Steve Tendon's Agility Narrative on Theory of Constraints for Knowledge Work

We started this conversation with Steve Tendon talking about the theory of constraints and doing a deep dive into how Steve has taken the theory of constraints and applied it to knowledge work. It's fascinating. It's interesting. We didn't end up getting to the agility narrative until 25 minutes in. Feel free to skip around this interview. 

Steve has applied the theory of constraints to a service based business. The big challenge when searching for the core constraint in the system is the high variance in knowledge work. i.e. The wandering constraints symptom

He explains that he instruments for work execution signals studying how work moves across the organization. Logging when work starts and finishes provides the core data to manage operational flow.

Steve introduces Herbie as the constraint that limits the performance of a system. How can we find Herbie? Finding the constraint in each team i.e. the constraints in the work process. And the constraint that limits performance of the entire system - The constraint in the workflow. And finally the constraint in the work execution

And much more…. 

Steve talks about the value of agility and who is his protagonist in his agility narrative. Steve's theme for the Agility Narrative? 3 Foundational Patterns - Inspired leadership, Unity of Purpose and Community of Trust. And he identifies the more important question is to connect intent with "what I should not do today?"

More about Steve and his work can be found on https://tameflow.com/

Transcript

Martin: [00:00:00] 

We started this conversation with Steve Tendon talking about the theory of constraints and doing a deep dive into how Steve has taken the theory of constraints and applied it to knowledge work. It's fascinating. It's interesting. We didn't end up getting to the agility narrative until 25 minutes in. Feel free to skip around this interview. Go to the 25 minutes, listen to that component, and then skip back and hear the the theory of constraints and how he applied that to knowledge work.

Martin: [00:00:39]

 Welcome to the Agility Narratives podcast series Citation I co-host, unfortunately, is not able to make it today, we hold this space. So as a community, we can listen to leading change makers and enterprise agile leaders talk about their agility narratives. Hi, Steve, very pleased to welcome you to the agility narratives to talk about your personal journey as an advocate of change and the challenges you see and how you help organizations address them. Each narrative, we hypothesize that this gives us insight to part of the whole. Hi, Steve.

Steve: [00:01:17]

 Hi, Martin, and as I usually say on the podcasts and video appearances. Hello, friends of Herbie, and we might explain what that means as we go along. I'm really happy to be here with you and your audience.

Martin: [00:01:31] 

I'm glad to be a friend of Herbie as well and a fan, so look forward to hearing more about that. Steve, you helped organizations learn about themselves and how they perform together better, how teams can be more effective together, how to optimize team flow. And you have an optimization process that covers at least four flows. I've heard of. Help us understand what's in your toolbox and how agility relates to those practices. 

Steve: [00:02:00] 

Wow, that's a great question to get started. You mentioned correctly that I work with four fellows. They are in order, I would say, the flow of money. So the financial flow, the flow of stuff, the things that need to be done. So that's the operational flow. And then the flow of information and finally the psychological flow. So these are like the four facets or dimensions or perspectives that I have chosen to use when I work with our organizations. The background story is much deeper. The time flow approach is constructed and based on pattern theory with pattern languages. And for the last, I would say, almost 12 15 years, it has subsumed a lot of theory of constraints, which I would say my claim to fame is that I have figured out how to use the ideas of the theory of constraints to see for short in this context of knowledge work where everything is a moving target and we are blindfolds standing on on quicksand.

Martin: [00:03:24] 

So when I was reading the goal and I was actively trying to do that conversion from the manufacturing environment into the services business that I know. So I'm really interested in your take on on theory of constraints and how you translate those dynamics that you see in the manufacturing environment into a service based business.

