 So let's do a bit of housekeeping to start. You're here for webinar. It's how to make every bet count when you can't do it all. That's the name we came up with. I think we're going to dig into something called Driver's Constraints and Floats today. If that's interesting to you. I've written about it a couple of times, and I'll also mention the inspiration for that. Johanna Rothman wrote this great book. It's probably eight years ago now that changed how I saw things. So we'll talk about that too. Okay, whoops. Okay, housekeeping. The session will be recorded. We'll email it to you later this week. It takes a couple of days, but don't worry. We will email it. And next, if you have a question, use the Q&A feature. We'll not be able to track the chat for questions. Hopefully that's not a problem. And we'll try to answer questions throughout the discussion, but I'll leave more time towards the end for that as we're doing that. And I think that's about it to get started. Okay, let's get going. I hope you all are doing well. I'm still waking up. Hold on a second. One more cup of coffee. Okay. So to set the scene for today's talk, I wanted to talk about spider sense and specifically product manager spider sense or sometimes known as Spidey sense. And so Spider-Man will occasionally get this sense that something is wrong. And we call that Spidey sense. In this case, he senses that he's the bread baking experiment is not going to work out well to do these things. But that's a decent example. But I want to talk about another example. And that is of a climber and skier named Cody Townsend. And if you like skiing at all, you probably have seen some of his crazy videos on YouTube. And he does these expeditions to climb that sort of top peaks in North America. It's incredibly good. Someone wants to throw a link for this. But the thing is called the 50. And there's an episode in the 50 where they hike back into the deep back country, deep in the back country of Rogers Pass, British Columbia. If anyone's from around there, let us know that you're there. That'd be pretty amazing. And there is a ski line called the Comstock Cool Wire. And they attempt to ski this. And they get all the way back into the back country. It's an incredibly difficult sequence. And here they are on the ridge. And they're taking it all in. And their Spidey sense is firing at this point. And you can hear them talking. It's actually an amazing. You probably have to advance towards the end of the scene. It's pretty amazing. And so Cody Townsend and Greg Hill and Sam smooth. And these folks are sitting there. And they start saying things like, well, it's cloudy to the northeast. The slough, which is sort of, I think it's something related to snow. You know, it's like crusty snow, maybe. There's they're predicting they're going to be in a whiteout in the glacier. And they're starting to realize that they should not be moving forward. And they say things like it would really be pushing it. Mount Dawson is stuck in the clouds. Are they going to try to get up there? And they say, we can't swing for the rafters every time. And it's this really awesome part of the episode. And they turn around. They turn around after trekking and getting all the way to this ridge and being close to being able to film. They've got a film crew and everything. And they hit the point where their Spidey sense goes off. And they realize that the conditions are not optimal for them to be able to proceed. And I want to start with this example because what we're going to be talking about pretty much all today is something that I've noticed among experienced product managers and then more broadly maybe experienced product organizations that they build a sixth sense, a Spidey sense to kind of weigh the conditions and think about whether something is going to work out and then maybe shape it in the right direction so that it maybe has a better probability of success. So I wanted to start with that idea. Usually I don't like sports metaphors, but this is not exactly sports. I mean, this is more like an expedition. It is actually pretty similar to product as they go about it. So what you notice when you talk to a lot of product managers is that less experienced product managers tend to adopt this maximization mindset. So they start thinking about things like how much can we do in the allotted time? You know, how many value points, how many boxes can we check? How can we have the broadest impact across all of our customers and then their schedule? You know, how can I hyper optimize my schedule? How can I make if maybe if they're in marketing or whatever, how can I make the message land with everyone? And how can we be good at everything? And it's intuitive in many sense, but it's wrong. As you become more experienced as a product manager, you start asking things like what is the core problem? You know, where can we have leverage? Where is there a one plus one equals three effect? What's important? What's the crux of the issue? What's the kernel of the issue? Maybe with Richard Romel, what must we prioritize? What can we be not that great at? And how can we increase our chances of success when it comes to what really matters? So it's a slightly, it's a different mindset. They're obviously still thinking about value, but the mindset has shifted a little bit. And I'll explain how and why, but this is the critical thing before we start jumping into driver's constraints and flows. I realize here too, some folks might be more experienced but work in an organization that absolutely incentivizes the things on the left so we can get to that in a bit. It absolutely happens. It's not just about the experience of being a product manager. I thought I would draw it out like this and I don't know if anyone has, I won't punish chat with this question, but if anyone as a product manager is sat down in front of a prioritization spreadsheet and each column has a good thing. So there's like a good thing, A, B, C, D, E and F. And then right at the end, there's a thing called effort. And you've found yourself thinking about something that's already been decided on that you're gonna build in your mind and you think to yourself, well, yeah, I think maybe four for that, maybe 3.5 for that. Oh, good, good things, D, E and F, 5, 5, 5. This is gonna be