Fast flow with Flow Engineering | Steve Pereira

Thank you very much for having me. It's great to see the full room. I'm really excited about this. For me, you know, it's partially developer experience, - it's partially performance improvement, - it's partially strategic, it's partially tactical. Flow engineering is the culmination of everything that I learned - all the way from starting in tech support to CTO - and advising large companies, small companies of all different types. So, it really is a collection of different techniques. There's a bunch of different mapping in flow engineering. And really, it's the application of the mapping, - what you learn by doing the mapping, - that's more important than the maps themselves. But I will take you through all of that. And we're going to start with the idea of individual flow. So, I think about flow in a couple of different directions or dimensions. The first one that I think everyone is familiar with is- kind of the work in the flow state, right? When we get into this groove of working productively, - and time melts away, and then all of a sudden, - you know, it's lunchtime, or we're being called to dinner - or whatever else interrupts that. That's kind of the ideal productivity, the ideal developer experience, - the ideal contributor experience. But as you'll notice, it's kind of made up - of these little bursts of time - that is actually contributing value to what we're trying to do, - and we spend a lot of time struggling through toil and confusion - and battling tools and trying to get things working. And what we have to do is really maximize this value at a time. That's part of the exercise of flow engineering. And surprisingly, there's very low, single-digit calculations - attributed to how much time we actually get to spend - in stuff that matters to customers, stuff that matters to our business. We spend so much time really fumbling through processes and friction - and approvals and all the other things that get in the way of that flow state. And I've worked with organisations where that number is as low as 2%. And they're more common than you think. It's very rare to see that number very high. And what we're always trying to do is maximize that number. So, how do we do it? Flow engineering is kind of targeted - at maximizing that number. But even more importantly, the collective sense of flow is really - where you start to see the return on investment for these activities. So, this idea that everyone working together - to deliver value in the organisation is part of a chain of activities - and this coordination effort that it's necessary - to deliver to customers real valuable outcomes. But we see this mix of waiting and producing - and battling toil and friction and navigating the organisation - that slows us down or impacts that time that is value added time - that really unlocks the ROI, - the financial outcomes that we're all working towards as part of teams. So, how does this manifest itself? We can see some of these effects. So, if we start tracing the path of work - through somebody's day or even a release workflow, - you'll see that there's a number of different hurdles - that pop up in the way, that really get - between us and productivity or us and effective flow. And so, that's just the parts that we can see. But there's many factors that are invisible that even have greater impact. And this is really challenging and really where you get the benefits of mapping. So, first of all, we have multiple different perspectives - across any value stream, any flow of work in an organisation. And that really is a challenge because we are talking - different languages in some cases, we're talking past each other, - over each other's heads, we're not connecting. We're not understanding each other. And that means that we can be working against each other. Another way that we're directly working - against each other is having different goals. Some people are thinking about financial outcomes. Some are thinking about career outcomes. Some are thinking about customer outcomes. And it's very difficult and seldom that we reconcile - these things and align them together. And then, the last one that I think is also really problematic - and challenging is different scopes. So, we can be talking short term versus long term, - big picture versus detail, any number of differences - between what scope are we actually talking about. And that's another area where mapping is really powerful and effective. So, I've talked about this collective flow of work, the value stream. What does that look like? So, if you look at your typical organisation, - we're delivering value in various places, - but the path to delivering that value - often traverses many different areas in the organisation. You know, we have marketing as a portion of this, - we have sales activities, we have operational activities, - developmental activities, design, coordination, - and they're all part of this flow from idea to value delivery. And we call this a value stream. It's really just a label for something that is holistic - and contributing value to customers and delivering value to the organisation. But we really only ever have a portion - of this visibility available to us at any time, - which makes it very difficult for us to understand - where is the most impactful improvement opportunity, - where is the proper investment of our resources and our time and our efforts. So, how do we avoid micro-optimizing under these circumstances, right? If we look at this overall flow of work from ideation to delivery of value, - we have our siloed visibility - inside of our department or our sphere of influence, our team, - and that's where we often focus our attention - to the detriment of the overall performance of the flow, right? In many cases, what you'll get from this micro-optimization - is that you'll actually exacerbate the constraint. The conditions that are slowing down the flow of work - or impeding the quality or margin of what you're producing, - can be made worse by not addressing the constraint. So, we could be working against ourselves effectively. But even worse, and I think, you know, any individual contributor in the room - will recognise a bunch of effort dedicated - towards something that they thought was really meaningful, - running up against the wall of business or leadership - that thinks that this is a complete waste of time, - and they just wish that you would focus on this thing over here, right? So, they might be right. You might be right. But we don't know unless we look at the entire flow - and scrutinize it directly. So, how