Agile Strategy - A conversation between Roger L. Martin and Sohrab Salimi
Sohrab: okay welcome everyone to our next session of the agile insights conversation today. I'm honored again, after I was able to host him last year. Roger Martin is joining us again for those of you who don't know Roger. Roger is the author of many books. That's the way I got to know him. Of course he's done a ton of other interesting stuff in his life and he can tell us more about that a bit later. But he wrote Playing to Win which was one of the main, maybe most defining books on strategy in general, where he also laid out his framework for creating strategy. We will deep dive into this today last year I spoke to him about his book When more is not Better and this book was really interesting because it also, especially for those of you working in agile ways of working, he laid out the mechanistic view or the view of on the economy and on organizations as a machine and thought about like we need to look at organizations differently because they're living organisms and there's a lot of insights in that book. And I think Roger if i'm not mistaken, if i did my homework right, this year you're publishing a new book and it's called a New way to Think.
Roger: That is true.
Sohrab: Which I'm really really curious to read about later and maybe you can give us a bit of thoughts around that throughout this conversation. But before we get this conversation going I introduced you as the author of these three books, maybe you can share with our audience a bit more about yourself and what you've done in the past. Especially connected to the topic of strategy which we will deep dive in today.
00:01:33,440
Roger: Sure well um I guess, I got interested in strategy right from the time I was at business school. And I happened to go to business school at Harvard at the time Mike Porter, a friend and colleague, was publishing competitive strategy and he became a superstar for doing that. But he always wanted to be an academic first and foremost and so when companies called them up and said: “Mike I read your book and I'd like to do this competitive strategy stuff, he didn't really have capacity for it. So we assembled this group around him that could just essentially deal with the demand for his new way of thinking about strategy. So we built a firm called Monitor Company which I did for a decade and a half with a bunch of colleagues and then I got talked into being the dean of the business school in my home country. You can tell by my Canadian accent that I am a Canadian born and the President of the University of Toronto convinced me that I should give up my high paying job in Boston, come back to Toronto and run the Rodman School of management at the University of Toronto. So I did that for fifteen years. But I continued working on and writing about strategy. Including Playing which was written during that period when I was a dean.
Right toward the end of the period, I think it was in 2013, I retired from academia and do the two things I love most. Which is write and advise Senior Executives on strategy. So that's me. I am a strategy geek. I love strategy and I love writing about it. I try to take a little bit more of a holistic view of it and that's why When moore is not Better is not a strategy book. It is, but it's a book that relates to how one needs to think about strategy I think to be effective. But it also applied Strategy to the economy as in general.
00:03:47,778
Sohrab: Yeah, absolutely. So when I was reading When more is not Better I absolutely saw the link to strategy. Especially when you think about strategy also needing to consider for responsibility that we as organizations and that as individuals have and bringing this into our thinking. Because in the past, everything was about more and bigger, but you can also differentiate and we will talk about strategy more today.
Roger: Yeah
Sohrab: Differentiation is one key aspect of this when you don't aim for more but when you aim for better in different ways so…
Roger: Absolutely
Sohrab: I consider When more is not Better also as a strategy book, especially for CEOs. In that case now you mentioned that.
Roger: Yeah, but if I can just say, if I can just say you know it. I would hope one of the takeaways from When more is not better is to think about sustainability in a holistic way. Is what we're doing sustainable? Now, obviously environmental sustainability is a super important piece of that, but it's not the only element of sustainability. And you know for what it's worth, Imean I think one of the things that I find attractive about agile, is I think embedded in agile, it is a sense of sustainability.
So, if we're talking about being at war with our clients and talking about contracting, so that we can get a favorable deal with clients. That doesn't create a sustainable relationship with a client. And if you make people work in a way that creates all sorts of useless rework that's not sustainable either.
So again. I'm increasingly thinking about “how can you think holistically about what you're doing? In a way that you can keep doing it for the benefit of all the players in the system, rather than some? People would say strategies about how you get advantage and then put the pedal to the metal and take the greatest amount of advantage for you of whatever advantage it is. And that, in my view, is not sustainable and I'm seeing companies that I think are making themselves not sustainable these days, by not thinking holistically about what is not working for everyone.
00:06:19,618
Sohrab: Absolutely, so taking a full stakeholder perspective and not only a shareholder perspective is just one example. You mentioned today, you're doing the two things that you like most which is writing and advising other people.
Roger: Yes.
Sohrab: As part of writing, just to put it out for everyone. Every week Roger publishes an article on the topic of strategy at Medium.
Roger: Yes.
Sohrab: One of those articles was about agile and playing to win or agile and strategy development based on your own framework. This is what I want to explore. And in our show notes both on YouTube and later in a Podcast you will have the links related to the articles that are referred to here but if you're curious right now just go to google type in roger martin and medium you will find his page and all the articles are out there. It's really good. Now before we talk about agile and strategy, I want to get your perspective Roger on strategy in general. Including like, what are some of the biggest myths out there and how you think about strategy. From my experience going through a lot of literature around this topic working at Bain myself for three years, your perspective is unique and different to what many organizations do and think about strategy.