great. These prioritization spreadsheets are, you know, anti-science, right? It's not a great word for it, but there's all sorts of biases mixed up in these. I described them as Willy Wonka in the chocolate factory. You know, if you're familiar there, you know, all these kids, there's a high maximization tendency. Now, is Willy Wonka a murder of kids? Like that's another question that's been debated, but overall he has a lesson and all the kids sort of have this kind of gluttonous sense. And that's kind of how I imagine the less experienced product manager, they're going through and they see the spreadsheet and they just say, we're gonna be great at everything. Now, what's going on in the head of a more experienced product manager? And the best way I can describe it is with this drawing. So here we have all the good things. We have this shape and they're imagining the solution sitting there and it's getting pulled in a bunch of different directions. And what they realize is there's what's called a focus tax that intuitively it might make sense that you can be good at all these particular things, but you're gonna have trouble doing any of this well because you're trying to sort of hyper-optimize across all these good things. So you can see here, they've given themselves a focus tax. They've said, I don't care if you add all these up and if you say that it's gonna be a medium, but I sense just like how Cody Townsend sense that things wouldn't go right, I sense that there is gonna be a problem because it's hard to do all these things well. That's what they realize. It's hard to fit all the factors for getting up the glacier to go up the comms.core to that. So this is the difference. The maximization idea on one side and then the more experienced product manager intuitively understands that there is a focus tax and they sometimes have trouble explaining that to people. And so what we're gonna talk about today is an activity or a tool to help you explain it to people and then also improve your own skills at this. Another way that I've thought about how to describe this is that in this chart here, on the x-axis, let's say we have all these sort of different goals, different problems, different dependencies, different trade-offs, and then on the y-axis we have value. So what happens for less experienced product managers is that they kind of imagine that the more goals you layer on, the higher the value and then it kind of levels off at that point. And then they can intuitively understand that if you try to do too much, meaning if there's too much effort in the project, the line will go down over time. So they sort of get it, but they don't really get it. The experienced product manager understands that quickly, if you have a couple goals and dependencies, you hit this point of value right here and then after that, it's not just that the value goes down, but it actually ventures into negative territory, right? At a certain point, as you add these different dependencies and trade-offs and things onto the effort, the chance of success goes way down and the value goes way down. So maybe that would help you visualize what I'm talking about. There's a slight difference in how they perceive things. Now, one of the interesting points here, the reason why I'm zeroing in on experience, I normally don't do this when I think about experience, but there is a component of experience, I don't know if you remember earlier in your career, someone would, if you were really ambitious and someone would say, well, we're gonna do this, this, this and this. And what went on in your mind? You thought, I can do this. I can thread the needle. I can make this all happen. So here I have an example of a less experienced designer. Sounds good to me. In general, they all want the same thing, right? Yeah, those two platforms that deadline, sure, I can do this. I'm enthusiastic, I can do this. Now contrast that with, let's say, a more experienced designer, the more, as you gain experience in your career, you begin to be able to make use of the additional information. So in this case, the designer knows that they would take a different approach if they knew they were targeting one person or the other. They understand that they would be mediocre if they tried to do something to help this type of customer and another type of customer. So if you see what's happening there, as we gain experience, we gain knowledge about how to make use of the information we're getting. And earlier on, we don't have that. And that's why I think sometimes, less experienced maybe executives or leaders say, well, why are you all worried about this? Well, they haven't been on the front lines. They don't understand that actually the differences matter and you would take a very different approach if you knew that you were doing one thing versus the other. Okay. So what this boils down to, and this is again, this is not just about experience, but I think this is maybe more a combination of cognitive biases. If anyone's a cognitive bias expert, maybe tell me what the real names for these are in the chat, but I think that what we could say is there's three sort of general trends among PMs and then maybe the more experience you have, maybe you're able to battle these a little bit, but we underestimate the value of narrow focus and we overestimate our ability to have a broad impact. So we desperately want to have a broad impact and that persuades us to not take a narrow focus. We underestimate the negative impact of constraints. So we don't realize that adding one or two or three constraints can kill an effort and we overestimate our ability to juggle constraints. Sure, if we just hire enough project managers, we're absolutely going to be able to manage this complex delivery schedule. And the third one is we underestimate the value of flexibility and we overestimate the value of certainty. There's certain a number of cognitive biases that mention this. So we desperately want the answers, we desperately want certainty. And so there's a tendency to push for that when there's a lot of value and flexibility, the sort of two-way door decisions that I think Jeff Bezos or someone mentioned. So these biases are real. It doesn't really matter how much you experience you have. And I think over time, as you gain experience, you gain