GA 560 | Flow Engineering with Steve Pereira

This is where we should focus and getting everybody on board, which I think is another like, huge realization for me was that individually we can realize all kinds of things, but if we can't convince other people or bring them along or meet them where they're at, then, you know, we're often just wasting our time.

Today I am excited to welcome Steve Pereira back to the show. And no, he's not related as far as we can tell, but I suppose we are sort of connected in this whole world of continuous improvement and making things better. So it's been fun getting to know Steve over the years. And so he was on back in, oh gosh, it's been a couple of years ago. And we'll link to his previous episode below where he came on talking about value stream mapping. So as you'll hear, Steve has got. He's more in a sort of. In a software world. And so he's passionate about many of the same concepts that all of us sort of traditional lean thinkers are passionate about, including things like value stream mapping and then fleet flow. Flow in particular. And so in fact, he's recently written a book that was just released called Flow Engineering. And now this, this book is. Is definitely targeted towards sort of the software community. And so during the episode we get into really just the topic of I, I opened up with like, what do you mean by flow engineering? Because for lean thinkers, we hear flow, we hear engineering. So what. What just sort of defining the term. And I thought he had a really good job talking, got into some of the key maybe tenants of what he's passionate about and the importance of value, clarity and flow. And so he sort of walks through those and how they play along with one another and how sometimes it's a chicken and the egg sort of thing with, with value and clarity in particular. But, but I think he does a nice job sort of wrapping it all up. But yeah, great guy. I really enjoyed this episode and I hope you do too. Show notes can be found over@gembapodcast.com just look for episode 560. Again, that's gembapodcast.com, look for episode 560 and as always, scroll down on whatever app you're listening to. And please go connect with Steve over on LinkedIn and check out his book if it sounds like something that would add value to you. Okay, enough from me. Let's get to the show. Steve, welcome back to the show. How's it going?

It's going great. Thanks for having me me back. I'm really excited about this. I'm, you know, I'm such a fan of this topic of conversation and, you know, the work that you've been doing. So it's. It's a pleasure to be back talking to you again.

Yeah, I was just checking it out before we. We jumped on that. You were last on in episode 382. That was a while ago, man. We were talking about value streaming and all kinds of stuff, so.

Yeah, it's amazing. I remember I was. I was off by. You know, I was. I was. I think I was at a lakeside property trying to get away from the city for a bit. During COVID I think it was during.

COVID It might have been, yeah. Well, good stuff, man. Time flies. So where. Where are you calling in from now, Steve?

I'm at home. I'm in my office, so. Familiar territory. You know, as I mentioned before we started, my. My office setup's been thrown into a bit of a turmoil, but I'm trying to get it all sorted out. I'm in the midst of recording a new course. We're doing a course based on flow engineering, and that's. That's why I'm experiencing with. Or I'm experimenting with lighting and stuff. So my setup looks a little. Maybe a little severe for.

It's all good, man. It's all good. You look great. So. All right, well, hey, Steve, you know the drill. We like to begin all of our episodes with our guest sharing a quote. So what do you got?

Yeah, so I. I dug in the crates and looked, and I pulled one out from Michael Nygaard's book, release it from 2018. And it's all kind of. It's focused on delivery, which is, you know, what I was kind of focused on at the time. When I read this book, it came out, and I read the book, and the quote I really like is, a new architect will focus on the boxes, and an experienced one will focus on the arrows, or an experienced one is more interested in the arrows. And to me, this separation between structure and dynamics or the architecture of things and the behavior of things is really interesting. And I find that the more experienced I am, the more I am focused on that dynamic piece, like, what's happening in between the structure, what does the structure allow for in terms of behavior and what is happening in the system, rather than how the system is kind of structured. So I like that one to remind me that for as much as I like the tech and I like building stuff, it's what that enables that really matters.

Got it. Got it. All Right. Well, hey, in case folks don't remember, from episode 382, which was a few sleeps ago, tell us a little bit about yourself, Steve. Besides having a great last name, tell us about your background, how you got involved in all this continuous improvement and flow work, and then tell us what you're up to these days.