00:07:38,720
Roger: Well I'm a little bit bearish on strategy these days. In fact one of those medium posts was The Lost art of Strategy. I really feel that it has become a lost art and it has been replaced by a very bureaucratic technocratic thing called strategic planning. I am a lifelong strategy guy. I love strategy but I generally hate strategic planning. It's because strategy has become an exercise of defining a bunch of initiatives you know we're going to do. We're going to do that, we're going to build a factor, we're going to expand this, we're going to hire more people. Whatever it is and that's the perception of strategy and it's a very sort of technocratic view. It's sort of like: “okay so, Rob, you're the head of manufacturing, here's what we're going to put in the strategic plan that'll make you happy. And there's Bill. You're the Head of Marketing. We're going to put this in the strategy. In the strategic plan. And people, what's our Strategic plan?”
For me, to a great extent, it's a budget with pros. Put the right budgets in our numbers and then you add at pros and you call it a strategic plan.
My view is rather that strategy is about choice. It's about making a set of choices that are integrated together to position you on a playing field of your choice. The place on your playing field in a way that you can deliver the most superior value equation in that space.
Which is what I think of as winning. And people will say, well, we don't have time to do strategy and I say sorry, you can't not do strategy. What you do is your strategy! Not what you say! So in my experience it is relatively rare that a company's strategy is similar to its strategic plan because its strategic plan will say: “oh we are going to be the best in our industry and the most innovative company and blah blah blah. And then they're not spending anything on R&D. They're not listening to their customers and whatever. That's their strategy, right? The strategy of many organizations is to punish customers as much as they can. Punish them without them leaving. They're making a set of choices like you know. You know, you want to return those goods. Oh wow, then we're going to make that super hard for you! That's their strategy. Because whatever the customer feels that is what the company is doing. And so what I try to do when I'm advising people is to say let's reverse engineer what our strategy is today. Like what must it be today that caused us to do the things we're currently doing?
And then how can we migrate that to one that is kind of better, more competitively, sustainable in that journey of making choices. Making a real set of choices where you can write down the choices and say these are the choices. That we are making by our actions. That is what strategy is.
00:11:03,938
Sohrab: That is what strategy is. What you do, not what you say or what you put on a poster. And in many cases now you mentioned that for a lot of organizations, and I see this today maybe more than it was in the past. You've been around a bit longer and maybe you saw that already in the past.
Roger: Yeah.
Sohrab: But due to the fast pace of change, a lot of organizations speeds startups being organizations and high growth industries and I know, I read your post today about high growth industries which is always like just a temporary thing not a permanent thing, right?
But also people in larger organizations they usually, or not usually but many of them, don't take the time to really think through a strategy and either they don't develop a good strategy or they have someone else develop it for them, right?
Roger: Yes.
Sohrab: Now if an executive comes to you, if you're in a conversation with one and they say no I don't have time to do this. What is your response to that?
Roger: My response is, no you are doing it. And they were like: “what do you mean?” And I said: “You're, I believe that you're the CEO. And that Chief Executive Officer means a whole lot of choices come to you. And you make those choices, am I not correct?” And they were like “yeah, but that's not strategy” and I say “No it is! It's deciding what to do and what not to do. And last time I checked, you know, you can't decide to do everything. You have to say yes to some things and no to some other things. So if you want I said I can just sit in your office for a week watching you and then I'll tell you what your strategy is.”
So you have only one question. For you as CEO the only question is how thoughtful you want to be about those decisions. And if you want to make each one of them, kind of on their own, as if they were sort of independent from the others, I mean you can do that.
In my experience, that doesn't make for great results. It means a little bit of this, a little bit of that. And somebody else is figuring out exactly how they want to beat you!
So my suggestion is that we think about what those choices have produced for you.
That's a gap between what you wish for, and what is actually happening. Let's just identify that. What about where we're landing now? Is our relationship with our customers growing? Or is it shrinking? How are we doing against competitors? Is our profitability lower or higher? Whatever problem we've got, now which is a gap between results and aspirations?
Why don't we dive in on that and ask: “could we make a set of choices that would make that problem go away?”
So in some sense, I think people have this notion of what is strategy. It's getting everybody together for all sorts of long meetings in order to come up with a strategy plan, right? And they say iIdon't have time to add all those activities into my day, plus it's not very useful. Which is true, so I say our strategy is just a problem solving technique. You tell me your biggest problem that you're going to work on anyway?
That is going to be solved by making a different set of choices because you don't want to be insane right? Insanity is doing the same thing and assuming that the results are going to be different. So if we don't want to be insane we need to make a different set of choices and let's work on that. And so in some sense that turned strategy from something where: “here's my daily job, and i've got to go over there and do strategy to come back to my daily job.”
Isn't this your daily job? We're going to do your daily job in a way that results in something that is more positive and productive and hopefully makes the problem that you've identified go away. So that's how I deal with this “I don't have time for strategy” kind of thing.
00:15:34,738
Sohrab: I don't have time for strategy, yeah. I mean looking at this to be very honest as the CEO of a small company I'm constantly in this battle, right? How much time do I spend on actively doing strategy? And if I don’t spend that time we're still doing strategy. It's just maybe not that conscious, right? And maybe we're not identifying that key challenge and maybe we're not aligning all of our daily activities to address those key challenges.
So when I go through your books but also through your posts, which are super valuable. It's like: “okay, I need to take a step back, I need to make that investment in time, to ensure that my daily actions that I’m taking anyway but that I make better choices on a daily basis to achieve the aspiration that we set out to achieve as an organization.
So in addition to what you've already mentioned, the where to play and the how to win. Prior to that this aspiration is also important and I remember, you wrote one post where you talk about transformations and I want to go into that a bit later in this conversation.
