#54 – Startup org design at 50+: Team leadership, setting objectives, coordination, & performance evals | Aaron Dignan & Sam Corcos

When overseeing a startup, a CEO needs a well-functioning organizational design. This helps evaluate the performance of the team, make sure functions are coordinating and executing efficiently, and communicate and set objectives. In this conversation, bestselling author of Brave New Work, Aaron Dignan, talks with Levels CEO, Sam Corcos, about performance management, accountability versus responsibility, and the power of mixed roles.

17:29 – Don’t turn performance management into a binary

Many organizations view their employees in a binary: they’re either top performers, or they aren’t. But performance isn’t always black and white.

The one thing that I have noticed throughout my career is that organizations love to do this thing of, these people are top performers or high potential employees, and these people aren’t. It’s binary and it’s very simple. The top performers get to go to Hawaii and the other ones get a performance plan. My first question to you would be, for Levels and for you, what does performance mean? What is performance? Because actually, this feels a lot like the budgeting question for me when people are like, “Should we budget or should we not budget?” I’m like, “What is budgeting? What are you actually trying to do with the word budgeting? And then let’s talk about it.” And in actuality, most people are trying to do three things with budgeting, not one. And then you can talk about the right way to do each of those three things. I think the same story is true here where it’s like we use this catchall phrase of performance management, but we then have to break it down. So how would you define performance? What does performance look like?

21:21 – Commitments and outcomes

Performance is a spectrum based on commitments and outcomes. Are you doing what you said you’d do? Are you playing the game according to the rules that we play?

Performance, to me, is broad spectrum. So if I have an expectation, I don’t care if that’s a principle level or a value level expectation or a mathematical expectation, it’s still just, are you doing what you said you’d do? Are you doing what we said we do? Are you playing the game according to the rules that we play? And then are you able to deliver? So part of that to me is commitments and outcomes, which is, what are you committing to? What outcomes do you need to deliver that you’re bought into? And are you able to do that? Part of that is actually more about developing mastery. So what is the vector of your development, your skill development, your rate of improvement? All that stuff matters too, because even if somebody can do what they say they can do, but they don’t grow at all over five years, that’s probably not a great sign and it’s going to sort of stall out the culture in some way, shape, or form.

22:10 – Performance is a team sport

Performance shouldn’t be determined by one person. It’s more of a team sport than an individual one.

What’s weird about performance management to me is a couple things. Number one, it generally tends to be unidirectional. So it’s, whatever this leader thinks about this individual is performance management. In actuality, it’s like performance is a more of a team sport than an individual sport. There’s very vanishingly few things that we do alone completely that result in actual business performance. Most of it is collaborative. And then once we are doing it, who decides what good looks like? And the classic example of this in my business, in facilitation is, someone who has a certain kind of identity and background and life story will say, “This is how you should facilitate. You should be strong and vocal and loud, and you drive the room.” And then someone will come in with a different background, a different identity, maybe a different gender, identity, color, skin, et cetera. Maybe they have a totally different style. So it’s like, “Oh no, you’re not performing as facilitator because you don’t facilitate the way I facilitate.” But actually, maybe the audience is happy. So a lot of it boils down to who do we ask. So a lot of performance for me is about reputation. We’ve struggled with this too in different periods of our history, but as much as possible, we’ve tried to build systems where reputation drives a lot of outcomes for people, how they’re perceived by their peers, who wants to work with them, which clients want to work with them, at what level, how much demand on their time and existence there is. Those social signals tell us a lot.

37:10 – Create a culture of constraint

Rather than forcing people to ask for permission before every action, let them lead the way and only add constraints when necessary.

Clarity around decision rights is absolutely key. And I would say what that looks like to me at an ultra, ultra varsity level is, it’s not just all decisions, it’s in which decision spaces do I reserve the right to be CEO and in which decision spaces does it benefit us to make decisions a different way? So we have essentially, if you think about all the authority in the organization starting here, what gets distributed and what doesn’t and how do we claw back things that we disagree with is a really interesting part of the journey. And also, I’ve talked about this a little bit, but for anyone who’s listening, the difference between creating a permission culture versus a constraint culture. Because if you’re in a permission culture, then you can’t do anything until you’re told you can. But if you’re in a constraints culture, you can do anything until we say you can’t, which means there’s a lot of authority at the edge. Most of the agreements and decisions we’re making are actually about restricting authority or focusing energy or clarifying or constraining what’s possible either with a policy or a process or a role or whatever the hell it is. That’s a very different motion than folks are used to. So it’s clarity number one, and then number two, actually getting into the subtlety of clarity about which kinds of decisions and what are the best ways to make those decisions.

42:59 – Don’t micromanage

If you create space for your team to operate in, don’t go in and change things constantly. It’s your job to elect the right steward, not micromanage every team.

I look at the decisions that each successive team, whether it be the leadership team, the CEO, the board, whatever, every successive team all the way down, the decisions they’re actually making are decisions directly related to what they’re doing. And then decisions about which decision spaces and purpose spaces they’re carving out for other teams to operate within. So it’s more like a two-step accountability than a one step. So you’re talking about they make a decision, they bring it to you, you don’t like it, you change it. That’s direct connection. The way I think of it is, I’ve given the studio circle, along with my other leadership team members at what we call our source circle, we’ve given the studio circle that produces all our content a purpose around building the brand through content and some authority and decision rights around what we produce and where we produce it and where we send it and all that shit and even a budget around how much to spend. Once we do that in our model, I do not have the right to interrupt or disrupt that process except by changing that charter or who fills that role of stewarding that circle. So I can be like, “I don’t like the last three decisions that y’all have made, I’m going to make a proposal that we reelect a new steward for that team.” Or I can say, “I don’t like the last three decisions y’all have made, so I’m going to propose that we change the charter for that team.” But I can’t just go in and interrupt a decision in flight. It’s the design, it’s the blueprint of the organization that we actually have direct authority and accountability over. The executional reality is one step removed.

51:27 – The hierarchy of work

Determine who is responsible for which tasks and keep your hierarchy clear. This makes it easier to allocate responsibility and keep projects on track.

At the end of the day, what we’re talking about is who’s on the hook to actually do it and make it happen. For us, that’s always one role or one team. So that’s clear. If we’re wondering who is responsible if it doesn’t go well, who needs to fix it, it’s pretty easy to just say, “Well, is the role in a team or a circle? Great, then it’s their problem. Is that nested in a super circle? Great, then it’s their problem.” It’s very clear always where the work lives in the hierarchy of work. So if something is going haywire in our growth team, we’re going to hear about it in the source team and then we’re going to have to be ultimately responsible for fixing it. And if that doesn’t happen, then the board’s going to hear about it and be responsible. So that’s very similar, but we don’t feel the need to write that down for every freaking project we do or every role we create. It’s obvious because the structure is transparent and intuitive.

56:16 – The power to object

Instead of gatekeeping the decision-making process, allow projects to move forward unless there is a reasonable objection.

Most of the power that happens is actually the power to object, not the power to consent. So things are going to happen unless it’s proposed and somebody’s like, “I don’t think this is safe to try.” And that’s a power that typically within a team is shared. Every role has that capacity to be like, “Wait a second.” Because frankly, we play from position. So if someone in a finance role and someone in HR role and someone in a growth role and someone in the sales role are seeing this thing, this idea, I actually want them each to look at it through that lens. And if sales is like, “This is a terrible idea,” I want to hear that. I want to actually play with that a little bit before I try to say, “Well, I love it, so it’s happening.”