Yeah, so it's been a winding road. Been in tech for over two decades, and starting out in tech support and kind of work my way into software and software delivery, and that was. That's always been a weird space. It's always been kind of a niche area because people love building stuff, but actually getting it out into the world and then keeping it running and performing and improving over time is often an afterthought. But I was always really interested in this idea of, like, what does it take to get this done? How do you enable this kind of collaboration that it takes? How do you make it easy to do the right thing, hard to do the wrong thing? Create systems that help people do work rather than focusing on the work itself or the, you know, the artifact or the software that is ultimately built. How does it get built? How do you make that easy? How do you make that sustainable? How do you allow the talent to sign and contribute without getting bogged down in the toil and friction of everything that it takes to get these things done? So that's been a passion of mine forever. And then I kind of stumbled into value stream mapping as a way of seeing that picture from a broader perspective and then having the measurement and quantifiable and qualifiable data to say, this is where we should focus and getting everybody on board, which I think is another huge realization for me was that individually we can realize all kinds of things, but if we can't convince other people or bring them along or meet them where they're at, then, you know, we're often just wasting our time.

Yeah, got it. Okay, well, and you've also been busy writing some books, is that right?

Yeah. So the first book, Flow Engineering, was kind of the culmination of what we talked about in the last podcast was, you know, these lean practices of value stream mapping and the. The other practices that kind of support value stream mapping packaged up into kind of a cohesive program that you could run yourself. And at the time, I just had kind of the ebook that you could download from my website, and then we made the leap. And the reason that I made the leap into writing a book was my co author, Andrew Davis, got in touch with me after running A lean coffee on value stream mapping at a conference. And he said, have you ever, what books would you recommend on value stream mapping? And of course I said Karen Martin and Michael Osterling's book on value stream mapping. But there wasn't anything at the time focused on the DevOps space or the software native space, the, the modern large scale digital engineering space exclusively. Karen's book is wonderful and it spends a lot of time on knowledge work, but it's still got a significant portion of its focus on manufacturing. And that sort of stuff kind of bounces off your typical engineer, your typical software or product development team. And luckily we got Karen to write the foreword to the book, so it's been wonderful to collaborate with her on that. But Andrew and I wrote that book as kind of like something that we could hand people and say, this is how we do it and it works really well. And you could follow the recipe essentially in the book to do value stream mapping. But all the way from getting initial buy in and figuring out what should we map and why are we mapping and what do we expect to get out of it, all the way to building a roadmap that says this is what we want to experiment on or change and when it's going to happen and who's going to own it and how are we going to measure success. All those things that tend to get forgotten after you map the value stream and it becomes this artifact that people will put, put on the shelf for, you know, put in a knowledge base and forget about.

So. Okay, well, I always, like sometimes when we talk about specific things, like you mentioned flow engineering multiple times and obviously name of the book. We got a bunch of people listening right now who are, you know, sort of like in the lean world, whatever, it's manufacturing, office nurses, doesn't matter. And they're all about flow. They're all about making value flow and all the rest of it. So what is. Define flow engineering for us, for the lean world out there? Who's. We're all about flow. I mean, Womack and Rather wrote about flow many years ago. Right. Kind of the og. So like what do we, what are we talking about here when you say flow engineering?