In the post where you emphasize the importance of that aspiration. How meaningful it is to have a true aspiration, not one of those lofty ones. But one that is tangible and that is achievable in the midterm and then align everything that you do on a daily basis towards that? So there is no disconnect between strategic work and our daily operations. Where our daily operations are strategic work. That's at least what I take. Or am I interpreting too much?
00:17:16,800
Roger: No, no not you're a very good kind of reader and student. So absolutely right.
I would just add to what you said that's why I think it's really important to toggle back and forth between your aspirations and where to play, how to win a choice because again many aspirations, as you have alluded, aren't helpful to the organization at all because they're unrealistic. It's like well we're going to be the best company, the most innovative company in the industry.
You know I'm from Canada originally and practiced consulting in Canada for a while because they opened a Monitor Office in Toronto. So many Canadian companies that were national, not global, would say: “well we're going to be the most innovative company in our industry”.
So okay, Canada's population is 33 million and you only sell in Canada. How is it that you're going to fund R&D, that would match one of your US competitors with a home market of 330 million or some European EU competitor that's got a market of 500 million people? That's just made up.
You're not going to be the most innovative company. You could, if you decided to serve the world. But you haven't. And so your “where to play” is utterly inconsistent with your aspiration. So you sort of have to toggle back. I say don't start with an offsite. The wordsmith of your vision mission or aspiration. But start up and say “what would we love to become”? Let's table that for a minute and then ask if there is a where to play, how to win, when that is doable for us. Hard but doable for us that would make that aspiration come true? And if the answer is no, then try another aspiration, right? It's just back and forth and that's one thing that again the technocrats who dominate strategy are far too linear. It's like having an off-site to decide the aspiration. Check, we've done that. Now go and think about where we're going to play. Okay, check. Good. Now, how can we win there? And all that does is create constraints along the way. That makes it even less likely that you're going to be able to come up with something matching that works. And so I'm much more into sort of this, you got to circulate around and be more iterative in your strategy. Think about an aspiration, see if there's a where to play, how to win, that works there. Revisit the aspiration. Okay I think we've got a place to play, how to win. That's kind of the last two boxes: capabilities and management systems. Can we really build the capabilities that are necessary to win there? And if we can't, then we said, well, how could we modify the how to win? What does that mean for where to play, what does that mean for the aspiration?
And just go back and forth. A lot of people just say “oh that's failure”. You have to go one, two, three, four ,five along the way. It's not a strategy. Again, the key thing about strategy is, it's a holistic set of choices, it’s not a strategic plan where you've got a list of choices that don't relate to one another and don't add up. You have to have a set of choices that work together.
00:20:41,618
Sohrab: And what I always liked. So, you referred to this back and forth, right, this iterative approach. I always also looked at those things as feedback loops. So when you have this aspiration, now you decide on where to play, you decide on how to win and so on and so forth. And maybe, you make a decision but based on that decision you move into action.
So you might decide to run some experiments, to assess whether that where to play is applicable.
Roger: Yes.
Sohrab: And whether the how to win is actually a winning strategy.
Roger: Yes.
Sohrab: And when you realize, that it's not. You circle back and you reassess where to play, you reassess your big aspiration and based on this for me this in itself is very well aligned. With what we refer to an incremental and iterative approach to product development particularly but also organizational development in the agile space. Now, when we think about agility what I see happen a lot is that organizations say “wow everything's moving so fast we cannot make a longer term decision on strategy et cetera. Now I would be interested in two things. One is when you think about longer term or midterm and for strategy, what is the time frame that you usually look at and that might be different from industry to industry and I would be interested in your perspective on this. And the second thing is getting into the core about how do you connect your play to win framework with agility, with the key principles, not a specific framework like scrum. But with the key principles and values in agile ways of working.
00:22:23,120
Roger: Sure, on the first question of how long people saying “well things are moving so fast I can't make long term decisions”.
One that's just factually not true, right? Lots of the decisions companies make are long term decisions, whether they know it at the time or not.
You know, if they decide to install SAP as their ERP instead of Salesforce. They go for a license product. That's a long term decision. That's a decision for like ten, wouldn't you say probably ten years minimum. They're not going to rip out SAP or rip out Salesforce after having installed it. So one message I say to you is: “just don't don't fool yourself into thinking somehow the decisions you're making do not have a long term kind of impact!
So you know, so recognize that but to the extent that your world is, you think of it as VUCA, you know it's volatile, uncertain, complex, rather ambiguous.
You need to think about how I can modify it? Is this strategy right? How is this a choice that we're making, that is going to last ten years whether we like it or not? Or could we make that choice in a way that leaves us optionality to iterate, get feedback and modify it and so I think that's a part of good strategy practice. It’s having that sense of how have I left myself the ability to iterate or have I boxed myself in? That having been said I would argue that all good strategies make bets.
They do box you in, they do say we're gonna do this, not not that.
Sohrab: Because that's the definition of a choice.
Roger: Good. Got going again and said we are not gonna sell licenses, that boxed them in. If you can't make the benefit of software as a service, profoundly manifested for clients, it doesn't matter what you think. It's what the clients say. Then you've made a choice that is going to be a terrible choice for you. And you do make choices, and that was a fast moving kind of industry, saas is not sort of some stable industry, so you've got to make those choices but think about what ones you have to make that are irreversible.
And try to make those in a way that's consistent with your environment and in terms of time frame. You asked specifically about the time frame and then you alluded to maybe it's different by industry. Yes you are, I believe, correct.