01:05:08 – Find the right role

Sometimes, if a person isn’t doing well in their role, it’s not because they’re a bad employee. They just need a different role.

You’ve heard me barking about roles for a long time now, but it’s another place where I think roles are really helpful. Where it’s like, in this role, this role was unable to do this, or this person was unable to inhabit this role successfully because of X, Y, and Z, as opposed to saying, “Phil’s a low performer,” because Phil’s not a low performer. In the right situation, in the right context with the right instrument in his hand, he’s a high performer, but in this role, it wasn’t working and here’s why. And then people can be like, “Oh, shit. I see that that’s what’s expected in that kind of role and why there was maybe some issues there.” If I want to know more, I can go find out from the team, but just something about that gives it a little distance where it’s no longer about the person, it’s about the context. And that’s what really matters anyway. It’s like, “This was not working out and now we need to find someone who can inhabit this role and kick ass.”

01:06:45 – The power of mixed roles

A successful organization will always outgrow its people. This is why it’s so important to put people in mixed roles, so that when one role becomes redundant, they can move to another.

You want the organization to outgrow the people because you’re so successful in each of the stages. And that’s one of the reasons also why we pride ourselves so much on a role mix rather than a role singular. Because it may be that it’s like, “You know what? We’re pulling you out of this role because you’re not performing there. You’re really performing in this role and we’re going to build another one like it that we can put you in, or we want you to go part-time, or any number of other possibilities.” But suddenly, we have more tools in the toolbox to play with to get maximum performance. And that’s why when people ask like, “How do you achieve the levels of profitability that you do? Or how do you achieve what you do with the number of hours you put in?” I’m like, “We just get more out of ourselves because we don’t treat ourselves single action figures in the box.” Every skill that we have is deployed against a mix of roles over time. As we get better, the mix changes and the rules change.

Aaron Dignan (00:00:06):

When people ask, “How do you achieve the levels of profitability that you do? Or how do you achieve what you do with the number of hours you put in?” I’m like, “We just get more out of ourselves because we don’t treat ourselves like single action figures in the box.” Every skill that we have is deployed against a mix of roles over time. And as we get better, the mix changes and the rules change, to your point. It’s absolutely the case that roles are going to outgrow you or morph out of you. As long as you’re a player on the team who knows how to find and create new roles for yourself that add value, and in our system, you can literally propose them. And then suddenly I’m in a role where everyone’s like, “Wow, that’s the ideal scenario.” So there are some people that are not in a place in their life or with their skills where they can hang and they do have to go, or that they are disingenuous about commitments, they have to go. But a lot more people are just accidentally in UX when they should be in UI.

Ben Grynol (00:00:59):

I’m Ben Grynol, part of the early startup team here at Levels. We’re building tech that helps people to understand their metabolic health, and this is your front row seat to everything we do. This is A Whole New Level.

(00:01:24):

Building companies is inherently hard. When you’re working in a startup, things move very fast, teams grow and teams scale, and you can maintain the vision and maintain the values. Those are things that hold true. The culture is spread from one person to the next. It’s something that we talk about with different thought leaders on an ongoing basis. But when it comes to org design, it’s something that if you’re not intentional about it, it can be an afterthought. So Sam Corcos, co-founder and CEO of Levels, sat down with Aaron Dignan, author of Brave New Work. We ended up reading this book internally and there were so many takeaways, so many things about the way to think how really a business has an operating system, a way of working. And if everybody agrees upon that operating system, well, things can start moving pretty fluidly.

(00:02:09):

So, Sam and Aaron sat down and they really dissected this idea of org design, what does it actually mean and how can you think about having teams that essentially work autonomously and independently to keep pushing work forward and getting things done. The idea is to constantly remove bottlenecks and do what we do, focus on the work. Here’s where they kick things off.

Sam Corcos (00:02:35):

We’ve read your book. We’ve implemented a lot of the ideas. There’s a lot that we’ve liked about it in terms of, it’s certainly aligned with a lot of, we’ll call it the more idealistic way of building a company. So what I really want to talk to you about is, we’ve started implementing a lot of these things, and where the rubber meets the road is a lot more friction and complexity than maybe one would hope for in doing that sort of thing, I think. In some organizations, it’s just like cultural resistance to change, people don’t change. That’s not really the case for us. People actually really like the ideas. I think the biggest category that I wanted to get your thoughts on is the systems, we’ll call it the API model. Are you familiar with the… Many companies have these different types of organizations. Uber calls them platform teams and project teams.

Aaron Dignan (00:03:57):

Sure.

Sam Corcos (00:03:57):

Are you familiar with that divide?

Aaron Dignan (00:03:59):

Yep.

Sam Corcos (00:04:00):

SpaceX has something similar where some organizations, hypothetically, it might be the thermal team, which is they don’t quite have enough work to just be completely independent, but every piece of the puzzle has to check in with the thermal team to make sure their stuff doesn’t melt.

Aaron Dignan (00:04:19):

Right.

Sam Corcos (00:04:20):

So they work with a consultancy and they have this API model like you described in the book. But what we found is that, that works really well in a specific set of circumstances, and it seems highly dependent on people performing extremely well and being very responsive because there’s little to no accountability when one of those APIs or consultancies doesn’t perform well.

Aaron Dignan (00:04:51):

Right.

Sam Corcos (00:04:51):

Because hypothetically, if the consultancy is accountable to me, but I don’t really work with them, I’m not paying attention. So these other people are not getting what they need, and they get really demoralized and they can’t fire them. They can get them feedback, but they don’t really have to listen. So the only way for them to get anything done is through what feels like this back door politicing of whispering to the CEO to get people in trouble. And that feels really bad.

Aaron Dignan (00:05:19):

Yeah, totally.

Sam Corcos (00:05:19):

Whereas if they were on your team and they worked for you and they reported to you, you have much more authority and the ability to remedy that situation. So anyway-

Aaron Dignan (00:05:30):

Yeah, even if they just are on your team.

Sam Corcos (00:05:32):

Yeah, exactly. So I’m curious what your thoughts are on that and if this is just an inherent fragility in that kind of a model.

Aaron Dignan (00:05:43):

Yeah. Well, I mean one of the primary trade offs and pendulum swings and dynamics that we see in org design is between centralization and decentralization between functionalization and functional integration. What tends to happen is, when we have functional organization, more like an Apple model or something, then it becomes all about, well, these functions have to collaborate really well together, which means they have to be highly coordinated, which means people have to do a good job and communicate well and all that. And then when we actually need to get something done, it requires an elaborate dance to make it happen. You can’t force it with reporting lines, and you also can’t force it with just identity.

(00:06:26):

In the functionally integrated system where you have essentially business units or micro factories or anything where you bring all the functions together into product service units in one way or another and make a marketplace of them, the trade-off there is that sometimes you lack a little bit of that cohesive identity around what are the thermal standards, what are the thermal protocols, how do we know this person that’s on this team is any good at thermal analysis, et cetera. So you have sometimes a gap in expertise locally or you just have a gap in standards and alignment across the different units. So those trade-offs are always there. And anybody that has listened to me talk for more than five minutes knows that org design is all trade-offs.

(00:07:06):

The way we would play with the dynamic you just described is, we’d start by saying, is there an agreement that we can make about whether it makes us faster and more customer centric to have this as an area of practice or to integrate it within the teams themselves? So usually, it’s like what are the things we want. It might be speed, it might be quality, it might be performance, but whatever those variables are, we might have the debate of, do we get more of that stuff if these skills are embedded inside teams amongst all the other skills? And then the other question is, do we get them actually better when we isolate them? But the problem there is throttling and making sure the line isn’t too long or that the performance is really high.