some ability to be able to address these, but not that much actually. So we still fall into these traps over and over again. Now, I like to mention this because I don't think this is just our work. We do this in everything. I thought of some fun examples. Like here is someone's vacation planner. And I don't know about you, but if I had a vacation that was planned to this degree, I'm not sure I'd have a fun vacation, but it's obviously intuitive to do that. Another great example is when your in-laws, if you happen to have in-laws, are visiting. Do you know that feeling when your in-laws are visiting that suddenly all the constraints and dependencies make your life extremely difficult? Oh, my mother-in-law does not like she spread and she wants to be the one who does the cooking. My mother's Greek, so there's a lot of mom will do the cooking element no matter what. So you see it's too many constraints. It's too much stuff, but we fall into that trap. I love this one from this wedding. There's obviously a lot going on at this wedding. There's a lot going on here. The choice of outfit, the cornets, the little, whatever those things call, the jeans, there's a lot going on in this particular picture. And so as a product, they're maximizing and it's not quite working out. Another example that I liked as I found is that this couple was here to see this migration of crabs. So they put their campsite directly next to this migration of crabs and their campsite was attacked by crabs. That's kind of over maximizing. Family dinners is a great example. I like this from the New Yorker. This is the classic family dinner. Look, this is the sibling on the phone to let this out. This is the cheering on sibling. These two guys actually don't like each other, but they're fighting through their sons or requisite people. There's a lot going on here. So hopefully your Thanksgiving is not like that. But my point is, is that all these examples are sort of cases of over maximization, conflict, too many dependencies, too many constraints, trying to do too much and potentially falling flat. Okay. So which brings us to the topic for today, driver's constraints and floats. It was probably about 2015. I read this book by Johanna Rothman called Management. Manage it, your guide to modern pragmatic project management. And there was just so many knowledge bombs in this book and it's so practical and such a wonderful book. And I've since kind of, I think I've extended a bit on this idea, but it's very important to give a shout out to Johanna Rothman. Most ideas that you see people like me talk about were either implicitly or explicitly inspired by the people often underrepresented who did the work decades ago. So let's be honest about that. So driver's constraints and floats, check out the book. Next, let's get into it. So what is driver's constraints and floats? I'll give the quick definitions and then we can dig into more details. Drivers are the positive impacts we hope to achieve. And you could imagine this by filling in the sentence to increase, to decrease, to advance, hour to maximize. So for example, increase ease of onboarding for SMB customers, decrease the 90th percentile time from sign up to end through revision and template publish. Great, to increase, decrease, advance. Limiting constraints are things we must navigate on our ways to achieving those goals required to need to limited by. So in this case, our constraint by regulatory environment in Europe required to notify customers four weeks in advance. You get the idea. Floats are what we would call like degrees of freedom or areas of flexibility and they increase our range of movement. And this is pretty intuitive because you sort of know it when you feel it. So in this case, we have the company, we have flexibility when it comes to how extensible or integrated this needs to be, maybe with the admin screen and flow coordinator, which is a product that we have. So you're giving yourself some degree of flexibility. And then finally, enabling constraints is a very subtle concept, but they temporarily decrease our range of movement, but they enable us to make progress. So for example, you could agree to not agree. So in this case, they say, for now, let's ignore the downstream reconciliation workflow implications. You could say that's a bit of a float, but what they're saying is, we are creating a constraint here for the purpose of enabling the team. And the difference between that and limiting constraints we'll discuss in a bit, but for a sake of this fill in the blanks we're choosing to for now, we for now let's ignore, later we'll revisit. These are the three central ideas that you Hannah Rothman had where drivers limiting constraints and floats, and I added enabling constraints because I thought it was an important component and something we've talked about a lot in complexity, for example, but these are the four central ideas. Let's just go into some examples, talk about how we put them together. So here is the vacation example, we're going back to it. So I want you to think about this vacation and on the left, we have one plan. They want amazing food, for example, Michelin star restaurants. They want outdoor adventures, great for kids. They want romantic time together. They want to see all of the historical sites and they enjoy meandering, walking around the town. And their limiting constraints are two weeks in July. We know July is a great time in Europe. They want to visit their relatives while they're in Amsterdam. They must bring two-year-old and the eight-year-old. They're feeling feeding schedule. They have to feed the two-year-old. Oh, they have seafood allergies too. And they thought it would be a good idea to invite their grandmother who's an absolute foodie. Okay, now compare that to the right, which is we want to have outdoor adventures with kids. We like meandering. We still have to go two weeks in July. Yes, we have seafood allergies and we are going to bring our two-year-old and eight-year-old. Now, anyone have any bets here? Oh, I should have done a poll for doing these things. But anyone want to have a guess here, what's going to be more likely to succeed overall as you go through? Has anyone been on the leftmost example? I've certainly gotten close. Well, you get the idea here. You intuitively sense