Steve: [00:03:48] 

Of course, the goal was written some 30, 40 years ago. I don't even know how long it was. And it was set in a manufacturing floor. [00:04:00] So that is the origin of TLC, but TLC has evolved a lot. It then came to handle even the retail distribution supply chain aspects of businesses. It came to handle sales and marketing, and it came to handle project management with the critical chain project management approach. But none of these are really addressed. What nowadays we would say is the realm of agile, of agility, of quick moving, flexible, adaptable organizations that thrive in a situation of uncertainty. And it is quite surprising that TLC has not been applied to knowledge work. Why? Well, because the premise of TLC is is really that of getting away to be able to manage a set, let's say, a direction for the organization, despite the presence of uncertainty despite the ever recurring visits of Mr. Murphy and all his friends. The fundamental idea of TLC can basically be summarized is that we try to aggregate all the actions we take locally to safeguard ourselves from the impact of variation of surprises and aggregates them into one single pot that is shared across the organization. Now this idea is really powerful, and my quest has been how can it really apply it in a situation of knowledge work? The problem with knowledge work is that the sources, the frequency and [00:06:00] the variety of variation are orders of magnitude higher than the other domains.

Steve: [00:06:09] 

And that's why people say no, it's impossible. You will just fail and bump into the wandering constraints syndrome, which means that you find the constraint, the bottleneck, if you wish. At one place and the next moment before you can even think about how to handle it, it does jump to another place. So it is a futile goose chase you never get to to the point where you can put your finger on the constraint. So I set out to figure out how to do this, where we have a multiplicity of projects or products, and where you have a multiplicity of events. That is to say deadlines. And you have a multiplicity of stakeholders. And finally, of course, you have a multiplicity of teams the way the team organizes work handles exactly that situation. I instrument the organization with work execution signals. So you get the right signal at the right time, even if there is a lot of variation. I set up a system that can filter out the real signal from the noise, and I think that is one of the most powerful things that Tamiflu has to contribute and vastly supersedes. The ways that scrum can ban safe is, and, in my opinion, overreaction that produces an excess of activity.

Martin: [00:07:50]

 What is the inventory in this model?

Steve: [00:07:54] 

You can take the the user story of Scrum Extreme Programming [00:08:00] and Kanban and consider that as the unit of work, the work items are the basic units of accounting. In order to construct the buffers, we must study how these work items. User stories move across the organization flow through our value networks. So many nodes, many teams where different pieces of works diverge converge are being elaborated. So it's this idea that work flows through a network that at the end comes down to these units of inventory. A value network is composed of a multitude of teams which are connected with what I call the workflow the network of work moving through the system. And when you are at the team level, we do something which actually the Kanban method already does. We simply log the time from start to finish. We are interested in knowing when a piece of work starts and when it finishes. We are very disciplined about this, that information, the duration of any piece of work becomes the foundation of everything else we are doing to manage the operational flow. So the only thing we are measuring, we are not chasing key performance indicators and many other numbers. Why is this so powerful? Because then you can start looking at this system from a more scientific point of view, looking at these collections of durations as a data set so you can do time series analysis and that will uncover many things for a single team. We simply plot out a histogram which is the flow time distribution, [00:10:00] meaning that historically we have a certain amount of user stories that were completed after a certain duration.

Martin: [00:10:11] 

Herbie hasn't come into the picture yet, and I'm sort of curious about how Herbie fits into this.

Steve: [00:10:18]

 So let's clarify for the audience Who is Herbie? Herbie is the nickname that in the world of the theory of constraints, we used to indicate the constraint that element that mostly limits the performance of our system. And it comes from the story of a Boy Scout and his team that have to go through a hike across the woods and get to base camp before sunset. So there is a deadline and there is a mission, a goal to be accomplished. How that story unfolds is the foundation of the theory of constraints. It is a way to find the constraint. Step one is always to identify the constraint unless you know where the constraint is, you cannot apply all the wonderful ideas of T or C, and that was really the greatest stumbling block in this knowledge work. Also become method started off from a tech perspective, but then abandoned the quest for finding Herbie because it their way of doing it was not successful. They were taking the manufacturing metaphor of looking at the pile ups of work in front of some machine. That's how it works in the manufacturing world. But it isn't logical at all because there is a huge difference between a machine who's working on a very well defined cadence with very small tolerances and the performance of a team [00:12:00] or an individual. Now some days are good, some days are bad, and our performance varies incredibly. So the ideas of managing the queues in terms of work items just doesn't hold. That's where the idea is generating noise rather than giving us the signal. So how do we find Herbie? We are able to measure the end to end time of any work item. So as we have a flow time distribution for the entire team, we can have flow time distributions for every single column.