(00:07:45):

Most of the complaints I hear with centralized functions like a thermal unit or really any quality control is, it’s just too slow. We need to do something and now we’re waiting on them and they have a long line, and so then the whole organization grinds to a halt. So one of the agreements that we’ve used within The Ready and now a little bit within Murmur and certainly with our clients is, if we’re going to have a unit like that, whose customer essentially is an internal user group, and you can call that a help desk, you can call that a center of excellence, you can call it whatever you want, the rule is that, that engagement, that purchase behavior that’s happening between those two units is optional, but the process is mandatory. So we might have a policy that says you have to do thermal review in order to move forward with a part, but you can do it with the internal unit or you can do it with an external unit.

(00:08:38):

So now we have a marketplace dynamic happening where it’s like if they’re not performing, if they’re not getting the job done, and this works as an example even better with a creative group or an internal agency or something like that, where you can use them or you can use the market. It’s your P&L, it’s your dollars. So go figure it out. And in some of these more sophisticated systems that I’ve studied, they even literally move money that way. So it’s like, the HR unit isn’t too big because it’s literally paid by the business units. So if they don’t want that service, if they don’t want that annoyance, they won’t pay for it and that system has to shrink. And if they offer something they really want and they’re really great, then it grows. So it becomes a real service provider to your point. So I don’t know if that answers your question, but those are the things we play with in that order, and then we try to make sure that the agreements are there to support whatever that structure looks like.

Sam Corcos (00:09:26):

Yeah, that makes sense. It’s interesting also in the internal, external. The marketplace component is interesting because it’s sort of like, they talk about this in why AWS has been so successful, which is if they open up their call center technology and nobody uses it, not even Amazon uses it, that’s a pretty good indication that their product is actually not as good as they think it is.

Aaron Dignan (00:09:51):

Totally.

Sam Corcos (00:09:52):

So it’s a painful but really powerful-

Aaron Dignan (00:09:56):

Necessary.

Sam Corcos (00:09:57):

Yeah.

Aaron Dignan (00:09:57):

Necessary. Yeah, totally.

Sam Corcos (00:09:58):

Yeah.

Aaron Dignan (00:09:59):

Because what you see over time is, any organization that’s older than 10 years or bigger than 5,000 people, these functions take on a life of their own and they just get bigger, they never get smaller. It’s like, are they actually getting better every year or they just getting bigger every year? So if we want to avoid that, we have to create the rules of the game early that essentially put market forces on those functions.

Sam Corcos (00:10:19):

This idea ties into performance management. So we’ll table the content, we’ll table the content around platform versus project team. We’ll probably get back to it, but it seems to be a very common thing. I’ve actually been hosting dinners with a group of founders, I think I’ve mentioned this, that are all similarly staged, usually 50 to a hundred employees. It is shocking how identical all of the problems are.

Aaron Dignan (00:10:52):

I know.

Sam Corcos (00:10:53):

It’s just so predictable, the types of problems that each stage of complexity. And as you know, there’s some fudge factor. It’s like, if it’s 50 employees, if it’s 70 employees when we hit this problem, but there are these very distinct phase changes of complexity. One of them that seems to be very common is around this stage, I was talking to Ale Resnik who started Belong, he was saying that 70 was the point where they get these problems where they have to implement performance management. Every company has some threshold where they realize there’s some missing piece and they have to fix it. We are there now. We need to figure out a good way of giving people more direct feedback. I think culturally we already have that. It just hasn’t been institutionalized in a way in terms of a process, which I think is a problem. One of the other things that I’m thinking about now is, we treat compensation data as confidential.

Aaron Dignan (00:12:08):

Mm-hmm.

Sam Corcos (00:12:10):

I personally don’t care, but the fact that Bridgewater doesn’t share compensation data and they share basically everything else to me is an indication that there’s probably a good reason for that, of like, the people who are taking it as far as you can take it, haven’t taken it that far. I don’t actually know. Do you have a sense of why they don’t make compensation data public when they make everything else? And public, I just mean internally available, not public external one.

Aaron Dignan (00:12:37):

Sure, sure. Yeah. I have spent some time there and talked to a few folks. I don’t know that I’ve asked this question directly. My sense is that in their ecosystem there’s quite a big differential around pay based on the different levels of role in execution that can happen in that kind of a marketplace. And there’s a sense that that might be disruptive rather than helpful because when you have a 10, 20, 30, 50X pay gap, that’s just a whole different animal than Phil makes 10K more than me.

Sam Corcos (00:13:07):

Sure.

Aaron Dignan (00:13:08):

So I believe that that may be one of the reasons. I also think that their goal and their orientation is around personal development. It’s not necessarily around equity and balance. So everything they do is about growing people fast and aggressively in a certain kind of way. And it’s not necessarily about other values that might be held. So those are my two assessments. We’ve spent so much time looking at open and closed comp models, and I definitely come down on the open side having seen all the evidence, but to their each own. The main thing you’re taking on when you go to open comp is actually just the period of disruption while all the assumptions and stories get broken. And then once you get on the other side of that, it’s fine. And it’s also fine if you don’t do it, you just have made a trade-off around certain principles.

Sam Corcos (00:13:55):

Yep. That makes sense. So this ties into the second category, which I think we’re currently wrestling with. When we have this conversation about do we make compensation data available to everyone at the company, the conclusion was, I think I was the only person saying yes. And it wasn’t like I actually cared. It was just like, “Yeah, sure, it’s fine.” And everyone else was a really firm no, so it was not the hill I was willing to-

Aaron Dignan (00:14:22):

Really climb yet.

Sam Corcos (00:14:22):

Yeah. And the second category which we’re wrestling with right now is around performance management, which is, we’ve historically had things like one on ones confidential. We share all of our reportings, all of our meetings are reported, everything gets shared.

Aaron Dignan (00:14:46):

Yeah. I’ve watch a lot of them.

Sam Corcos (00:14:47):

You’ve watched a lot, and not a lot of our one on ones though. Some of them are, but most of them aren’t. And people’s performance is not made public. The Bridgewater example is a really interesting one of like, I think they call them baseball cards, where you get to see like, “Here’s Aaron. His strengths are organizational design, project management. His weaknesses are, he’s a poor communicator, he’s bad at this, he’s bad at that.” And you see it. You see it and everyone else sees it. You know who is performing well and who’s underperforming. People’s status, I would say performers, is visible to everyone and it’s not a secret or a surprise.

Aaron Dignan (00:15:30):

Right.

Sam Corcos (00:15:30):

And we haven’t done that. We’ve kept it very confidential. Some of the people on our team use softer language of like, we don’t want to damage people’s reputations. We want to make sure they’d say it face. We don’t want to create problems for them or disrupt their lives. But at the same time, when that person departs and they depart for ambiguous reasons… I’ll give you a counter example. So when I did a podcast with Marc Randolph, who started Netflix, one of the things that he said is, people notice what you tolerate.

Aaron Dignan (00:16:15):

Yeah.

Sam Corcos (00:16:16):

If they see poor performance and it’s not called out as poor performance and you don’t say we are firing this person because of performance reasons, and if you just let them create their own narrative and you just let it go, then you’re not really reinforcing it and people don’t learn anything. You’ve actually, I think, created more anxiety, but it’s also feels kind of mean. Mean is a childish term to use, but it can almost feel like you’re using somebody as an example, like you’re burning them at the stage as a threat. So I don’t know what the right balance is there. We’re struggling to figure out, do we make compensate… Sorry, not compensation. Do we make performance conversations public to everyone? Do we have a chart that says like, “Aaron is a high performer and Bob is a low performer,” and that’s just visible to everyone on the team or not? I don’t know what the right model is here.