that the vacation on the right is a bit more feasible. And it's more likely that you're going to end up really having outdoor adventures with the kids and meandering versus the example on the left. You can imagine it's going to drive everyone insane. Okay, so you get the idea, hopefully, drivers limiting constraints, drivers and limiting constraints. Here's an example of drivers, but how there could be too many of them. This poor product manager is here and they're wondering, you know, someone's, maybe this is their internal voice, but it could also be the external voice. So we want to make sure they know about the silver plan. We want users from channel X, we want existing users and savvy users and make sure we collect their information to do the remarketing brand appeal. The product is easy, get them doing actual work, get the data into the product, get them to add coworkers, they're going to experience the breadth of the product. You can intuitively tell that there's going to be issues here. That's why our puppy product manager is worried because you can tell, these are drivers that they have, but you can see that there are too many of them. Okay, similarly, they've got their raincoat on. Now, let's think about some constraints. You know, we have to use framework X, we can't bother Mary, Darren's going on vacation, we have to use the old blue logo color, that's an inside joke and amplitude. We need to funnel people to pricing plan XV3 because V4 is not ready. Hopefully this isn't inspiring too much anxiety among PMs because we deal with these, my favorite, oh, you can't use the back browser button. Yeah, that's like a classic one. All right, so you get the idea, these constraints feel a little different and there are too many of them, right? We can intuitively sense that there's too many of these. So drivers, these are things that are valuable, that have impact, but there's too many of them. And in this case, these are things that are constraints and there's too many of them. Why is this important? This was my critical understanding, reading, managing it, Johanna Rothman's book. And she makes the bold claim in that book, and we should just take this, not with a grain of salt, but you have to scale it to your organization, that she says basically that if something has more than two drivers and more than two constraints, you are only able to be successful on those with sheer brute force attention. So there is a way to thread the needle when you have a lot of drivers and constraints, but it's really hard. The fewer the better, and at a certain point, your chance of failure goes up to 100%. So I use numbers like two, four, six, eight, but you get the idea that if you have a total number of drivers and constraints like two, you have a pretty good chance of success. And as you start to add these, your chances go way down. And her point is that, sure, if you have 55 project managers and you try to dot every I and cross every T, you might be able to coax this in, just like if you have a vacation planner and if you have three people to help you, you might be able to get through that vacation, but this is really hard to do that. So this was the critical understanding when I read the book was like, wow, okay, seems intuitive, but you layer more of these things on and that gives you a sense of what's not gonna work. What I realized later in my career was that the best PMs do this intuitively. They don't know they're doing this. They don't say, well, I think there's too many constraints and I think there's too many drivers. They just get it and they've intuited this concept and that's what gets them to kind of keep laser focusing in on the particular problem. The major difference if you think about it is that the less experienced PM is thinking effort versus value. Oh, team, do you think that will take two weeks or three weeks and then they think about the value? The experienced PM understands that in many ways this might sound heretical, but it's more important to shape your efforts and make sure that the driver is clear than it is to know the estimate for the thing or try to estimate value or try to estimate effort because they are layering in the dependencies and limiting constraints intuitively. And so instead of playing spreadsheet bingo with prioritization, they are understanding that the shape matters and it's better to have 10 well-shaped efforts with clear drivers and fewer constraints than it is to have the thing that checks all the boxes. So our goal is then to limit the number of drivers, limit the number of limiting constraints, so limit the number of drivers, limit the number of limiting constraints, increase the floats whenever possible, and then artfully use enabling constraints to make progress and learn. I would, my one way that I describe agile to people is actually the artful use of enabling constraints. The goal is not to sit there and break up your projects into 55 parts of exactly two weeks and deliver n number of points into do all this. That's crazy talk. But if you think about agile at some level, it's basically about artfully using constraints like batch size, regular integration, maybe even pairing with people which seems counterintuitive but might be more beneficial. There's just a series of artful enabling constraints. What you find in the agile world is that people have forgotten that and these things become process hoops instead of techniques and tools to get some desired effect and we'll get into that in a bit. Okay, here is, let's get down into the exact details of this. So when defining drivers, what I've noticed is they can be very difficult to define drivers because the intuitive thing is it would be something like delight customers or it's not very clear what the driver is. So I came up with this pretty simple tool you can use nothing super original here although I just came up with it. I'm sure it's coming from somewhere. Here's a way to think about the drivers. So right now, the template that I've, you have access to the mirror board, we can share it again and I have a bunch of, you can copy these as much as you want. But what I do in the workshops is I force people to be explicit. So instead of saying, what's your objective? I asked them to use this template right now we are working to increase the and I have a