Steve: [00:12:34] 

And the way to find the column that is mostly limiting the performance of the team is to find the column that has the greatest, average flow time. It's not looking at the cues why it is because these flow times distributions typically have long term stability, and if we want to act on the constraint, we have to find the constraint that is limiting us in the long term in the big picture. And by finding the column that has the longest average flow time, well, that is the constraint in the team. More precisely, it is the constraint in the work process. Every team represents our work process, so if we have 15 teams, each one will have their own constraint. But then of the next question is, well, if we have 15 teams, which one of those teams is really the constraining team. So that brings us to the next level and that is in the systems view picture of all teams, which one is the one that will mostly limit the performance of the entire system. And that's where we then focus on the constraint in the workflow. Remember, the system is the work that flows [00:14:00] through a network of teams. We want to find the constraint in this network and that is the constraint in the workflow. I must also add there is a third categorization, which I call the constraint in the work execution. And in order to be able to manage this entire system, we must concurrently have our eyes on all three. The work process constraint, the workflow constraint and the work execution constraint

Martin: [00:14:29]

 In reflecting on what I'm hearing is a scientific approach to work that you're taking everything. It comes down to how that actually reflects the work that's been done, the constraints in the system and how people interact together. Is that correct? Why that unit and why is that more effective than other units that you've considered?

Steve: [00:14:57] 

My approach is very scientific, and it builds on cubing theory for the management of these networks. And then when it comes to the financial parts, we use the ideas of throughput economics or throughput accounting of the theory of constraints when it comes to the information flow. I look at the the span, the frequency and the latency of the feedback loops, especially the feedback loops that connect the bottom of the organization. So the people performing the work to the top, to the very top and any layer in between, because that's where there is a lot of dysfunction in most companies, they do not know how to communicate, what to communicate about, when to communicate. And we need to be able to approach that in a scientific way as well. And finally, [00:16:00] even when it comes to the psychological flow, the human dimension is really put at the center stage and where people are more happy than before. Well, there I use a lot of the ideas of myalgic sent me Ali, who, by the way, recently passed away at the American Hungarian psychologist who first defined what psychological flow states are those states of optimal experience where where we are in the zone, basically. So in all these four flows, I try to have a very scientific approach and try to make them all work together. We create a unity of purpose where all people in the organization, from the CEO to the latest junior higher know what they have to do. They all go in the same direction because they have that shared common interest. They will not waste time and energy in friction and conflicts, which is something that most other approaches just do not address at all. To the contrary, many of the agile approaches are actually conducive to, well, it's a divisive mechanism whereby there is a detachment between management structures and operational structures.

Martin: [00:17:29]

 I heard you mentioned your approach doesn't create this conflict. And what is the what is the fundamental element that is aligning people in your method that you're not seeing in the agile and Kanban methods?

Steve: [00:17:44]

 It is defined by goal drought, even though it's crude in its expression. He stated that the goal is to make more money today and in the future. Now that might seem very antiquated and not very [00:18:00] much aligned with the human values of a fragile. So I say it doesn't matter what the company is doing, the company is a collection of individuals, individuals in their life, they grow. So your personal need for money is on a growing curve. I think that is the rule, not the exception for people. Now put all those individuals together in the company. They expect pay raises. It's a clear need. At the same time, we have another dimension, which is time that creates the mechanism that we want more money for less time. From a company perspective, they want the same from the market. And that money velocity money over time is really the unifying factor. Why is it so powerful? Because all activities can be translated into the elements of throughputs accounting, and therefore it becomes easy to decide which brings most bang for the buck. It's extremely simple to to explain, but it's more difficult to do because the companies are using cost based accounting, even for management decisions. Every department has their own KPI and these KPIs are in conflict. So that's where we are seeing the conflicts between departments coming about because of these KPIs, which are not coherent, not consistent, not based on one unifying model.