Aaron Dignan (00:17:23):

Yeah. I mean in some ways, there’s so much to say about this. You could do an entire podcast just on this question. The one thing that I have noticed throughout my career is that organizations love to do this thing of, these people are top performers or high potential employees, and these people aren’t. It’s binary and it’s very simple. The top performers get to go to Hawaii and the other ones get a performance plan.

(00:17:45):

My first question to you would be, for Levels and for you, what does performance mean? What is performance? Because actually, this feels a lot like the budgeting question for me when people are like, “Should we budget or should we not budget?” I’m like, “What is budgeting? What are you actually trying to do with the word budgeting? And then let’s talk about it.” And in actuality, most people are trying to do three things with budgeting, not one. And then you can talk about the right way to do each of those three things. I think the same story is true here where it’s like we use this catchall phrase of performance management, but we then have to break it down. So how would you define performance? What does performance look like?

Sam Corcos (00:18:20):

I would say, a piece of it is, this is one that we have not done as good a job with, is people should know where they stand. It should not be a surprise when somebody gets a note that says like, “You are not performing and we’re firing you.” So we’ve done a lot of work in the last six months to make that much more clearer to people. I think we still have a long way to go.

Aaron Dignan (00:18:48):

Sure.

Sam Corcos (00:18:48):

I think we’ve come some weight. What’s interesting, and I actually didn’t expect this, the people who were asking for performance reviews were not managers. It was the people being managed.

Aaron Dignan (00:18:59):

Of course. Yeah.

Sam Corcos (00:19:00):

They want to know like, “Am I doing well?” It was like there’s a subtext as well like, “Am I going to be fired or not? Please let me know.”

Aaron Dignan (00:19:07):

And also, am I valuable as a person and how do I feel about myself?

Sam Corcos (00:19:11):

Yeah, totally. We need to do better with that. So there’s that people component. The way that I view performance is if you hire somebody with a set of expectations of what they will deliver, are they delivering against that or not? Are they reliable? Can you trust them to do things? Do they communicate? Do they match the culture? We have had some people who really did not do the things that we thought that they were capable of when we hired them.

Aaron Dignan (00:19:42):

Right, right.

Sam Corcos (00:19:44):

In some organizations, those people just coast. This is going to be a mildly offensive statement, but I have a friend who works apps at an old large company. He joined shortly after college, and he left that for a few years, very frustrated. He described it as baby boomer daycare, where it’s just like there’s all these people, they don’t really do anything. Nobody-

Aaron Dignan (00:20:16):

Just move paper around. Yeah.

Sam Corcos (00:20:18):

They’re just there, and the company is just slowly dying and nobody seems to care. So he left. But if somebody isn’t paying attention, poor performers will become the entire organization because people start to look around. It really is not fair that those people are at the company being compensated in the same way when you have people who are at the same level with the same expectations who are delivering extremely high-quality work. So if you want to maintain a level of talent density, really, it’s not just about finding really good people and retaining them and promoting them, doing all this. It’s also, you have to remove people who are causing problems, who don’t perform the way you want them to. So I think of it in several dimensions. So those are the two right now that I’m thinking about the most.

Aaron Dignan (00:21:15):

That makes sense to me. I have some similar dimensions. I mean for me, first of all, performance, to me, is broad spectrum. So if I have an expectation, I don’t care if that’s a principle level or a value level expectation or a mathematical expectation, it’s still just, are you doing what you said you do? Are you doing what we said we do? Are you playing the game according to the rules that we play? And then are you able to deliver? So part of that to me is commitments and outcomes, which is, what are you committing to? What outcomes do you need to deliver that you’re bought into? And are you able to do that? Part of that is actually more about developing mastery. So what is the vector of your development, your skill development, your rate of improvement? All that stuff matters too, because even if somebody can do what they say they can do, but they don’t grow at all over five years, that’s probably not a great sign and it’s going to sort of stall out the culture in some way, shape, or form.

(00:22:10):

What’s weird about performance management to me is a couple things. Number one, it generally tends to be unidirectional. So it’s, whatever this leader thinks about this individual is performance management. In actuality, it’s like performance is a more of a team sport than an individual sport. There’s very vanishingly few things that we do alone completely that result in actual business performance. Most of it is collaborative. And then once we are doing it, who decides what good looks like? And the classic example of this in my business, in facilitation is, someone who has a certain kind of identity and background and life story will say, “This is how you should facilitate. You should be strong and vocal and loud, and you drive the room.” And then someone will come in with a different background, a different identity, maybe a different gender, identity, color, skin, et cetera. Maybe they have a totally different style. So it’s like, “Oh no, you’re not performing as facilitator because you don’t facilitate the way I facilitate.” But actually, maybe the audience is happy. Right?

(00:23:07):

So a lot of it boils down to who do we ask. So a lot of performance for me is about reputation. We’ve struggled with this too in different periods of our history, but as much as possible, we’ve tried to build systems where reputation drives a lot of outcomes for people, how they’re perceived by their peers, who wants to work with them, which clients want to work with them, at what level, how much demand on their time and existence there is. Those social signals tell us a lot. Then there’s the feedback component which you were talking about, which is, people want to know how they’re doing. They want that feedback and they want it for personal validation. They want it for career development and pathing, and they want to do it so they can not be scared they’re going to get fired.

(00:23:45):

So figuring out a way to make sure that the right people are getting feedback is critical. And again, not just feedback from one perspective, but from the perspectives that matter for performance in that role. So that really matters to me. And I think, at the end of the day, what you’re trying to do is stitch together a system where you don’t have to do a once a year, one person top down performance review, but actually you have a set of agreements and a set of rituals and habits that are creating that clarity all the time. And then if something isn’t working, there’s a way to disrupt it.

(00:24:15):

So as an example at The Ready, if you are not a good teammate or clients don’t like working with you, eventually you’re not going to be on a project. When you’re not in a project, we have an agreement that says if you’re not in a project for a certain amount of time, we need to have a conversation. So we let the marketplace decide, “Uh oh, this isn’t working, let’s talk about it.” But by the same token, maybe you’re killing it in the market and you’re billing a lot, but people are not responding to what it means to work with you, maybe you’re an asshole, whatever it is. We also have an agreement that’s called member review where someone can say, “Hey, I propose that we review the membership of Sam. And why is that? Well, for these reasons.” And then a set of different roles and perspectives that we’ve preordained come together and they review Sam’s membership, and is this working? We go get the data we need and have that conversation.

(00:25:06):

But both of those things are kind of organic and emergent. They’re not happening on a schedule and they’re also not happening directed by me or by any other kind of figure who has power. They’re just part of the rules of the game and how it works. And that’s not perfect, to be honest. They’re real trade-offs in that. But as much as possible, I like to design these things to be automatic and scalable rather than big time sucks and time drains, especially when they’re inaccurate. One person telling me that I’m good or bad at something is not very useful to me, and it’s a huge time drain. But if I have ways to get information regularly from lots of different perspectives and I have an overall reputational score in the system, then that adds up to something a little bit more meaty.

Sam Corcos (00:25:52):

Yeah, it’s interesting you say that because one of the performance challenges that we had, somebody we had to let go from the company, the first indications that I had that something was wrong was, we have signed them to a project and the person leading that project said like, “Could I have somebody else on this project, please?” It was like, that’s a weird thing to say.