do we do that? So, this is just a very, very simple representation of a value stream map - that collects specific activities that happen in sequence - in order for delivery of value to occur. And some of these things could be happening in parallel, - we can show branching here, - we can show many, many different dimensions. But the basics are these are activities that are happening in sequence. We don't want to get too detailed because we don't want to get lost in the weeds. But what we can notice by looking at this basic data, - this experience data of the participants who are mapping, - is that we have a clear opportunity, right? So, we can take everything that we could possibly focus on - and forget about 99% of it for now and focus on the thing - that is screaming out to us that is obviously a waste, right? Obviously an opportunity. And this depends on multiple factors, obviously, - but we discover these opportunities through mapping, - through conversations with the participants, - through the diversity of perspectives represented across the value stream. So, what does this look like? We have multiple different maps in flow engineering for very specific purposes. We start with the identification of value. What are we trying to achieve? What is the outcome that we wish was true? What can we create in the future - that will bring us to a better state of performance? So that the initial target outcome is outcome mapping. And I'll take you through some of the detail of that. But it's very important to identify value - because it helps us ignore everything else, right? That's incredibly important because we can very easily get lost in the weeds. We can easily forget. What are we doing, again? Why do we have 18 people in 18 different directions? Why can't we get on the same page about what we're trying to do? So, we get it out of the way first. We invest in initial clarity to set off in the right direction - and align everyone to a common goal. That's very important. It's also something that we can bring to leadership - and get their buy-in and support on. Then, we want to look at the value stream as a way - of building clarity on our current state. So, understand the current baseline performance - and what the current process or flow looks like. And we dig a little bit deeper in that, where we find a constraint, - we will uncover any dependencies that are having an impact on that constraint. So, what are the causal factors that are causing the issues - that we're seeing in the value stream? And then, the future state mapping is really all about - understanding how to create a better system. How do we unlock improved flow and performance across the value stream? And that can be faster, better, sooner, safer, happier. And this all culminates in a flow roadmap - that collects all the insights and information - that we gather through mapping into something actionable - that we can go distribute efforts on - and act on what we've learned to make things happen. A lot of people will value stream map and then go back to their jobs - and wonder why things don't get better. You have to have something actionable come out of it. And it should be in the form of, - what are we going to do today? What are we going to do after? And what can we just forget about for now until those things are done? So, we incorporate all this into flow engineering. And the next question that we often get asked is - who's doing this, who is involved? And I think that one of the most important factors - here is executive support. If we want dramatic action, dramatic improvement, - investment in automation and change, - it's incredibly important to have an executive sponsor - who can make space for that, invest in your opportunities, - but also help you coordinate with other groups - that are typically needed in large scale improvement, right? If you want to have a large impact, - you've got to have many players involved in that process. So, an executive sponsor can kind of pave the way. It helps to have a facilitator like me - who can be somewhat impartial to the process, - not steering things in a specific direction, - and also, somewhat of an expert - in navigating the mapping process and guiding the mapping process. You may have coaches in your organisation. You might have facilitators in your organisation who can do this. And so, I would say that you probably have talent - in the organisation who can act as a facilitator. But it's very important to have someone who's outside of the flow - guiding the process of analysing the flow - because you don't want this to seem like a political exercise - or something where someone is putting a thumb on the scale. And then, you have representation from across the value stream. So, the various roles and stakeholders, - you want to have up to a maximum of 10 or 12 people - because you don't want too many chefs in the kitchen. That can help you kind of manage the input and coordination - of getting everyone to contribute and managing all the information. And the second question is, where are all these streams, right? So, you might think that you have an idea of where these streams are. It might be confusing trying to identify them. I think I take a maximalist stance on this. And I will say that you have streams far beyond product development, - far beyond the things that you think are primarily providing - value to the business or value to your customers. I see activity flow everywhere in an organisation. So, from everything from incident response to experimentation, - planning and providing events. You can be working on anything that has a sequential flow of activity - and delivers value either internally or externally considered a value stream. And I think it's very helpful to do so, because the same practices, - the same principles and the same return on investment applies. You can use the same practices and get incredible results - from all of these flows of activity using the exact same process. So, to me, that's very helpful. And what does this look like is usually the next question, - so that's an example canvas that I will commonly build upon - in Mural, this is Mural. There's a bunch of publicly available templates for Miro and Mural, - some of the most common platforms for building out collaborative maps. And what we'll do is, - through building these iterative segments of stream maps - or the various maps, we'll assemble them - into this basic flow of outcome, value stream, dependency, future state, and then the flow roadmap. And so, you can use any tools that you already have for this. You don't need to go buy something. You can use