Sort of the fundamental way I think about strategy and time is that you have to have a strategy that will last long enough.
And will serve you long enough to pay back on the irreversible assets you put in place. So if you're building a copper mine from scratch, chances are you won't get copper out of that mine for at least twenty years.
Sohrab: A long time.
Roger: Yes, from the time you say we're gonna do it, you gotta go and negotiate with the guy because they're mainly in kind of latin american and asian countries. You are going to go negotiate with them, you often have to put in all sorts of infrastructure etc. And if you're lucky maybe fifteen years later, but probably more like 20 years later you will get copper. Well, you know that, whether the world is VUCA or not, you know you're going to have to invest billions and billions of dollars. Maybe ten billion dollars at least five, probably more like ten billion dollars to have copper 20 years from now to 50 years from now.
You have to have a belief that regardless of the cycles of the copper price, the cost position of that mine, once it's built the way you want it to be built, it will be robust for any reasonable range of copper prices out twenty to fifty years from now. So if you build a copper mine that's at the highest end of the cost curve, chances are that's a bad strategy because if copper prices are toward the low end for any period of time, you'll be out of business. If that copper mine perhaps is the lowest cost kind of mine in the world when it comes online and you kind of know what the cost position of all the mines are right now and nobody else can build one less than twenty years, then if copper prices are attractive, then you'll make an unbelievable amount of money. If copper prices are not attractive, you will still make some money because somebody else is the marginal cost producer setting the margin price. But that's very different than starting up in an internet business or you write about some code you know to get out there and your assets if you make a year or five years of return on those assets before your competitive advantage goes away.
Then you'll be happy and you'll have earned a good return on it. So it kind of depends on how long your advantage has to last to pay for the investments. The irreversible investments that you need to make to deliver that. If you will do that where to play, how to win a strategy.
00:28:49,618
Sohrab: So, you always have to look at the barriers of entry for others and what kind of modes you basically can create in your specific industry?
Roger: Yes.
Sohrab: Thank you for that. Now taking this to the way that organizations want to work in the future, being more agile. And saying or even if they admit we're already making longer term decisions like installing or getting SAP or Salesforce or whatever. or maybe even deciding to build a certain product in a certain market.
Roger: Yep, that's a longer term.
Sohrab: A longer term strategic decision, but how can they make this? Maybe it's not even building a bridge between these two because of the way you frame strategy, creation and execution. It's so close to agile ways of working but what are the biggest gaps that you perceive people have in their minds and their heads when they think about on the one hand of building a strategy and on the other hand executing it in an agile manner?
Roger: Ithink the reason that it feels so to you is, that there's not some big gap between my view of strategy and the agile view of the world. I would argue that both my view of strategy and the development of agile were essentially anti technocratic movements. At least now, if you're the expert on agile, I am not, but here's my reading of agile. It was a reaction to
increasing technocracy in software development now I know it's broader now, but it sort of arose out of software development which is waterfall you can sort of plan this stuff all out. You can figure upfront exactly what's going to happen and figure exactly how to contract with the customer. And then the customer goes by and you just follow that technocratic plan.
Sohrab: Exactly.
Roger: Agile at least feels to me from reading it as a revolt against technocracy and to say: “no, no, no, it's this more organic thing you've got to iterate more how you treat the customer during the process is super important to an outcome! If you tell them to go by and we'll come back with the finished software you know is bad, you know just bad things are going to happen. But all of those, to me, are a sort of a technocratic view of the world. And that's in some sense what my view of strategy is a revolt against!
It's like no, you don't say to a customer “oh you want a strategy, okay great thanks Rob. I will see you in six months and I will deliver on a plate your strategy and to do that I have to engage in all these meetings where I meet with everybody and agree on. We are going to rank the initiatives and come up with the ten initiatives. And then we'll be done. There's no iteration, it's like, it's like you start from here you say “let's get everybody together round the table let's see what manufacturing wants. Let's see what Bill in marketing wants, let's see what Rush wants in the supply chain. Then we'll put them up on the board and we'll rank all of them and fight about them and come up with the list of the ten things we're going to do.
Then we're done with strategy. A very sort of technocratic approach so I guess I think in and right right then convert it into the budget and then get buy in. Tell everybody that this is great and they should agree with all of it. So that's why I think it feels to you, an agile guy.
What's the big difference? There's such an attitude similarity. I mean like I guess I think you know agile is generally it started in the world of how do you develop software it's now much more broad. It had a different purpose. My mind is how do you make those decisions about what the company does and does not do.They're in essence about somewhat different stuff but the mindset behind them is nearly identical.
00:33:36,098
Sohrab: It's very similar and I like how you put this right. That the agile movement is some kind of a revolt against that technocracy.
And in When more is not Better you talk about this mechanistic view on the economy and on organizations where you basically can determine if I do this, exactly this will happen. And it rarely ever does because things are so much more complex and I think in the development of software it is and it was certainly very similar.
That's why this movement happened.
Roger: And can I just say, sorry to interrupt, but you know the revolt to me, like in and When more not better. I mean I'm an economist by original training so I always say I can meet up with an economist because I'm not beating up on somebody else I'm beating on me there. There is such an arrogance to me in this technocratic approach. I talk about a great Nobel prize winner, an economist called Yanti of Russia, who came to America. But
He got the Nobel prize for the notion of input output models. He could take the US economy, break it into five hundred segment sectors and say here are the inputs into this and hear the outputs in others. And to have this sort of model for the US economy that could say I now know how this works. You know to me that's sort of a level of arrogance that I could never sort of presume. That you can figure out how this incredibly complex adaptive system is going to work and write equations about it.