Aaron Dignan (00:26:18):

Yeah. Yeah.

Sam Corcos (00:26:19):

It’s an interesting mechanism.

Aaron Dignan (00:26:21):

Yeah. We have all of our teams managing their own membership, so we don’t assign anybody to anything. It’s always a two-way match. And that means that if anybody on a team at any time is voted off the island, they’re voted off the island. So it’s like, “You’re not on the UX team anymore. Uh oh, what do you do here then?” And then there’s a point to have a conversation. So that can be helpful. It’s more social and more lord of the flies, but I prefer it.

Sam Corcos (00:26:47):

Yeah. So Miz and I, who runs operations I know you’ve talked to, we’ve been thinking a lot about organizational design. He maybe sent you a document that you wrote around this recently.

Aaron Dignan (00:27:04):

Yeah.

Sam Corcos (00:27:05):

One of the things that became more clear to me, and tell me if this resonates with you, is that the first model that we wanted to have was much more around this idea of, you can have people accountable to many other people. There’s not a strict hierarchy per se. You can have, a person on the support team can be accountable to the CEO for a project. It doesn’t have to go through a manager in the typical hierarchy tree. It’s much more like a network graph in terms of accountability and responsibility. We kind of already have that, but what we realized in talking to a lot of people, specifically people who worked at Apple and Google and other larger companies, is that that works super well when everyone is performing. But it seems like you need a simple model at the bottom of the pyramid that is just pure hierarchy of who reports to who because somebody needs to decide who gets fired if somebody’s not performing. That’s kind of a blunt way to say it, but somebody needs to make the call and you can’t just have the CEO make every decision.

Aaron Dignan (00:28:31):

I hope.

Sam Corcos (00:28:33):

Yeah. I have a graphic internally at the company of like, you can almost analogize it to the Maslow’s hierarchy of needs, of like, at the base of the pyramid, you need a clear, simple hierarchy of decision making. If that’s all functional, you can have this really great matrix system where people can be accountable to every other department in the organization. One of the people on our team, Braden, is accountable to I think eight other people in leadership for specific projects. And that’s amazing. Only one of whom he reports to. But that’s because Braden is performing really well. And if he wasn’t and he didn’t have somebody who he was clearly reporting to, we wouldn’t really have a way to take action to solve that problem. So I’m curious how you see these sorts of things fitting together and how hierarchy and a traditional leveling system fits into the sort of model.

Aaron Dignan (00:29:43):

Yeah. Well, I think the first thing to think about is hierarchy of what? So we definitely believe in hierarchy at The Ready and at Murmur, but we don’t really believe in a hierarchy of people. We tend to focus on hierarchy of the work and the roles because the people are malleable. I can be CEO today and not CEO tomorrow. I can hold three roles. I can hold 10 roles. What matters is actually the work and how the work is configured so that the work gets done. One of the things that I jokingly wrote about in the book is, there’s a philosopher/teacher who talks about, there’s no word for accountability and finish. Accountability is something that is left when responsibility has been subtracted. So I just generally don’t go in for the accountable too part, or at least not at a one-to-one level.

(00:30:36):

When I think about who’s accountable for what inside our businesses, they’re accountable for lots of things. We’re responsible for lots of things. They have objectives, outcomes, responsibilities that are written out as things they need to do on a regular basis, et cetera. Those things all matter. And I think, I guess, if I had to force it, I’d say they’re accountable to the team or teams where that work lives and is delivered. So in some cases that accountability is to the user. In some cases, that accountability is to a team that they service provide for. In other cases, it might just be to their colleagues on a team. But the accountability or the responsibility is quite fluid and multi-variant in that way.

(00:31:15):

So what I’m really looking for are the right collection of roles with the right purposes and the right responsibilities and the right decision rights, configured into the right teams or groups or circles who also have clear purpose and clear decision rights and clear spaces to operate in, in which they are responsible to for their work and to each other. And if anybody is breaking agreements or commitments, then there are methods for dealing with that. And there are ways to create the accountability or the outcome of somebody being asked to leave.

(00:31:47):

I just generally find that to be more interesting as a social phenomenon rather than a one-to-one phenomenon because I get a lot more data. So if I’m a kiss up, good at managing up person on a team and I manage my boss correctly in a traditional hierarchy, I’m kind of good. CYA is covered. But if I’m actually in a system like ours, I have to be in a net positive relationship with all these different roles and all these different configurations to maintain that reputation. And if I don’t, I’m either going to hear about it in the form of feedback or I’m going to hear about it in the form of a request that I no longer hold a role because I can call for an election on any role in the business at any time. There’s going to be a consequence, a real consequence. But it’s not going to come from one person. It might come from a place that we don’t expect.

(00:32:33):

And if there’s a disagreement about how I’m showing up or what I’m delivering, et cetera, maybe that’s an occasion for the whole team to update the agreement about that role or update the expectation around the outcome or mix up or change the way we’re approaching it. Maybe no one is to blame here, it’s just a system’s problem. But too often, I find that when people just go down the path of like, this person’s a bad performer, this person’s a good performer, they’re missing out on the fact that each of us performs different levels, different ways in different contexts.

(00:33:03):

So in the beginning of the pod, you said, “Aaron, you’re good at this, but you’re disorganized.” Maybe I am. So if I’m on a team where organization is paramount and demanded by the leader, I’m a low performer, but I’m Aaron Dignan. I’m not a low performer, I’m a high performer in the places I could be a high performer. So are you going to fire me from your company because I’m not organized? Probably, that’s a mistake. I should be in a different set of roles where I can really grow for you. So I think those are the dynamics that I like to pull on when I’m resisting the status quo way of doing this.

(00:33:37):

That being said, a lot of people do have pattern recognition and pretty deep grooves around those dynamics and like, “This is my clear line of sight. This is my person that I need to make sure I’m okay with. It tells me I’m good or bad and gives me feedback. It’s all in one nice package.” And that is nice, man. It’s nice to have one tidy little package. Mom and dad are right here, and I can just go to mom and dad whenever I want. It’s very hard to be like, “We got nine brothers here, and we got to figure out who we are and how we’re going to succeed in this world.” And the dynamic is very, very different.

Sam Corcos (00:34:10):

This is an interesting thing to dig into because I think you and I might have a substantive disagreement on some points.

Aaron Dignan (00:34:18):

Oh, good. That’s more fun and it’s better content.

Sam Corcos (00:34:23):

Yeah. One of the things that you said, sometimes the nobody’s fault is a system’s problem. In my mind, that means it’s the CEO’s fault. Fundamentally, it’s always somebody’s faults.

Aaron Dignan (00:34:36):

Yeah. Oh, I agree with that. The problem is that, that’s not who gets blamed in those scenarios most of the time. It’s the person who’s the low performer, and they’re out the door and the CEO goes back to lunch.

Sam Corcos (00:34:47):

Right. Sometimes they are not the ones responsible if they’re operating in a bad system. It’s really then the manager or the CEO who are ultimately responsible for whatever performance issues. I get that.

Aaron Dignan (00:35:01):

Yeah, I would agree. And I would say, for what it’s worth, and you understand this as someone who has several co-founders, that responsibility can be divided and allocated to different folks in different directions. But also, if you’re a member of a team, if you’re a member of a football team and things are not going well, yes, it’s the captain’s fault and the coach’s fault, but also, we all need to show up with a “Let’s fix it” hat on, and ideas could come from anywhere that might fix it. So there has to be a little bit of an all hands on deck, we’re responsible for our own house kind of vibe. But to your point, if the buck has to stop somewhere, it stops with the person that sets the table, and the first person that sets the table is whoever starts the damn thing.

