But we'll have to talk about maybe having you back on and taking a different angle to these topics.
So thanks again for coming on the show, and I appreciate you being here. Thanks for all of you who watched. And if you want to get a hold of Sevak to get some counsel or just have a conversation, his details are in the show notes. He's written a bunch of publications, so please look him up. And you can always contact me as well for anything that you need. Thanks a lot again for watching. See you all next week.
Sevak: It was a pleasure. Bye.
Kumar: Bye.