Martin: [00:19:38]

 I think the core message about the unifying effect of money and people's need to be more effective is a strong message. A couple of things I'd love you to cover just briefly how do the patterns fit into this? And then the second question is how does mental models fit in?

Steve: [00:19:58]

 I develop team flow with [00:20:00] patterns, Alexandrian patterns. Christopher Alexander was an architect that invented this way of describing reality, and patterns have two big use cases. One is as a descriptive element where you can describe what you are observing. So just like a doctor looking at a patient will make a diagnosis. If you have a collection of organizational patterns or an organizational pattern language, you can observe an organization and quite precisely describe what is going on there, so it becomes a diagnosis.

Martin: [00:20:42] 

Give an example of one of those.

Steve: [00:20:45] 

Let's take the most common one. We know that multitasking is bad. Someone decides that something else than what you're doing now is more important. So you interrupt that work and create this context switching. What results there is that the working process increases, that is a pattern you define a context is multi-tasking creates a problem. It is too much work in process and then you define a solution. A solution to a problem in a context is one of the canonical definitions of a pattern. And what solutions can we have? Scrum invented the time box time boxing limits the amount of work in process. So it's one solution to that problem. The Kanban method invented the column without limits. It has the same purpose to limit the work in process, but it's a totally different solution than a time box. And then Team Flow has a third solution, which is called drum buffer rope. I know that's opening another chapter in the conversation, but let's just say it's a third way to resolve the same problem in the same context. And why is this powerful? Because depending on how you choose to solve that problem one, two or three, when you apply that solution, you're changing the context. So once you have chosen to do time boxes, sprints like scrum does, there are certain things that you will no longer be able to choose. And conversely, there are other things which you can do just in virtue of the fact that you're having time boxes.

Steve: [00:22:35] 

And this gives this idea that patents are composable. It's like functional transformations. You apply a pattern, you have a new context and so on and so forth. Once you understand that limiting working process is important, that is a mental model. It is an interpretation of reality. I see that my inbox is overflowing, the calendar is full. I don't have time for anything. I must limit the working process. So you start to have this idea that in reality, there is something that is too much, too much work. And I must reduce that that amount when you apply the pattern and say that, for instance, let's do it the Kanban way with column limits. That decision is transmitted to team members that are involved in implementing that solution, so they must understand what they are doing. You must transfer that notion that, hey, we apply column with limits. And once people understand the reason why certain things are being proposed, well, most likely they will have a self-interest in applying those. And when you offer them a solution that addresses that pain point, they will own it. They will adopt it because it results one of their problems. [00:24:00] But notice very well. It does so because they got the the mental model, the interpretation of reality, which then turns into operational, actionable decisions. Let's use the word limit.

Martin: [00:24:14] 

I love the concept of mental models because it gives you that picture in your head of a problem to be solved or solution that that can be delivered, and it provides that mechanism to navigate quickly around around topics. So let's start on your agility narrative. So as a change maker, how does innovation and agility fit together?

Steve: [00:24:38]

 They are tied by the hip because when you are acting in the business environments we have today where change is constant and very, very frequent, your competitors will come up with new ideas every day and it's not sufficient to keep up. You must be able to out innovate them. My definition of agility is is very dictionary. For me, it is the ability to change direction at speed and I would add even at scale. So agility as the ability to change direction at speed and at scale is essential for innovation. Why? By using the idea of throughput, if I am able to perform ten times faster than you in terms of organizational performance? Well, maybe in the next quarter, I can run 10 experiments and test 10 ideas on the market while you are busy implementing only one. So agility gives you the capacity to explore new ideas faster. And innovation happens mostly by discovering what does not work. And then you try the next thing and the next thing.