Sam Corcos (00:35:40):

Right.

Aaron Dignan (00:35:40):

So you are ultimately responsible for all the patterns that flow from that moment in those choices.

Sam Corcos (00:35:47):

For sure.

Aaron Dignan (00:35:48):

At least you see you step away.

Sam Corcos (00:35:50):

Yeah, I’m lucky in that this is my fourth company. I made a lot of mistakes in past companies that I-

Aaron Dignan (00:35:58):

That’s enough.

Sam Corcos (00:35:59):

Yeah. I have not repeated most of them. One mistake that I made in the past was having co-founders and not being clear about who makes what decisions. When we brought on each person at Levels as a co-founder, we have five co-founders total, but we made it very clear that at the end of the day, I’m the CEO, which means we have to agree that even if it’s four against one, it doesn’t matter, if I’m the one, that’s what we’re doing. Getting that alignment early was really important and saved us a lot of pain because there have been decisions that we’ve made that I was the only person who was really pushing for it.

Aaron Dignan (00:36:42):

Sure.

Sam Corcos (00:36:42):

A lot of the stuff around transparency, we had several of the co-founders who were very strongly against it and we had a couple who sort of pass it, and I was the only one saying, “Well, I got the sphere, we’re doing it.” And I think that turned out to be a really important decision that could not have happened if it was not made clear and if we didn’t all agree and have this culture of disagree and commit, because we really did see it through and it went really well.

Aaron Dignan (00:37:08):

Yeah. I mean clarity around decision rights is absolutely key. And I would say what that looks like to me at an ultra, ultra varsity level is, it’s not just all decisions, it’s in which decision spaces do I reserve the right to be CEO and in which decision spaces does it benefit us to make decisions a different way? So we have essentially, if you think about all the authority in the organization starting here, what gets distributed and what doesn’t and how do we claw back things that we disagree with is a really interesting part of the journey. And also, I’ve talked about this a little bit, but for anyone who’s listening, the difference between creating a permission culture versus a constraint culture. Because if you’re in a permission culture, then you can’t do anything until you’re told you can. But if you’re in a constraints culture, you can do anything until we say you can’t, which means there’s a lot of authority at the edge.

(00:38:00):

Most of the agreements and decisions we’re making are actually about restricting authority or focusing energy or clarifying or constraining what’s possible either with a policy or a process or a role or whatever the hell it is. That’s a very different motion than folks are used to. So it’s clarity number one, and then number two, actually getting into the subtlety of clarity about which kinds of decisions and what are the best ways to make those decisions. If you’re trying to guess the weight of a cow, Sam’s guess is not the best way to do that. Best way to do that is to average out all the guesses. If you’re trying to figure out how to get out of a burning house, listen to Sam because that’s a totally different context. It’s a chaos context. So we do a lot of work with the Kanban framework to look at which of our decisions are complicated, complex, simple, chaotic, disorder, and then what does that mean for where and how they should be made.

Sam Corcos (00:38:49):

Yeah, interesting. So I want to dig into one of the comments that you just made over, like the CEO ultimately being responsible for the decisions but coming up with some decisions that you give to people who are not the CEO, which obviously that is how organizations scale. Right?

Aaron Dignan (00:39:09):

Totally.

Sam Corcos (00:39:10):

In my own, I wrote a memo on decision making. For me, the definition of leadership is people who have been entrusted with decision making authority of something that’s strategically important.

Aaron Dignan (00:39:22):

Yeah.

Sam Corcos (00:39:24):

Scaling and org building leadership is really just the transfer of decision-making authority. That said, it’s still a clear single line of accountability for every decision ultimately to the CEO. I’ll give you some counter examples where I have seen these things be extremely toxic. So one example, we’ll use the board as the people who can make decisions that are not the CEO.

Aaron Dignan (00:40:01):

Sure.

Sam Corcos (00:40:02):

I know one company in particular where they’ve been performing really well, but the CEO has these two executives who are just really bad. They are incredibly culturally toxic. They gossip a lot. They create all kinds of problems, but they have this really cozy relationship with the board and he needs board approval to fire them. And he can’t. He’s tried and they won’t let him. Having some sort of approval rights, I am a believer that the board should only have one responsibility, which is, do you fire the CEO or not?

Aaron Dignan (00:40:44):

Right.

Sam Corcos (00:40:45):

If you believe that they are making the right choices, then leave it alone. Otherwise, you should fire the CEO if you don’t trust their decision making. I think that’s totally reasonable. All those other stuff of boards taking control from the CEO and micromanaging their decisions, I cannot think of an example where this has been a good thing. This is one of the things where it’s like, hypothetically, I’m trying to understand the model that you’re presenting, of like, we agree that our head of growth should make decisions related to X, which we obviously do that right now. But ultimately, he’s accountable to me for that decision and I can tell him whether or not we’re going to do the thing.

Aaron Dignan (00:41:33):

Mm-hmm.

Sam Corcos (00:41:33):

It’s like, I think he’s going to say, with his group, they’re going to come up with a proposal. Ideally, my role is just to say, “Thumbs up, let’s do that.” But he doesn’t have independent authority. In my understanding of decision-making frameworks, all decisions have a single line of accountability to the CEO. Is that a misunderstanding or is that a different framework for how you think about it?

Aaron Dignan (00:42:00):

Well, it’s a different and very acceptable framework. I mean, there’s nothing wrong with that, but it’s definitely different than the way we tend to look at it and advise around it. In some context that can be really helpful, so I’m not even saying it’s a bad or a good thing. But what I would say is this: First of all, the board situation is really easy to clear up. We want to put decision rights where the contact is greatest. So having people that are one day a month or one day a quarter looking at the business, making a decision about who I work with as a CEO, F that. That is not even close to our philosophy.

Sam Corcos (00:42:36):

Yeah.

Aaron Dignan (00:42:36):

I agree, they essentially exist to do two things in my mind. One is to essentially bring different perspectives to bear societal, economic, social perspectives around our purpose and how we’re manifesting that purpose in the world, and picking somebody to set up the table. So we’re right on the same page about that. Where I think we have a slight difference in approach or style is that I look at the decisions that each successive team, whether it be the leadership team, the CEO, the board, whatever, every successive team all the way down, the decisions they’re actually making are decisions directly related to what they’re doing. And then decisions about which decision spaces and purpose spaces they’re carving out for other teams to operate within. So it’s more like a two-step accountability than a one step. So you’re talking about they make a decision, they bring it to you, you don’t like it, you change it. That’s direct connection.

(00:43:30):

The way I think of it is, I’ve given the studio circle, along with my other leadership team members at what we call our source circle, we’ve given the studio circle that produces all our content a purpose around building the brand through content and some authority and decision rights around what we produce and where we produce it and where we send it and all that shit and even a budget around how much to spend. Once we do that in our model, I do not have the right to interrupt or disrupt that process except by changing that charter or who fills that role of stewarding that circle. So I can be like, “I don’t like the last three decisions that y’all have made, I’m going to make a proposal that we reelect a new steward for that team.” Or I can say, “I don’t like the last three decisions y’all have made, so I’m going to propose that we change the charter for that team.” But I can’t just go in and interrupt a decision in flight.

(00:44:19):

It’s the design, it’s the blueprint of the organization that we actually have direct authority and accountability over. The executional reality is one step removed. And their trade-off is there. And the trade-off we like about that is the level of ownership, intensity, stewardship that we see from people when they can’t be interrupted. And when they’re fully accountable for the result and the outcome, there’s no like, “Well, I checked with whoever.” We like that. The trade-off is that sometimes stuff happens that we don’t like.