I just can't make myself be that arrogant and sometimes people sort of say Roger like there is some inherent arrogance in your mind, you're saying we're going to make choices in this uncertain world and so you're saying that you can be absolutely right.
But that's a misunderstanding of what I am saying. I’m saying you could make choices. You can't not make choices and a bunch of them are exactly the same. Even not making the choice is a choice. We're not making a choice, it's a choice and what you say or what you end up doing might be wrong and you have to take that into account. But it's better to make a choice consciously and say this is what would have to be true for this to be a good choice.
And then watch and adjust if the things that would have to be true aren't being as true as you wish they were. Then adjust and adjust and adjust and adjust, which again, is consistent with agile where you just have to keep adjusting and I think people who are technocratic inclined and have never done that before. They find it scary and worrisome and painful but I think, again you're the agile expert, but I think in the world of strategy if I've convinced somebody that it's making bets. All you can do is shorten your odds of success rather than from ten to one if you got seven to four that's better.
And don't worry about needing to adjust, that's what happens. Then they get quite comfortable with it. It's sort of a mindset thing. If they think that that's an error, then they're
uncomfortable and don't want to ever admit an error. If they instead say this is going to happen then it's, you know, it's just no big deal and I assume that when you're kind of doing software development in an agile fashion, when the customer comes back and says well
I know that's sort of what we said we want, but now that we see it we'd like it to be a little bit different, the agile person is not saying “oh my god this is a disaster”. You know we got to go back to square one. They're like I believe life has just unfolded as life. It won't do whatI think is right.
00:38:17,938
Sohrab: Exactly. You expect the changes. So when I’m teaching agile ways of working to organizations that believe in the waterfall system one of the questions that I ask them is how often, if you look at a longer term project or product development initiative, how often do you get change requests? And they always say “oh we get to change requests all the time” and then I ask them do those change requests that come in, do they disrupt the plan that you've made at all? Obviously they disrupt the plan that we made, I guess. So when you have that defined approach to work, defined process control, that's the term that we use, you are not expecting changes to come. So every change that comes disrupts your plan.
Roger: Yeah as if it was a bad thing.
Sohrab: Whereas when you're working in an agile manner, if you believe in empiricism and
empirical process control, you're expecting those changes. So whenever they come you can embrace them.
Roger: Yes.
Sohrab: And you're already planning for them to come. You don't know what the change will be, but you know there will be a change.
Roger: Yeah.
Sohrab: I like the point that you brought up about some people being very arrogant. When predicting the future. You are admitting to yourself I cannot predict the future because I cannot.
It's not about spending more time and doing more detailed analysis. Because then
I'm trapped in analysis paralysis. That's also a choice. It's a choice of not moving forward. But it's about being okay with the information that I have and with the logic that I can put into play. I'll make a choice but I'll put in the feedback loops to evaluate whether my choice was a good one or not so good and then I move back and make a better choice. And if needed I'll move back again and I make a better choice because now based on the actions that I'm taking I'm getting more and more information.
Instead of just sitting there and trying to get a lot of information without acting and then hopefully coming up with a better decision. That's I think very much aligned with what you put in there and I read one of your articles. It was about one of the fallacies that people have in their heads when it's about analysis and that they don't see that there's two types of analysis, one is the long like the time taking cost intensive market evaluation and all of that. But you can do the other one, which is logical thinking.
So really sitting there thinking through your current business and your future business in a logical manner and based on that logic it can be false but making decisions and then evaluating them. So I love this piece because I think a lot of people in the agile space.
They move too much with the pendulum on the other side and they're like no we can't decide anything. We just have to do research. But that's also making a decision.
And you're not taking the time for that logic.
Roger: Yes. I love the point Sohrab. And yes, what I'd say about analysis is, we often think of analysis as a thing. I'd like to divide it into two things: it's a logical structure to which you apply data. That's what analysis is. So for instance, a logical structure might say bigger factories are lower cost because you can spread the fixed cost across each unit of production. That's a logical structure. There's a firm belief embedded in that. Now, if you wanted to have analysis you would then add data. You'd go and look at one hundred plants of varying size and say “oh, are the costs really lower in the bigger plants or not?” that's analysis.
What I say, as you said there, is in my view never ever under any circumstances an excuse not to have logic behind your choice.
But there are many excuses not to have lots of analysis. In fact virtually every entrepreneur that drove a successful entrepreneurial meet had very little analysis; they had a logical structure and they said all the established players aren't going to do this. But I'm going to do it and of course most of them are told you're going to go out of business right.
Sohrab: That's really scary.
Roger: And then, they just go. Do it and in some sense they produce the data that actually supports the fact that their logic was good. And so I think always having a logic behind your choices is key. Depending on your situation, like if you're a giant publicly traded company with shareholders and a board of directors, you can't do what the entrepreneur does and say we're going to spend two billion dollars because I think it's a good idea.
They're going to ask for more analysis which is more data now that's going to slow them down and make them more conservative. Which is one of the reasons why they're getting disrupted. The big companies are getting disrupted by little, little, little guys and gals in garages who just say here's my logic . I'm gonna go on the basis of that. So just as you said you can have this balance in agile of not even thinking about or let's just do. There's the same sort of thing in the world of business and the world of strategy.
In general, which is having an intelligent balanced view of how much data can we have? Do we have enough? And how much do we want to connect based on our situation?