bunch of words. I could have done like 50 more but that would have made it too overwhelming but these are things like efficiency of productivity of extensibility of predictability of reliability of and what I'm doing is I'm forcing them to be explicit about the variable. So in this case right now we are working to increase the ease of integrating the SSO SDK. This was from a workshop just a couple of days ago might be common for people. For partners in our XXX group maybe they've got some integration group which we could potentially measure by tracking these things. Why is this important for drivers constraints and floats is one, I didn't mention anything about what they are delivering. There's many ways that you can increase the ease of integrating the SSO SDK and I asked them to be explicit. In this example right now we're increasing the probability that let's see, customers will create greater than or equal to two functional integrations within 15 days of first product use for customers arriving through maybe paid search sorry organic search which we could potentially measure by doing these things by channel time percentage to achieve the thing. So this can be a helpful way for you to think about specifying your driver because a crisp driver definition never goes out of style. Once you get that right a lot of things fall into place. You can also ladder these, right now we're working on to increase this and if we are successful I would expect to see an increase in and you repeat the process that's more advanced for another day but if you're feeling ambitious you could do that particular thing. Now when I present that the natural thing and this is Adam Taylor I'm picking on him because I tried to explain this and he said it doesn't make any sense he brings up a situation where I've read the material but let's say you need to ship by ex-date due to a legislative reason that would make the data driver but also constraint. So dates are often get people tied around with this they wonder where dates figure in. I'll let you read Johanna's response but what she basically says is that she, I love what she does here she mentions that she has two clients and one misses deadlines to their advantage because they're able to produce revenue and one client number two doesn't actually treats the deadline as a driver and what we find when we do that that client one date is a float not forever and for client two date is a driver so what this indicates is it really depends on what business you're in for fixed price project oriented work your ability to hit the date might absolutely determine your margin in product work, time and date is often a float to a point a deadline might be an enabling constraint to help you learn faster but it is not the driver and this is what's different between this and what's commonly known as like the iron triangle of project management why because well iron triangle of project management was largely for organizations working on a different type of work. So you can see how deadlines fit into this so we're doing on time. Okay, we're going, we're going. The next concept here is the idea of a limiting constraint I tried to pick some examples one that guy has to walk he must drive over the burning logs although you could argue that that's what makes him have fun. This other person is the person who has the smallest department in New York it certainly makes it harder to have a dinner party and then one of my phase pictures of safe of a visual board with all the dependencies between the work that should give people anxiety for sure. And then I love this quote I posted something on LinkedIn that said what tools do you use and Louise mentions that her corporate environment couldn't use any of the tools that people mentioned and she said, unless it's Microsoft it's like pulling teeth to get access this limits collaboration, creativity and general communication severely limiting constraint. Just using Microsoft is a limiting constraint on collaboration, genetic creativity. So Louise might mention for her effort that we are forced to use Microsoft products because it's a limiting constraint. How do limiting constraints differ from enabling constraints? So here's an example back and forth so limiting constraint will be something like we are required to request X from Y. An enabling constraint would be focus on tier one accountants for now. So you can see I prevent these little things like we are required to need to or constrained by or limited by enabling constraints are different and they use words like for now we're choosing to for now we for now let's ignore later we will revisit. So for example, we've run out of budget for X in 2022 could be a limiting constraint. However, a team could say, you know what? We're gonna limit this first effort to $100,000 and then reassess that's an enabling constraint because they're using that as a forcing function or creative constraint to help them make progress. So the distinction here seems subtle but it's not when you think about it enabling constraints are meant to kind of move you forward and make progress limiting constraints are things that to Louise's point she would ideally not have to navigate around. So obviously we always have limiting constraints but enabling constraints are something that we can use in a valuable way to do this. Floats are where we have flexibility. This is pretty self-evident things like we don't need to ship to all of our customers at once our budgets flexible. One of my favorite from marketing and amplitude is we don't need this message to land with everyone. That's in a wonderful float because it kind of releases you. And so what this does is it provides us flexibility where it matters. And you know, people could say, well, isn't this kind of common sense? How rarely do teams explicitly mention where they believe they have flexibility? Not very often, right? So it's often like here's our objective here's our key result but that doesn't really say here's our objective here's our areas of flexibility here's the constraints that we acknowledge and then here's the enabling constraints that we're gonna use, it gets pretty interesting. So what this becomes, so when we talk about shaping an effort to have your bets count the fundamentally means is that in a before shaping environment