Yeah, so part of it is appealing to a cohort of professionals who are really heavily in this digital space. And what gets their attention are things like DevOps where they're like, okay, well this is a term that appeals to me because it's inclusive. It's like, I understand the combination of those terms and that's like a new value Creation because it's unlocking this new collaboration between these two siloed departments. And that's really exciting because now all of a sudden we have this, this term that we can collect a bunch of ideas underneath and it makes it easy to talk about. And, you know, we have everything nowadays. We have things like prompt engineering for AI, like, how do you create the best prompt? How do you refine that over time? We had product ops, which is like, how do you operationalize product and create some systems and platforms and capabilities that support product? And so part of the goal for flow engineering is to give folks a focal point for this area of practice that will perk people's ears up and say, oh, that sounds, you know, that sounds kind of interesting. How do you engineer flow? And the first thing I'll say about it is it's not a technical practice by nature. When I think about engineering, I'm thinking about solving problems, but systematically solving problems, empirically solving problems, thinking about them in a way of we have a target and our current state is deficient in some way, and then we have to reconcile the current state with our target state. Very kind of simple engineering practices, but not technical as much as they are human. In the case of flow engineering, right, I mean, we want technical improvements, but the way to get those technical improvements in a cross functional, siloed, bureaucratic ecosystem is bringing people together and solving people problems. So the people problems are, they could be anything from we have separate incentives and we need to align those incentives, or we have a variety of goals and what's our overarching objective that brings all those goals from parochial concerns where, you know, everyone is pulling in a different direction, to now we can all pull in the same direction because we can agree on a common target. So the engineering is kind of a broad term to represent all the problem solving it takes to enable flow. And the flow that we're talking about is quite inclusive as well. It's everything it takes to contribute and deliver value with minimal friction and toil and delay and waste and all those things that we know kind of get in the way. So flow engineering is really just a term for folks to bring together a bunch of separate ideas that tend to be put aside and not collected under something that folks can really focus on. We have continuous improvement. We have Toyota Kata. We have all these things that are floating around and we tend to forget about them over time. Right. So it's also an effort to kind of renew an energy and pull some eyeballs towards something that I think deserves a lot of attention. Like, as I mentioned, this is like a. Tends to be an afterthought. Like how we do the work tends to be an afterthought to the work that's being done or the artifact that's being produced, or the product that's being produced. But, you know, it's. It's super important. So it's trying to enable some. Some attention, but also give folks a. Something to rally around. Yeah, we have a lot of people now talking about, like, I'm. I'm gunning to get my position title changed to flow Engineer, because, you know, to me that's like a very meaningful title. So that's really interesting as well.

Yeah. So I know, I know there's three terms that you're. That you're. You're passionate about. Value, clarity, and flow. What is the. What's the relationship and why are those important to you?

Yeah, I mean, when we started to write the book, wanted to make sure that everything that we were talking about was kind of like, anchored in bedrock and that we could empirically build up proof that this was something that was a sound set of ideas and practices, and that we were thinking about this very thoroughly and rigorously. And so we kind of worked backwards from what does it take to enable flow? What does it take to improve flow in an organization? And we went actually back and forth between value and clarity. So we knew that you have to eliminate waste, you have to eliminate friction, you have to create these conditions for effective information transfer and work to kind of pass from silo to silo.

And.

Initially we were thinking that clarity is the most important starting point for that. You have to have clarity in order to focus your attention on the most valuable thing. And if you can do that, then you can kind of unlock the delivery of that thing by enabling flow. Right. So once we know what we want, we can sort of get everything out of the way of getting to that, and that would be really valuable. But then I was thinking about this and working through this and talking to Andrew, and we're going back and forth, and we sort of stumbled into this realization that unless you have a sense of value, you don't have a sense of direction, and it's very difficult. What are you clear about? Right. What is clear? If you don't have a focus, then you're sort of trying to do that thing where you capture everything and then you have no filtering mechanism. What's important, what's not important? How do we know that we're clear unless we have a sense of value? That's defined. And so these things are somewhat chicken and egg. But I think that this ordering of focusing on value first and getting a sense for what do we feel is valuable. What is our hypothesis, what is our North Star, what are our values sort of orienting us towards. And then clarity around. Okay, what does that mean? That we pay attention to what information is going to help us move in that direction. And then flow is basically how we make it happen. Right? So, and I was trying to, you know, I was trying to find, like, how do I know this is right? And, and that brought me to cybernetics. So the cybernetic connection to flow engineering is, you know, the, the classic cybernetic loop is like the thermostat, right, where it's like you set the temperature and then you measure the. The current temperature versus the target temperature, and then you reconcile the difference. You turn on the, turn on the heat, you turn on the cooling to get you to that target. And I was thinking, okay, well, that's very clear to me. You need a target first. And then you figure out what's it going to take to bridge the gap between your current state and your target state. And then you work to eliminate that gap. And the flow is really the activity of going through that over and over again and making it easy and productive for you to do that. So that got me really interested in the cybernetics space. And there's a lot there that we probably can't get into. But it was very interesting to dive into that and put flow engineering against it and see how does this actually fit. Can I point to some scientific basis that that supports it?