What I try to tell all my clients which tend to be larger companies, like I deal with two kinds of companies, all of my investing is, I don't invest in public stock. Only partially. Because it's tiny startups. So I deal with a bunch of tiny startups and a bunch with really big Dow Jones 30 size companies.
And in the tiny companies I say “i'm not going to invest in you unless you can give me the logic” and they're typically kind of blown away, because they'll show me their financial spreadsheets and I typically don't even look at them. I say that's all made up. That's about the future. Who the hell knows that. What is your where to play, how to win? And I invest a hundred percent on that.
But I’m super tough on logic. Like if they can't explain the logic, the where to play, how to win logic then there is no chance. No matter what their financial projections look like. No matter how many successes they've had before. No matter whoever it is.
Brin in the other VCs. I'm inconsolable, you cannot, you can't do anything for me. But then, when I'm talking to the big companies I say there are waves of these little people in garages who are taking aim at you. Are gonna disrupt? So if you insist on everything being proven in advance, that's a recipe for you not doing anything.
It's a balance that you have to have. To strike between how much data and analytics you need to feel comfortable. And if you say we need comprehensive proof they're gonna get you. They are gonna get you, if you say “well, we need some analysis of that because we have to be kind of responsible.” I'll say okay then, you've got a shot of keeping those people from eating your lunch.
00:46:30,598
Sohrab: Absolutely. There's a great quote that just came up in my mind. I think it's attributed to Steve Jobs, and most organizations say “I have to see it to believe it”. I need to see the data to believe there's an opportunity in this space. Now who would have thought you can sell books online before Jeff Bezos did it. But Steve Jobs says “when it comes to innovation you need to believe it so that it becomes real.”
Roger: Yes.
Sohrab: That's the point. Belief is not only your gut feeling or your instinct. It's the logic behind it. Before it is seen to anybody out there you have to make a decision based on that logic. That you say “all right we're going to build this product. We're going to make this business model. We're going to make this of course in small and incremental steps, but ultimately you're acting primarily in the field of innovation. You don't want to be the second or third or fourth mover in that field.
You have to primarily act on the logic piece and yes, you can then feed it with information that you get along the way. That's a lot of these things.
00:47:42,880
Roger: I think you're absolutely right. If we then dive into that sort of things, a natural question from that is “where did that logic come from?”
Sohrab: Where does it come from?
Roger: I would argue that it comes from seeing something in one place and saying “ooh I can apply that in this other place it comes from. Also just snippets, non-comprehensive data. Steve Jobs famously heroes in calligraphy. And he basically said, well people are attracted to this. We've had calligraphy for forever and people really value them, when they get a wedding invitation that was hand calligraphed.
If you do that makes them happy. Now, what does that have to do with a computer screen? You could say nothing because that's a completely different situation. Or you could say…
Sohrab: Everything.
Roger: I'm gonna have it as a thing in my head. What you see is what you get. It's gonna look the way I want it to look and I'm gonna care a whole lot about that. Is that comprehensive data analytics? No, but it's taking a snippet from here and saying I think it works there. It's like the iMac, right? The colors that Steve Jobs chose for the iMac. You know that ring and the lime and the grape color you know where the hell did that come from? The answer is it didn't.
Sohrab: I still remember.
Roger: Come out. Yeah, well it came out of him having noticed 30 years earlier a camera company called Savoy coming out with a line of cameras, little 35 million cameras in a set of these wild colors and the logic for Jony Ive was these cameras are kind of interesting. They were all black or dark brown until this company came up with this sort of frame breaking look. Of a camera and computers they're all sort of gray.
Monotonous kinds of colors could do that over here. And what choices did he make on colors of the exact precise pantone of the cameras. So that's the kind of thing that you have to and I would argue you have to develop if you want to innovate is this ability to see things and see how in a different context some of the same rules would apply so for steve jobs the rules that apply to to calligraphy in the kind of designed kind of works of art would apply on a computer screen and that's how I think you need to develop your capacity for innovation to say what are some rules about how i think human beings behave that i could twist in a new way and have a new application. So it's not to me random at all but it's not deterministic either in a technocratic way right Jony Ive could have been wrong. The computers are so different from cameras that people are gonna say I hate this but they didn't they said this is so cool.
00:51:57,058
Sohrab: This is so cool. So instead of looking at data.
Roger: First thing Steve Jobs did when he got back was to develop the iMac. Steve Jobs himself rebuilt that completely. He said no you know when I got back. I went to see Jony Ive and asked what's cooking. Saw the models of the iMac in his office and so what the hell is that and he explained it. And apparently or maybe a powerful servant apparently Steve Jobs said production was how long it took him to decide that. They were going to put that into production as one of the first signature pieces. In essence the new apple. So in my view Steve Jobs had this great kind of logical processor that said here's how I think about customers, how they think about. Here's how I think about technology. Whatever that fits or or that doesn't fit. He killed a whole bunch of things that don't fit at all. But that one fits because he asked. I apparently asked why aren't we producing that? I said well, you know, your predecessor is not sure that this is I mean maybe could be too radical blah, blah, bla.
00:53:20,658
Sohrab: That's the decision that he made, so what I'm hearing from this is instead of just aiming to get like tons and tons of data you can also look at patterns.
Roger: Yes.
Sohrab: And you refer to them as rules of human behavior etc. That you can look at those patterns and based on the patterns that you see in similar industries or sometimes even very different industries come to strategic choices and then actions.