you have lots of drivers you have lots of constraints and not many floats and after shaping we are shifting things to the right. If fewer drivers, fewer constraints, more floats more enabling constraints. And as I mentioned before, I don't care what anyone's level of effort and value is but if everything on your chart is poorly shaped you're gonna be out of luck. There's like an 80% chance those things are gonna fail and not really have the desired impact for doing it. So by shaping things by talking about this you're able to get a lot closer into doing it. So I'm gonna end with this example. So we have time for questions, right about 40 minutes. And so this is an actual one that I created yesterday for an initiative that I have at Amplitude. So let me describe the problem. I really love doing these webinars but they're long they can be kind of inaccessible. I love making templates but templates without anyone to help them are pain in the butt and you don't really know how to use them. So I'm up at all hours the night doing these very sort of synchronous activities and not really thinking about how to make things more async. So my initiative name is John Cutler's Async Learning Experiences Experiment. So what's the driver? The driver is we wanna position Amplitude as a trusted expert, not just a template peddler and not just, oh my God, it's the 55th effort just as value blog post or something. We really wanna be seen as someone that we're there to help our customers that were trusted expert. Seems kind of nebulous, but it was important to put in there. Now the real crux of this effort is create return traffic, real engagement asynchronously with helpful templates, job aids, videos, et cetera. The key driver here is that should it be async and there should be this kind of, you know, people don't need to be scrounging Amplitude for the right content here, you can find it. So we had some early success with the Amplitude North Star Framework Library and Miro and so it inspired us to do something more. So those are our drivers. What are the limiting constraints? At the moment, I'm the only one who can do it. Although if you're interested in maybe partnering with us, maybe we could work together. The other constraints is it does have to live on Amplitude.com, we wanna use Amplitude for analytics because that will be a helpful learning tool and obviously we want all the normal privacy, cookie control, et cetera. Now notice how I put these all in one group. You could be really pedantic about all your constraints but largely this feels like kind of a single constraint in the sense that it has to be integrated back into what we're normally doing. John can't go rogue. John cannot put the Miro library again. That was the word I got. Okay. Now look at all the flows, how coherent it is, the topics we choose from, the scope of the modules, the expertise of the people we target, whether we even attempt to capture leads, maybe we don't even try to capture leads and maybe we just hope that you'll like this. If you like these talks, let the world know, I guess. It's flexibility. I have flexibility to work with an agency. So there's a lot of flexibility in how I'm gonna approach this. What are my enabling constraints? One, we're doing a time box of eight weeks that started yesterday. Two, we're gonna get something into customer hands very, very quickly. We're gonna make sure that it lands. I am not doing any fancy master classes. So I could go crazy trying all these advanced video tips, but I'm not gonna do it because I know that that will not give me the best results. I'll rabbit hole. So I'm gonna create an enabling constraint, no fancy video. And then our other enabling constraint is I will not work alone. Because I know that if I work alone, I'm more likely to go rogue. So since we have flexibility to hire an agency, we're partnering with this company called Wavetable. These people are great and I met with them yesterday and it's cool and I have this team to work on here. You can see some other links to this. So that's a real example of one that I just was working on last night. And I've left you all an empty template that you could copy into MIRO if you want, or if you could screenshot for other things. So to wrap up before we get to some questions, the fundamental shift, you could forget about drivers and constraints and floats. They're just a tool to achieve these things. The fundamental shift here is this shift from this idea of maximization, juggling, peanut buttering. So, trying one message that will apply everywhere or trying to solve simplifying like that. Chasing certainty, it's the shift from those things to the right, much more about intentionality, single threading, so that focus and what I would call strategic ambiguity. You could think of floats as agreeing to be ambiguous because we'll produce better results versus what happens in most companies is that they have damaging clarity. All this kind of premature convergence on what you're gonna build, et cetera, to do these things. Why? Often because they have dependency constraints and therefore everyone needs to know exactly what they're gonna build so that some poor shared service team who's juggling working for everyone can actually help them. Another topic for another day. Okay, great. So we're gonna go to some Q&A. I have some Q&A here from folks who use the form beforehand, but I'm gonna weave in this with some questions that we received in the Q&A and I will stop sharing now to do that. Hopefully this was helpful and interesting to folks. Oh, hey, Mari, can you turn off your video share with the A or maybe other folks? And it doesn't matter. Okay, let's jump in to these particular things. Okay, hopefully everyone can see me. Great, let me go to some of the questions here and then we'll go through some of them. Okay, David Harvey asks, are options floats or enabling constraints? That's an interesting question. I would say that optionality, when I think optionality, I don't think of them as enabling constraints unless you had just a few number of options. So for example, if you said for this effort, we're gonna constrain ourselves to option one, two and three, that feels more like an enabling constraint. If you had all this optionality, like what we could go as deep as we want into visual design, we