What about. So obviously, just over the last couple years, artificial intelligence has exploded. What role is it playing in your world as it relates to flow and different various aspects of.

Well, I think it's really, it's a fascinating space, and I don't think we, we know a lot about what it looks like in the flow space yet. We're still in very early days of, I would say, kind of data collection and analysis for flow, like understanding how are we doing and what would make a difference for our throughput or quality, for instance, especially in the digital space, I would say especially because a lot of our activities are difficult to kind of attribute to financial outcomes or throughput. And a lot of that hinges on data quality. We're not attributing the right behaviors to delivery or quality. Then it's very difficult to connect the dots back to, okay, when we do more of this, we get this result or this outcome. And so data quality is really a major issue that's holding folks back from monitoring performance and making really good decisions about having hypotheses for improvement. And I think AI can really help.

Us.

If not clean up the data we have, but help us collect higher quality data as we move forward. We're always realizing that if only we collected this data six months ago, a year from now or a year ago, this decision that we have to make today would be so much easier. And so we're always in this battle against poor data quality that's holding us back from making better decisions. Think that AI guiding us to kind of record data at the right time in the right form and then helping us do that attribution between this was done six months ago and this is the outcome. I'm hopeful about that. I don't know how realistic it is. I know that in the value stream management space and in the digital world we have a massive barrier around data quality and data access. And I think hopefully AI can help us assess data and give us information about fitness for that data, overall quality and what information we can gather from it, but then also make sure that that data doesn't drift into poor quality or incoherence and help us manage new data sources and all these things that support our understanding. So that's kind of on the data side, but there's also what happens in flow engineering. We're talking to these teams and we're building out these value stream maps. And it's been really helpful to me to have AI as that facilitation copilot that's recording, transcribing, analyzing the, what people say and how those sessions go and then, you know, essentially inform what I can improve about my facilitation. But then surface things that I might have missed along the way. I can take a bunch of output on a board and synthesize it again through AI and say, you know, I've affinity map these things this way, but how would you do it? You know, and that, that helps, that helps me kind of see things that I haven't seen and also improve my ability to do things like affinity mapping. So I've really enjoyed having that as a support in, in a role, as a, as a facilitator.

What's your favorite application? I mean everybody's chatgpt is, I mean, what are you, what are you playing around with these days?

Well, it, it varies for different, for different things. I've, I've recently enjoyed using Perplexity for anything that I need attribution for. So when I'm doing a lot of research, I want to know the source of all the, you know, everything that comes back. So perplexity is pretty good at pointing me accurately to where it got information from. And that's really helpful. I use a, I use a tool called Aphorai, which folks probably haven't really heard of, but it's like I keep my whole library in it. So I have digital copies of all my books, I keep them all in there and then I can go and query it directly. So anytime I want to know one of the principles from Don Reinertsen's book, the Principle of Product Development flow, there's like 163 principles. So that's like my go to every time I'm like, okay, what's the, what does Don say about this? I can go directly to that and then for, for working on the course. Actually we're using Claude because Claude's really great at generating kind of like human output, you know, and, and so it's a great resource to, to again examine what are we missing? What, how is, how would Claude phrase this? So it's, it's different things for different reasons. But yeah, I'm still learning about new tools all the time. Do you have any favorites?

Oh, I mean, I mean, I mean, honestly I don't do a whole lot. I mean, if I use chat, gbt obviously for, you know, writing or check my grammar or something like that, but I've got one that I really like and I, I guess it's technically AI is Grammarly, you know, like, it's so powerful, man. It's like, it's like you think you can write, but no, you really can't. Or if you do, it's really bad, you know, at least mine is, you know, so, so I'm a big fan of that as well. Just, you know, to communicate better because I have to communicate with a lot of people, you know, so it's great.