Sam Corcos (00:44:47):

Okay. Yeah, it sounds like we’re not as far off as maybe I thought with the difference in language that we were using.

Aaron Dignan (00:44:54):

Yeah. Okay.

Sam Corcos (00:44:56):

So tying into some of the specific language that you just used which is, I’ve given them the authority, which means that you can then take away the authority if something goes wrong, but the whole point of giving the authority is because you trust them to make the decisions. So it’s not that you are giving them the authority because you have that trust and they will come to you with some suggestions of what they want to do. I can tell you, nine out of 10 times, because we have the right people in the right roles, people will come to me with proposal, it’s like, “Hey, all right, we just finished our content strategy. Here’s what we’re thinking.” And I might have a couple of things to add that maybe they don’t know because of context, different parts of the company. But usually I just go like, “That looks good.” And then that’s it. And then they execute completely independently.

Aaron Dignan (00:45:51):

The only subtleties I would add is, first of all, nobody comes to me with anything because they’re doing it, they’re executing. So I’ll see it, and basically I’ll find out about it, but we do have what we call an advice process. So if somebody’s making a decision that they feel a little over their skis about or a little under skilled, or they just want context and support, they can choose the right roles within the system to get advice from. And that might even include me or Rodney or one of the other people that have been here a long time, but that advice does not come with strengths.

Sam Corcos (00:46:20):

No.

Aaron Dignan (00:46:20):

So if somebody comes to me and is like, “Should I spend a hundred grand on this quarter from the Revolutionary War?” and I’m like, “No,” and they do it, they’re not fired for not doing what I said. It’s just like, “I told ya. Now, have you learned something and can we put this into practice in the next time?” So that’s the first clarification. And then the second one is, when I say I gave them that authority, to be very, very specific because we work in a constraints-based system, what we actually did is we, the leadership team, the source circle, which is mostly made of elected members by the way, it’s a very democratic team, but it’s a team with power, concentrated authority about making our content into that group of people before anyone could make content, any content they want at any price. We actually concentrated that authority and said, “No one else can make these sorts of decisions now because we’re focusing it into this team.” And that was a consent-based decision with up to 10 or 11 other people, so it was not just me.

(00:47:19):

The places that I make decisions that would be considered more of a small just me or almost just me would be at the board level where Rodney and I hold the only two board seats. And there, we do make fundamental decisions about the structure of the company and how we build it and what the rules of the game are with just the two of us, still with consent, but it really acts a lot more like a board. So yeah, it’s turtles all the way down, but that’s the way we do it.

Sam Corcos (00:47:46):

Interesting. Yeah.

Aaron Dignan (00:47:47):

Yeah.

Sam Corcos (00:47:48):

Definitely, I still feel like there’s misalignment, but there are so many things that are similar. I’m trying to parse apart some of these things.

Aaron Dignan (00:47:56):

I think, well, it’s just about what you want to prioritize and also what… I think the context matters. I think what kind of work you’re doing, at what timetable, with what expectations, with what level of complexity and lack of clarity, all those things count for a lot. But ultimately, it does boil down to principles in action.

Sam Corcos (00:48:17):

Yeah. I think about, a lot of these things come down to simple decision-making models like a RACI matrix. Right?

Aaron Dignan (00:48:25):

Mm-hmm.

Sam Corcos (00:48:25):

I think we put together a lot of thoughts around how we thought decisions should be made, and it ended up basically reinventing the wheel of a RACI matrix. Basically what we have is a concept of a DRI, which is a pretty common thing around a directly responsible individual. They are the ones who are ultimately responsible for the outcome of a project. No matter what happens, they are the ones on the hook.

Aaron Dignan (00:48:55):

Okay.

Sam Corcos (00:48:56):

Cross functionally. The accountable person is who you’re accountable to. Eventually, the bottom of that stack is the CEO. So you might be accountable to the CEO or you might be accountable to the person running growth, who is then accountable to the CEO for whatever it is. The goal is to push those responsibilities to the edge of the system as much as possible, as opposed to concentrating decision-making authority in the CEO or in a bare narrow set of people. You want to push that out as far as possible. Consulted is one that, this is where I think a lot of companies get tripped up, which is, in many organizations, people who are consulted have veto power over the project. And that’s where a lot of these organizations get tripped up where-

Aaron Dignan (00:49:51):

Kidding.

Sam Corcos (00:49:51):

Yeah. You end up with this one over n squared problem of decision making where it’s like, you end up with 25 consultants and the odds that that project is going to get completed is effectively zero.

Aaron Dignan (00:50:02):

Yep.

Sam Corcos (00:50:03):

So we make it very clear, very much in the same way that you’re describing, of like, if you want to spend a hundred thousand dollars on a marketing project, don’t ask me permission, but you can ask people if they think it’s a good idea. Some will say yes, some will say no. But where I feel like there’s may be a gap in the way that we’re describing these things is, you have this group that assigns a person to solve a problem, but the accountability model that I’m describing is, there’s a step before that, which is you have to align on what the problem is and that person has to agree to it. So like, they will say, “Here’s what I think growth should do for the next 12 months.” And then I say, “Great, you should do that, and you’re accountable to me for that thing.” Whether it’s to me or whether it’s to a group of people, it doesn’t matter, but it’s like, we have now agreed that you’re going to do the thing, now you can go operate independently.

Aaron Dignan (00:51:00):

Mm-hmm.

Sam Corcos (00:51:00):

Does that make sense?

Aaron Dignan (00:51:03):

Yeah.

Sam Corcos (00:51:03):

The accountability model seems very similar, but we seem to be using a very different language to describe it.

Aaron Dignan (00:51:10):

Yeah, and also I think different mechanisms to accomplish the same goals.

Sam Corcos (00:51:15):

Mm-hmm.

Aaron Dignan (00:51:15):

First of all, I think RACI model is generally kind of dumb. I think responsible and accountable, if you look them up in the dictionary, literally use each other’s word in their definition.

Sam Corcos (00:51:24):

Sure.

Aaron Dignan (00:51:25):

So it’s like, what the fuck does that mean? But at the end of the day, what we’re talking about is who’s on the hook to actually do it and make it happen. For us, that’s always one role or one team. So that’s clear. If we’re wondering who is responsible if it doesn’t go well, who needs to fix it, it’s pretty easy to just say, “Well, is the role in a team or a circle? Great, then it’s their problem. Is that nested in a super circle? Great, then it’s their problem.” It’s very clear always where the work lives in the hierarchy of work. So if something is going haywire in our growth team, we’re going to hear about it in the source team and then we’re going to have to be ultimately responsible for fixing it. And if that doesn’t happen, then the board’s going to hear about it and be responsible. So that’s very similar, but we don’t feel the need to write that down for every freaking project we do or every role we create. It’s obvious because the structure is transparent and intuitive.

(00:52:14):

So I think that’s the first thing. And then the second thing is, what does the definitional work around strategy or what we’re trying to accomplish is actually an agreement that we have about how we do strategy and how we do prioritization. So every 120 days, every circle in the company is doing strategic work in concert with the others and seeing what they’re doing and providing advice, et cetera, so there is a lot of calibration happening about what the goal is, but ultimately, that charter of that team that describes its purpose, that’s the main instrument for making sure we’re getting the right things done. And then when we look at the strategies, we can offer advice along the lines of like, “If that’s your strategy and your plan, is that actually the best way to pursue that purpose?”