don't need to worry about that for now, we don't need to worry about that for now, et cetera, that would feel more like floats. Hopefully that helped, David. Answer live, done. Okay, embrace your enabling constraints as Clay, I agree. Okay, we'll see. Oh, he has a question up above. Okay. Are enabling constraints a deliberate focusing? Yes, I think that they can have that effect. They can also just be things about how you work. So for example, I use the example of pair programming. A lot of people hate pair program. Yet the people I know who like gave it a chance, they realized after a little while, they're like, you know what, I thought I'd be way slower, but this is actually a lot more effective. And I spend a lot less time having to do these other things. So I would say that some enabling constraints help you focus on certain things and some things are more just related to how you work would be another way to put it. Or rubber ducking, okay. Oh, that's good. Tiffany Chang mentions rubber ducking, it's very funny. Okay. Okay. Could drivers also be about decreasing something? Yes, absolutely. It's so funny I debated, you know, whether it's saying increasing or decreasing. I stuck with increasing because you could flip around every decreasing comment to create an increasing comment and everyone likes things to go up, but you could obviously increase, you could obviously decrease the time to something. Yeah, that's what you could do. So let's answer that. Okay. Let me jump to one of the questions that we received beforehand just to mix it up. Oh, Tiffany asked, from your experience, which business functions or rules have the most challenge when trying to come up with floats? Why is that in your opinion? I'm gonna be honest, I think it's the business functions that have to deal with the BS. No, what I mean by that is that like, you have a product team and they're like, well, we're not sure what they're gonna release and for, you know, there's a float with the time here. There's a poor marketing team or someone trying to coordinate with them for that release. And legitimately they're gonna be inconvenienced by that. And so I think that we need to be empathetic to the other groups in our company and then sort of form a partnership with that group to explain why the float matters, how you can mitigate their particular fears. Maybe there's some enabling constraints you could create to make sure that you're integrating the knowledge with that team so they're not left hung out to dry on that. Kind of see where I went there. Do you see like enabling constraints could solve that problem of that particular floats by creating that integration? Okay, and Abhi asked, how do you build confidence in making strong bets and persuade your stakeholders when moving with imperfect information? Show, don't tell. I mean, if there's no track record in the business for watching any bet land and having an impact, you're gonna have trouble. And that's why I love amplitude and I work at amplitude, right? Because there's a little bit like, even if you're in a feature factory and shipping, shipping, shipping, it does not stop you from measuring impact. And then sometimes the way to have this discussion is like, look, we've shipped 10 things and none of them have had any impact. And therefore I think we should try it a little bit differently this time. Hopefully that helps. Okay, let's jump back to see if there's any questions in the Q and A. Oh, there's a couple more. I'm sorry, wow. John, Andy Nelson, what's the driver constraints for hanging the pictures in the background? Okay, my float, let's go through this. The driver is my son and my family and work now at the moment. The constraint is, I don't have those little, if someone wants to mail me the hanging screws, I don't actually have them in the house. So that's gonna be my excuse for now. The float is that I'm not too concerned about where the pictures go. And the enabling constraint, I don't know, maybe Andy, you can work with me and be my accountability partner and I will hang up. Okay, great. Arlo Boras asked, use fortune functions. I've used forcing functions in the past and just realized that they can be limiting constraints or enabling. I'll select them more carefully, bingo. That was me too. So Dave Snowden, and there's this sort of complexity thing in Kinefin, which is this complexity framework for understanding systems, kind of brings up that point around these enabling constraints. He has a bunch of different types of constraints now. You should look at his most recent work, which is extremely fascinating. But that was the big eye-opener to me that like most of the, most of the, for example, like the more pure, agile ideas, not the stuff that's been cookie-cuttered, corporatized or whatever are about enabling constraints that then have positive benefits. But we flipped it around somehow to limiting teams. How odd is that? It's just weird. So, yeah, Arlo, I came to the same conclusion. Another way to think about it is I come from a music background. You could walk into a studio and they don't really have enough instruments there. And in some way, you're constrained or maybe by the equipment. But then at the same time, sometimes when you're recording an album, you force yourself to only have a certain amount of equipment and record in a really pared-down studio so that you don't get distracted by all that stuff. That's another example. Okay. Whoa, oh, questions. I'm excited. There's a lot of questions. Okay. Andy, what's the driver? Oh, no, the wall. Okay, I've done the wall paintings. Okay. So Mike asks, would you consider external departmental engineering product teams within a large organization as a limiting or an enabling constraint? Whoo. All right. Let me give you the good example. Let's take Jack McCloy and our design systems team at Amplitude. His goal, he doesn't force people to use the design system, but if they buy into the idea of the design system, I believe it creates a little bit of an enabling constraint that moves them forward. But he doesn't force it on people. So it's interesting. In some ways, the choice to use the design system is a float, but if people buy into it, it's an enabling constraint. And I do believe that in large companies, sometimes they're those teams that their immediate