Martin: [00:26:04]

 Thank you. In your change narrative? Who is the protagonist? Who's that central character that you see at the center of this agility narrative?

Steve: [00:26:14] 

Well, from my perspective, no doubt it is Herbie, of course. Who else? The constraint,

Martin: [00:26:21]

 The constraint working out, the constraint.

Steve: [00:26:24] 

Let me say something important. You really care about this, that in in traditional organizations, there is very much a blame culture trying to find faults, and if it didn't work, someone has has to be the scapegoat. Now the idea of being the constraint is a very, very heavy one because you become the center of attention. And I work a lot with companies saying there is absolutely no dishonor in being the constraint. We don't want to find the constraint in order to blame with the constraint. We want to find the constraint in order to help the constraint. So it's this like a reversal of weight being the constraint is not a burden. It's a great focal point where all energies can come together.

Martin: [00:27:16] 

My story around that is that I find that good people do really great work and therefore they get more work because everybody finds that they can deliver, so they end up taking on too much work and becoming the constraint. People then look at them and go. This guy is not performing, the constraint doesn't necessarily mean you're not performing, and it should be almost an honor badge to be the constraint in a system and get needing the support of others to make sure that you are supported out of that situation.

Steve: [00:27:50] 

Absolutely, it's it's like being the most important person of all in the company.

Martin: [00:27:55]

 Yeah. So what would you say your theme for this narrative would be [00:28:00] and why? Why would you pick that theme?

Steve: [00:28:03] 

If I go back to the origins of time flow, I define three foundational patterns the inspired leadership, the unity of purpose and the community of trust. Those three things are what is at the center of all decisions in particular. Unity of purpose and the community of trust are very related to the idea of the constraint. Why? Because when you find the constraint, you have one one focal point where, as we just said the moments ago, the whole company can focus its energies. It is like a catalyst, an enzyme. It brings together all forces and therefore helps creating this unity of purpose. And because you're all working together with one thing in mind, it facilitates the development of a community of trust. So inspired leadership, unity of purpose and community of trust are at the center stage.

Martin: [00:29:03]

 Those are three very powerful elements. One of the conversations I've had recently is about intent. It's very similar to your community of purpose. The question is intent is really good, but how do you connect intent with?

Martin: [00:29:23]

What I need to do today,

Steve: [00:29:24]

 The most important question you should answer is not what I should do today, what, but what should I not do today? Because if you have this idea of a constraint which is one point one element, all others, which is the majority by definition, have more capacity than the constraints. And if you want the flow of work to move unhindered through the organization. And the intent to use the limiting work in process, you must make sure that no one [00:30:00] is working more than the constraint that everyone works at the capacity of the constraint. So the most important operational decision that you can take on is that when you have the urge to start a new piece of work, ask yourself, Is this at the capacity of the constraint? Or is it beyond? Because if it is beyond stop, go home, go fishing. As I say, take it easy. Be idle. And this is such a mind blowing statement for companies which are focused on resource efficiency and cost accounting. Being idle is acceptable behavior because it supports flow in nature.

Martin: [00:30:50] 

I'd love to respond to that and dig into that. That's a beautiful set of statements, by the way. Thank you. So in your change narrative? Who are the villains?

Steve: [00:30:59] 

I would say any individual that works against the unity of purpose. There can be many. And the usual suspects, which comes from the the finance side because of the focusing on cost cutting and and KPIs. But I would also say, you know, if you are in typically a scrum shop with agile in general, it's the anti management sentiment that is very common in this industry. And for me, that that is a no no because you are creating a division in a modern corporation. What matters is inclusiveness and broad collaboration, which means everyone from the CEO to the latest junior high.

Martin: [00:31:48] 

I also get frustrated with the agile community blaming leadership for not being agile enough. And my answer to that is generally that agile isn't meeting the needs of the leadership. [00:32:00] It's a two way street. So I'm really glad that you called out that as a villain. What do you see as the impact of that?