post-it notes on a wall. You can use a pad of paper and a pencil. It really doesn't matter. But something that is real time and collaborative - and remote friendly is an advantage. It really lowers the bar to scheduling these things and getting them done. Another thing about scheduling and getting them done - is we target 90 minutes for each of these mapping sessions. We highly constrain them, which forces them to stay out of the weeds, - but also makes it easy for you to fit them into a schedule - where you have to bring 10 to 12 people together. So, we want to lower the bar for mapping so that people do it more often, - and there's not such a stigma around, - well, this is going to be really expensive and time consuming, - so we'll do it once a year, maybe never. Lowering the bar makes it easier to fit into your existing schedule. And it more than pays for itself because you can easily get 20% savings - on a current state just by mapping and seeing what it tells you. So, overall, we've got this representation of outcome, - value stream, dependency, future state, and flow roadmap. And what happens? So, we'll see increases in throughput, dramatic increases in throughput. We're cutting lead time in half. We're producing more and less time with less friction. It's more enjoyable as a result - of minimizing waste and eliminating friction in the value stream. We lower things like mean time to repair. Deployment frequency, people will see DORA metrics reflected here, - but there's a lot of leading indicators as well - that are impacted by this, as well as lagging indicators. And I think that as a capability, - mapping is something that we really should just do more of. So, I'm a very big fan of just lowering the bar, - lowering the barriers for producing maps, - so that we can all do them more often. I think that we overcomplicate these things, - and they become intimidating, - or we think that they have to be perfect. All my maps are very, very simple for the purpose - of just making them easier to do and less scary [chuckles] to undertake. So, hopefully, by the end of the presentation, you feel - a little bit more comfortable going and mapping in your organisation. The practice of mapping, I think, is extremely underrated, - because it's a really excellent way of reengaging with your team - or people who you've never really connected to before. There's a lot of mapping sessions that I'll run, - and I get the comment afterwards that this is the first time - we've ever actually been in the same room together - or even on the same Zoom call together. And I think that's pretty remarkable because you work in the same flow, - you depend on each other, so having that first exposure be in a mapping session - is pretty remarkable and says something on its own. But it's a way of kind of recommitting to developer experience - or contributor experience and saying, we take this seriously, - we're looking at it, we want to improve. I think that's very helpful for engagement. It really develops a sense of shared understanding, - which is extremely powerful, and the sense of alignment - that you get based on that separate perspective, - separate time horizon, separate incentives, - bringing all that together is much easier in a mapping exercise - where you can see, here's what this person is talking about, - here's what this person is talking about, - how can we connect those things together? Then, this is not prescriptive - in the sense that you will have the same map as anyone else. You won't even have the same map as yourself six months in the future. These things change as the landscape changes. And that's kind of the value of mapping, but you can see a change over time, - and it's going to be tailored to your unique situation. Even in the same team, you could run - two mapping sessions back to back with two separate goals, - and have two different maps, because the goal defines - what you look for and what the map ends up looking like. And then, you remember all of these things. I mean, it's very difficult to learn all the principles of flow. It's very difficult to learn cycle time, lead time, all these terms. By the time you finish mapping, it anchors itself in your mind, - and you really start to see work in terms of flow - in a way that you never have before. So, I think it's very helpful from that regard. And then, as I mentioned, it's extremely remote friendly. There's no tooling restrictions or access restrictions. I wouldn't recommend trying to do it from a cruise ship, - which I did [chuckles] a couple of months ago. You do need some internet bandwidth to do this, - but otherwise, it's extremely accessible. So, let's get to the maps because I'm going to run out of time. So, first of all, outcome mapping is really about setting our target - and getting really clear as a group on what we're trying to achieve. So, the way that we do that is collect a whole bunch of raw data - about what people's experience in the work today. So, it's essentially a big brain dump - where everyone just puts all their thoughts and ideas - and questions on the board, and we affinity map them, - we vote on where we want to focus. And this is a huge alignment exercise. It's a huge clarity building exercise. It really gets us all on the same page, - and we can zero in on where we want to go. And this is not where we want to go forever. We're talking about the next six months. We're going to implement changes in three months, - monitor them for three months, and then do it again. So, even if your favourite target outcome isn't the one - that everyone wants to pursue, chances are - you're going to get to it, because we're going to start moving faster. We're going to start getting things done as a result of this process. So, coming out of this raw material gathering, - we coalesce all this into a map - that represents not only where we want to go, - but why we want to go there, what's in the way, - and then, what are we also going to do? In addition to the rest of the maps and flow engineering, - we can go survey different people, we can go have interviews, - we can go do assessments of different kinds. Again, not prescriptive. This is just a framework for building out - a process improvement, a performance improvement workflow. So, then we get into value stream mapping. Again, very, very simple. Basic data really gets you super far with this. You don't need a lot of detail, - but we can go into further detail if there's value. We can talk about roles, we can talk about quality, - we can talk about different tools - and artifacts in the process, inputs and outputs. You can layer on a lot of data