value is not known, but they're kind of creating a consistency or leverage that raises all the ships. And so for example, if your external department or engineering product within a large org is continuously slowing you down and there's no benefits from working with them and there's no benefits for the customer, it's a limiting constraint. But if using their service also feeds into the general good and helps maybe get that team, get resources and allows maybe platforms to be built that raise everyone, then I think it's an enabling constraint. It's a super, it's a fine, just the conversation could be helpful within your company. Just try that to do it. Okay. Brady Halahans says, would you consider difficult stakeholders, micromanagers as limiting constraints? Absolutely. How many startups are actually limited by the CEO? Like at the end of the day, if someone needs to review everything, it could be a limiting or enabling constraint. My book, it feels a little bit more limiting, but yeah. I'm not sure you want to go and like list that on all your boards at first, but you might want to figure out a creative way of putting that Brady. Funny. Okay. All right. More. Okay. Let's let me jump to one of the questions was asked before. I'm not picking, I'm picking the ones that have to do with today's topic. So I'm skipping some. Now I'm realizing that asking questions in advance is a little hard unless we focus you on an article. Okay. Some of these are really good questions. Okay. Oh, here's one. Do you have any advice for how to pick in action the best bets when the company sees the roadmap as a commitment and requires unmovable deadlines, delivery dates and a business case for every feature? Man, that's going to be really hard. I think that the only way there is you can't dress that head on. You're not going to rationalize in that situation. An example being that in their mind, the annual roadmap is a driver. The predictability is a driver. All their pre-planning and annual things are enabling constraints. And that's the funny thing, right? Like that actually boils down the issue here, right? So in that environment, they literally don't see the, they don't see that those ideas are limiting. They see them as enabling because they're chasing certainty. And so the only way to win in that particular situation is you have to show, not tell about how these slightly more open-ended, open-framed bets. You might even bring this framework to them and talk through some of those. Can you get a couple of efforts through the pipe there that are a bit more open-ended? What I said to someone the other day is could you shift this prefixed project to maybe a two-quarter effort that had a little bit of flexibility in terms of the solutions and then do that well and then try to demonstrate that it worked? Lots of questions. I'm sorry. I'm not getting into a lot of them. Okay. Sasha Colicut says, what would you suggest if the response to the issue of too much stuff going on to many drivers is to just hire a heap more product managers because then you get into issues of scaling comms and coordination within the product group? Yeah, this always weirds me out when you see a ratio of it's like, oh, we've got 50 product managers and a hundred engineers and you're like, what's going on? And what they're doing is you need, it's a shit umbrella. That's what it is. It's they're hiring all those people because they're in a wicked loop that because nothing can happen because they're not investing in a resilient system they need all these people to play defense. And the more people you have to play defense the harder it is to get anything done and it just becomes a wicked cycle. So yes, I agree with you that happens. What would you suggest beyond scope for today? Other than I know what's happening. Don't hire so many more product managers and also ask yourself maybe the part of the answer there Sasha is listing all those constraints to just make that really clear. Like what is the underlying problem that is causing to them to hire the shit umbrella group? And could you surface that and you walk into a meeting and say, I'm trying to shape my bet, but look at all these constraints and I have no degrees of freedom. Let's talk about that. That might be a way that you could have the conversation. So it's a tough one. I used to say, but just make the work visible and everyone will understand but I don't think that that's a hundred percent it. You know, there's gonna be a lot of political stuff to unravel there because people self-identify with being the shit umbrella. And then you ask them that their job has to be different and that's a big shift for them. Okay, Clay Vic says, you mentioned said four and less drivers and limited constraints and then I heard 80% failure. Any personal or private studies to give some? Yes, so let's go back to Johanna's book. So first of all, when I listed mine, what you noticed, I was kind of grouping them. I could have framed them with what the general constraint was. So I think that we probably have two and two or one and one or two and two or two and one. It's the spirit of it that counts and in that book manage it, I think that she references a famous study on projects that dug into this and demonstrated and actually there was some study on this. And so I'll have to go back and find it. I bought it off of Lean Pub and then that PDF is sitting on another machine somewhere and then I couldn't get back into Lean Pub. Okay. Oh, we're at time now too. I have two more minutes. Oh, Andy's gonna send the monkey hooks. Yeah, sort it out through Amplitude. This would be great. And I think we're good. I think we're good for now. I think we'll call it a day. We'll try to get to some of these other questions as we go through. This was great. Thank you so much for attending. Do me a favor. You have access to the mural board, screenshot stuff from there that you like, share it. Spread the word and keep a lookout for these more sort of self-contained, shorter learning experiences that we're gonna be doing. Hopefully that will fill a void in everyone's busy schedule. This is obviously a long session to sit through to do it. Okay, I have a workshop to do in 60 seconds too. So okay, thanks a lot everyone. Always fun to do these and yeah, I love it. Okay, talk to you soon. Bye.