Roger: Yes.
Sohrab: For your own organization now with regards to time. I'd like to talk about two more topics. One is smaller and the other one is a bit bigger.
Roger: Sure.
Sohrab: The smaller one is: You also wrote an article about Strategy and OKRs.
Roger: Yes.
Sohrab: And that we shouldn't take OKRs. I'm bringing this up because a lot of organizations currently say “oh we are agile, we're doing OKRs”. We can discuss that, but at least it falls under the agile umbrella from my definition and you talk about that OKR and strategy are not the same thing.
Roger: Yes.
Sohrab: But before I basically rephrase what you wrote in your article there I'd like to give you the time and space to share a bit of your thoughts on OKRs and strategy and where they are different but also how they connect to each other.
00:54:34,080
Roger: Sure and interesting enough Sohrab, that is my most viewed post in the whole 73 post series and by far. It just just blew up when it went out. The article as as most of these medium pieces are reactions to things that I see going on in the customers that I do strategy with and what I observed is this pattern of setting OKRs and somehow assuming because you've set them it will happen. I mean literally. That will cause the thing to happen. The thing you would like to have happen or be more likely to happen and it simply doesn't. We said, we did this. We set all our OKRs and then people just didn't execute.
And the reason for me is that yes, you can set goals like the whole process of saying what are we trying to accomplish you know. How are we gonna measure that? This is all good, but you have to have logic behind why on earth you would believe that you're going to accomplish those things. And the only way you're going to accomplish your objectives is to have a where to play, how to win that is consistent with it. If I call objective sort of a form of a winning aspiration, that winning aspiration has to be twinned with a where to play, how to win that is doable, sensible or it doesn't matter a wit that you set that. It isn't gonna happen.
And I think silicon valley is crazy about OKRs and all that it's gonna produce is more of this notion that we have the OK, we had them all set up and people just aren't executing. And so then you fire those people who aren't meeting their key indicators or key results and you say we gotta get some other people who can get the key results that our objectives suggest. And then you fire them, and you fire them, and you fire them- It's like you know, it's just a broken system and again. I mean this does kind of lead into the book you referred to before. The new way to think which is what I tend to do is ask the question, is a model that is in use now producing the results that you wish. It produces right and if it doesn't I'd rather have you rather than saying well we'll fire those people and give some other people who will be able to achieve the key results that we need hopes and they couldn't either until we'll fire them and get to more and then we'll fire them and get some more and we'll fire them and get some more. No, the model in your head was OKR make it more likely that we will achieve the key results and if you don't achieve the key results consistently I'd like you to ask the question that maybe there's something wrong with your model. Your OKR kind of model and maybe there's a better model for me. The better model is key results can only be defined after you have a strategy that would make it actually possible for you to achieve those key results. So that's what I think of as a new way to think. Actually OKRs as a system is not incompatible with strategy. It's just necessary but not sufficient. It's not unlike you saying if you take agile kind of over the edge and say we can't even think about the logic of what we're doing let's just do some stuff and hope it works out.
The sort of bastardization of agile. It's not going to work. It's not going to get the results that you want. What I find is that people fall in love with these models and then they blame everything else other than the model. So in the OKR case it’s blaming the people for not achieving the key results. We laid out the objectives, they were clear as a bell. We laid out the key results. Were clear about we deployed the key results and objectives to all the people and then those lazy or dumb or kind of not focused or whatever people went and screwed up. Incompetent people kind of screwed up our brilliant plans you know. And after that happens for long enough I say you know who's the idiot in this story.
Keeping this model sort of in this position of it. This it's the truth. No it ain't. Ain't even close to the truth.
01:00:09,538
Sohrab: It ain't even close. So I mean ultimately the OKR model, if you follow it like a dogma it becomes again the technocratic model.
Roger: Yes
Sohrab: Because you basically say everything we've determined. Everything now just do it. If you don't include those feedback loops and I think in the article you also refer to the objectives being very connected to the big goal that you want to achieve and you're not very like you're not dogmatic when it comes to terminology but to play and the how to win are the two missing pieces.
Roger: Yes.
Sohrab: If you think about OKRs and that's how you could integrate those two things. Now anyone who is interested in that. Obviously. it was the most read article that you wrote. So people can read up on this as well. Now being conscious of the final topic and probably we're not gonna to cover it but I still want to get your thoughts on this because a lot of my clients are struggling with the topic of transformation.
Big agile transformation, digital transformation, whatever xyz transformation. They all struggle with this and when I read your article on this you mentioned four points that are incredibly important to consider when you think about the transformation. One is the clear strategy and I'll give you time and space to elaborate on all of them. Two is to be patient. Three is to be resilient and number four is to have growth.
Now, can you briefly walk us through these four and for me the most important one probably would be the patient's piece because every client that I meet wants to finish their agile transformation in twelve months. Maybe you have a few words on this a few thoughts on this to share. Thank you.