Steve: [00:32:08]

 If you are able to get old people to stand behind one idea and break down the reasons, the internal reasons for having conflicts and friction, you will be releasing a lot of energy and time that can be dedicated to improving the company, improving the offering to your markets, serving the customers, being more true to the real purpose of the company. So if you get this and you see it from a perspective of of abundance rather than scarcity, there's no point in fighting one another inside the company. Because if we work together, we can make so much more that there is so much to gain for everyone. So why not?

Martin: [00:33:02]

 It's a great question what's at stake here for clients today, faced with the changes in the marketplace of not taking those steps to address the conflicts, the constraints in their environment and creating the better flow in their organization?

Steve: [00:33:21] 

Great question. And I would refer back to the later writings of Conrad, where he was addressing a similar theme. Who are the stakeholders? Well, the obvious ones are the owners, the the shareholders. They they must get their part. You have the employees. Employees need to be satisfied and happy or they will go somewhere else. You have all the suppliers, they need to be treated fairly. You have the market, the customers. They need excellence of service and products. And then you have the broader the [00:34:00] world where we live in the planet, Mother Earth, which we are using with how we are consuming all resources. So once you start reasoning in terms of unity of purpose, what is at stake is finding harmony between all these different components. And I go back to what I said moments ago that it is possible to have a unity of purpose and you're opening up a world of abundance. Well, that will resolve many problems that we have today inside the companies. But on a broader scale, the problems that we have on this planet, we have to start from somewhere. Let's start from our companies.

Martin: [00:34:43] 

Sounds like a great plan. So you're meeting with a leadership team and they're a little frustrated that there's a stock with where they're at and they see the opportunity and the threats in front of them. What is your call to action? What would be your call to them for action?

Steve: [00:35:04]

 Well, it's it's really very context specific, and that's where the pattern training would come into being trying to map out the territory and doing a diagnosis. But there is like a meta pattern behind this because if you dig deep enough, you know they will all have their pains and problems and aspirations and desires. So it's this tension between, let's say, feeling safe and feeling satisfied. If you dig deep enough and keep on asking why, why is that a problem, why like root cause analysis if you want? Sooner or later, you will be able to make that connection. I need more funding. I need more people. We don't have enough time. We have debt march deadlines. So as soon [00:36:00] as you can make that connection to the dimension of money and time that if there were more of them available, then some of those pains would would be resolved or at least alleviated. Well, then you have found the commonality and you do that across the board and say, OK, guys, girls, if we all get together to try to resolve that issue of the velocity of generation of financial throughput of money, then we can address the problems we are facing. So you try to bring about a situation where the organization develops the awareness that there is something very elementary that is common to all of us. And once that awareness is in place. Well, I would say you've done the hardest part of the job. All the rest follows by by logic and science.

Martin: [00:36:55] 

And how do you see the future? What is your outlook on the next 10, 20 years?

Steve: [00:37:02]

 I think that, you know, from a tabletop perspective, even if Tableau has been around for quite a while now, it's still unknown approach in the industry. Of course, I do not have a crystal ball, but let's say my wish is that more people discover these ideas and just try them out now. Try to apply them. Quite confident that if a sufficient number of people discover these elements, there will be a snowballing effect. And who knows, maybe tomorrow will become the norm, then that would be great.

Martin: [00:37:35] 

Thank you, Steve. As you mentioned partway through this, I think we opened a lot of chapters and there were a few chapters that weren't summarized at the end. And you've brought us a lot of really key information about theory of constraints, your perspective on patterns and the value of mental models. And I really would hope that we can do this again sometime and dig into [00:38:00] a couple of other areas because I think we could talk for a couple of hours without any issue here. So thank you.

Steve: [00:38:07] 

Indeed, just one story is not enough. We need an epic.

Martin: [00:38:12] 

We need an epic. A reference to agile. Thanks, Steve.