To have as a feedback loop.

Right?

It's great to have as this mirror that's like, I think this is okay. And then it's like, not really.

Yeah, yeah, yeah, no, for sure. All right, well, hey, I'm curious to hear. I mean, I've heard good things about the book, but what has it. What's been the reception been like for you and, and who's. Who's reading it?

I had, I mean, I, I may have over indexed in towards like not having any expectations for the book, just to maybe protect myself emotionally from any outcome. Sure. So I. I haven't, you know, benchmarked against my expectations very much, but I've been really, really pleased with the stories that folks are sharing about people saying, I never really had a job title that I resonated with, and I would love to be a flow engineer. And I'm petitioning to have my title become flow engineer because that, to me, encapsulates what I do and what I want to do in my organization. And so that, to me, is music. That's cool. That's wonderful to hear. Yeah. And I think that. That sort of. That sort of empowerment is not an outcome that I was anticipating. There's a lot of folks I hear from that have flow engineering teams, and they now have a group that's dedicated towards helping other teams with their delivery or getting rid of impediments or breaking dependencies. And, you know, they're using the book as a playbook. And so I think one of the magical things about a book is it gives people a little bit of support for things that they already knew and maybe a little confidence that says, like, it's not just me. I've got something I can point to and say, this is not just my idea or, I'm not the only person proposing this or doing this. That's what I've used books for in the past. Don't take my word for it. Right. I mean, you don't have to trust me. You trust this person, or you trust these companies who've shared their stories. So I think that's really powerful. And having that. The label to collect all of that underneath, and folks talking about flow engineering.

And.

In their internal conferences and in public is really, really powerful. I've heard a story from Toyota recently. They had flow engineering came up in one of their internal conferences, and they've been talking about it internally. And to me, that's like. We talk about Toyota in the book, and of course, Toyota is huge in the lean space, and so it's wonderful to see that come into a circle. Mm.

Wow. That's awesome. Well, congratulations again on it, man. It's. It's a huge achievement to. To write a book like that that can, you know, really have a positive impact on. On. On a lot of people. So well done.

Thank you. I appreciate that.

Yeah. Hey, if folks wanted to connect with you, Steve, what is the best way?

Well, I spend a lot of time on LinkedIn so that they can definitely find me there. I. We have a lot of really great conversations right now. There's a raging conversation about value streams, whether you have value streams that are only external to companies or if you can have internal value streams, And I love LinkedIn for that, we have a really great community. So that would be my suggestion and then I can point folks anywhere from there. We have.

Yeah, yeah, yeah. Well, we'll have it linked up here in the show notes. Yes. Everybody just scroll down on whatever app you're listening to this on and, and we'll have, we'll have Steve's handle there, so go connect with him. But hey man, thanks for coming on and we didn't get into your Agile work, but maybe we'll have to have you on down the road. Give us a sneak peek on what your, what kind of trouble you're working up there.

Yeah, I love to, I would love to do this. So this is the, the next book that's just recently come out is the Lean Agile Way. And I collaborated with a bunch of really smart folks on that. And you know, it's really kind of a reference for the convergence of Lean and agile in this kind of digital first world. So I'd love to come back and talk about that.

All right, man, well, we'll have to, we'll have to make it happen, man. So it's always good talking to you.

All right, thank you.

All right. Okay. Well, until next time. Thanks, Steve. Thanks for listening. Whether you've been on the continuous improvement journey for many years, or perhaps you're just getting started, Gemba Academy is here to support you. And while we're best known for our more than 1500 lean and 6 Sigma teaching and virtual tour videos, we also have a team of experienced Lean and Six Sigma practitioners available for one on one coaching, as well as variety of Lean and Six Sigma certification options. To learn more and to schedule a demo, head on over to gembaacademy.com.