01:01:57,760
Roger: Yeah. I'd love it if they could get it done faster than that. But it's just not been my experience I mean I guess I'm sort of in very much into logical coherence which is if you're going to use the t-word right, the transformation word, it means something transformational. Otherwise you wouldn't use that word. If you could say, if you can say we're engaging in an improvement effort right then I'd say okay we're trying to make ourselves a little bit better. I wouldn't suggest resilience and patience kind of for that. I'd just say get on with it and if you can get it done in six weeks good for you. But if you're going to have the topic as transformation i.e. the thing at the end of the process. It is going to be hardly recognizable as having any similarities with the thing at the start. So let's call that transformation. The idea that that's going to happen really fast is nonsensical. I mean in fact I don't know or I can't point to a situation where I saw that happen kind of very quickly. I’ve been part of the transformation, I talked about it in the article on the board of Thompson corporation which became Thomson Reuters and it was utterly transformational and it only took about 20 years. Maybe 22 years. So why the patience? The patience is if the change is extremely dramatic then. I just don't, I've just never seen it happen so quickly that having been said I would just say it's not a long dark tunnel right and and I think again this is something where my strategy and agile are friendly like agile doesn't treat everything like a long dark tunnel. Where we're in this long dark tunnel until we deliver the final software to the customer and they're seeing it for the first time. That if you make something into a long dark tunnel then people will obsess about how long this is taking. And how long is that tunnel? Can I see the light at the end of it? If instead you say you know we're trying to go from here to this completely different place we can see progress on a consistent basis.
Sohrab: Yes.
Roger: And again as I said in the article that was my approach at the Rodman School. It was unfortunately not a very good business school when I took it over in 1998. And that's not my assessment that was the assessment of the president and they said we need you to come in and make this. You know, a great business goal it isn't one now it's in fact in some sense it's a bit of an embarrassment at the University of Toronto compared to the Law School and the Medical School and the Engineering School all of which are fabulous. And this very important thing called a business school isn't. So I just set out. Just to say I want to absolutely transform this business school but my goal is to have the change in any one year be so small that nobody's gonna freak out and leave or freak out or push back or anything.
But then when they look back after fifteen years they can't even remember what we were like fifteen years ago. It will be so far in the past and that's what we did, we transformed it utterly. And that I'm not just saying that I was recognized as I mean you know I was at the last of my fifteen years as dean. I was named the business school dean of the year and the citation about why it's the most transformational dean of the era.
Other people said it was transformational but along the way it was just little steps, little steps and I didn't have to feel kind of badly about how long it took.
Because we were getting better every year.
Sohrab: Yeah.
01:06:26,080
Roger: And as I also said I mean our revenues in the last year were x times of that in the first year. So we grew and that's why growth is really important. It greases the wheels of transformation. I would never ever recommend a transformation for a non-growing entity. You're not gonna do it. You'll fail because everybody that you have to take stuff from. Resources to do the transformation. They will fight you every step of the way I didn't take anything from anybody. Now there were people who weren't the happiest when they saw something else getting a gigantic infusion of resources. But since I wasn't taking theirs, they were not going to go to the barricades to fight it. They would mope a little bit you know and I would have probably in that situation, too. But that doesn't inspire you to throw wrenches into the gears, to stop it from happening. I don't know there are people who will sort of say, well Roger, you're not being aggressive enough when you say be patient and and I say gee I'm an aggressive guy. I try to transform things. I think I did it. Anything that I'm involved in, like, I eat my own dog food. I do what I say on my own. On my own staff. I just think they're technocrat like I don't speak kind of well in some sense the technocrats. The technocrat wants to say no i want to be at the end as soon as possible and i think the playing to win / agile type of person.
I think much more obsessed about are we heading in a good direction, in a sound direction and are we making progress towards that.
Those are the two questions I ask. I get the sense that those would be the two questions that an agile person would ask is this a sound direction as if it's a stupid direction and getting there faster is not, it's no great thing is it sound direction. And are we making progress, are we acting in a fashion that makes progress? And I think the pin to win and the agile person both say i'm gonna go to bed at night saying if i do those two things every day then: this is all gonna work out fine. And so they sleep soundly uh they don't go to bed saying we're not there yet. We're not there, yet. We're not there, yet. And that's what the technocrat wants to be done with.
But you know what happens when you're done right? Do you think you're gonna be fine for all time because we're done? No, things are gonna change and you're gonna have to kind of modify it and adjust. Darwin didn't say it's survival of the fittest at all. He said its survival of the post adaptable and so I think having the mindset that says we have to have a logic behind the direction we've set out on.
We have to have a way of working that makes progress in that direction. That said, it's hard and that's again why I think I just said it's hard and that's again why I think I just have not had odds with agile even though lots of people think that strategy is at odds with it. And the reason they do is because strategy is dominated by technocrats and technocracy and technocratic procedures. I would say agile people have to say we don't like much of strategy but it absolutely does not have to be and I think I'm trying to fight a battle to make sure that it isn't completely the lost art.
01:11:01,538
Sohrab: No, we'll fight that battle together, Roger.
With that said two things. One is, it’s interesting calling someone who wrote the book called Playing to Win not aggressive.
Roger: Yes that is true. That is true
Sohrab: That is really interesting and I agree. I mean the technocracy it's there and whomever we can get with us along that fight and it's going to be a long term thing, we're not going to change the world from one day to the other. But as long as we have this momentum that's what I always refer to it. That momentum is on our side because we see the progress. I think we'll be fine.
Roger: I think that is right.
Sohrab: Roger, thank you so much again for giving us your time, your wisdom, your experience. It's an honor to host you and I hope I'll get the chance to interview again about the new book. I'll get that book very soon and then we'll talk.
Roger: That would be great. I would be delighted to do it again. These interviews are always incredibly, enjoyable conversations and I like the notion of an alliance between sort of the agile mentality in the plane to win mentality like that we should make sure we utilize that. Rather than seeing them as somehow very different.
Sohrab: Absolutely.