(00:52:59):

So if the purpose of growth is to create more leads, then we know what to do with, which is more or less is, more qualified, good, aligned, sustainable leads, then we know what to do with. Then everything that comes up strategically has to be put through that filter of like, is that happening? And if it’s not happening, why? And does this strategy seem like it’s well aligned to drive that? But again, we’re not reviewing that from the perspective of, we have to bless it. We’re reviewing it from the perspective of, they asked for advice on it from all the other people that they interact with and work with as a team, and they’re going to get it. And then if they don’t take that advice, two trimesters in a row and growth is not happening, then there’s going to be a consequence. But in the meantime, it’s the model of, let’s give everybody enough rope sort of the model.

Sam Corcos (00:53:42):

Yeah. Yeah.

Aaron Dignan (00:53:43):

So those are subtle differences, but I think they’re just very slight different mechanisms.

Sam Corcos (00:53:48):

Yeah. I think one of the things that I recognize in what you were just saying though is, I think we’re actually talking about different steps in the process, which is, when the growth team has an idea for how they want to achieve certain growth goals, they can ask me if I think it’s a good idea, but I’m in a consultative role, I’m not the veto in that conversation.

Aaron Dignan (00:54:11):

Yeah.

Sam Corcos (00:54:13):

It’s the step before of defining that accountability and that agreement model, of like, before we have said that you are responsible for doing a thing, having that agreement in place is something. That’s what I mean when I say the single line of accountability to the CEO is having that agreement first.

Aaron Dignan (00:54:36):

Yeah. Well, and for us, really, I think that makes sense, and I think it’s literally just a chain of agreements that are about the purpose and intent and scope of these different teams and roles so that it’s very clear anyone can click into Murmur and two clicks later be like, “Oh yeah, the growth team is supposed to be accomplishing this with these responsibilities, with these decision rights.” And is that happening or not? And then related to that, here are some expectations about how we do strategy and when we do strategy. When it gets down to the level of a particular initiative or project, if it fits in the scope of that circle and their budget, I’ll never hear about it and I don’t want to hear about it, but if it’s extracurricular, we have an initiatives agreement that’s separate as well.

(00:55:17):

So we make these non-linear bets every 120 days as a team on stuff that doesn’t fit in the structure. So it’s like, “Here’s the thing growth wants to do that’s a wild swing,” that gets proposed as a bet at our offsite and then people actually do weigh in on that. And there is ultimately a consent moment for a part of the membership to buy into that. But again, even there, it’s often multi-perspective and there isn’t a final say person. In some ways, what’s interesting about our model, maybe this is the interesting bit, because we run the whole company on consent, with the exception of decision rights that a role has. I hold a finance role, for example, and if I set the budget for those initiatives total, that’s not a debate. I set the budget. If I say it’s 300 grand, it’s 300 grand. So that is a role decision. We have lots of those around the company, lots of places where roles have total power to make shit happen. But that power was given to them through consent, through the group agreeing that it’s safe to try.

(00:56:15):

Most of the power that happens is actually the power to object, not the power to consent. So things are going to happen unless it’s proposed and somebody’s like, “I don’t think this is safe to try.” And that’s a power that typically within a team is shared. Every role has that capacity to be like, “Wait a second.” Because frankly, we play from position. So if someone in a finance role and someone in HR role and someone in a growth role and someone in the sales role are seeing this thing, this idea, I actually want them each to look at it through that lens. And if sales is like, “This is a terrible idea,” I want to hear that. I want to actually play with that a little bit before I try to say, “Well, I love it, so it’s happening.”

(00:56:57):

It’s funny because we had something in our last retreat, we have an initiative going around a campaign. We’re doing a marketing campaign. The idea that was selected by the group that was entrusted with that money in that initiative that came forward was really divisive. People that heard about it were like, “Whoa, whoa, whoa. We do not consent to that. We’re not okay with it.” And it was an interesting test of our model because ultimately, it boiled down to, did we consent to give this team the power to make this decision and the money to do it and the spate and the scope to handle it? Are they colleagues with good reputation? They got to do it. They’re going to do it, and it doesn’t really matter whether we like it or not. So that was kind of an interesting moment to see something happen that was not popular. Because I think sometimes when people hear me talk, they’re like, “Oh, everything’s kumbaya consensus popularity contest,” and it’s really not. Lots of unpopular stuff happens, but it happens in places where the power to do that was consented too.

Sam Corcos (00:57:53):

Yeah.

Aaron Dignan (00:57:54):

So it’s like we gave them the power to make an unpopular choice and they did. Now they’re doing it, and we’ll see how it does and we’ll learn from it as a system, which I’m excited about whether it’s good or bad. Personally, I’m very motivated by organizational tension that leads to learning. If we can get a little disrupted along the way and actually learn some deep lessons, I love that because that’s so resiliency building, so capacity building. It’s more dramatic and people sometimes don’t like me for that reason but-

Sam Corcos (00:58:27):

Yeah.

Aaron Dignan (00:58:28):

It’s like raising any system or a garden or a child or anything that you want just enough failure and just enough disruption to create strength. These are anti-fragile systems. Organizations are anti-fragile systems, so they can be.

Sam Corcos (00:58:42):

Yeah. Well, so circling back to one of the earlier conversations, talking about performance management, but specifically around the topic of communication around performance management.

Aaron Dignan (00:58:56):

Yeah. Yeah.

Sam Corcos (00:58:57):

If somebody is a poor performer, what is the next step? You have now made the decision we have to let this person go, or whatever reason, maybe there’s like a culture issue, they just aren’t delivering like they used to, or the needs have expanded and they’re no longer filling that role, do you let them set their own narrative for why they’re leaving and just be quiet about it? Do you announce to the team that somebody’s being removed for performance reasons to make an example of them? What is the correct approach here? And I realize that every culture handles this differently.

Aaron Dignan (00:59:38):

Totally.

Sam Corcos (00:59:39):

I’m curious what some of the models are that are not pathological.

Aaron Dignan (00:59:44):

That’s a great way to phrase it. I mean, one thing I’ll say as a caveat is, we’re running a ground here of employment law. So I think there are things that I think are good practice that are not necessarily legal practice. So what is allowed versus what is ultimately best for human beings, I think, are not necessarily the same thing. That’s the first thing I’ll say. Sometimes it’s true that you can’t really broadcast exactly what happened and that’s the way it is. So we live with that. But what I would say generally is two things. Number one, if we’re doing a good job in our feedback and reputation economy, long before someone is asked to leave, everyone is aware of what’s going on just because they’re there. They’re present for the kinds of feedback and information flows that are going to happen when something isn’t getting delivered because it’s constant.

(01:00:32):

We have a weekly action meeting where we’re asking, “What do you need and what needs to be unblocked? What is the status of all of our most important initiatives in every team in the business?” Whether that’s done asynchronously or synchronously, the writings on the wall, if shit is not happening and someone’s not delivering, people on that team know for sure. The whole system may not.

Sam Corcos (01:00:52):

Yeah.

Aaron Dignan (01:00:52):

So to your point, it’s local, not global. Right? The second thing is, once we’ve kind of come to that conclusion or things are starting to get clearer, there’s probably a conversation that’s not a formal conversation happening with that person, maybe with an elected leader or steward of their team, maybe with a formal leader in another model, whatever the case may be, where the conversation is like, “Hey, this probably isn’t working out. Are we going to go one more round at this? Or do you want to set an agenda for yourself?” We’ve had plenty of people opt out of our system on their own terms. When they’re like, “I’m an adult, I can read the room. This is not going well. I’d like to work for the next six weeks to get out of here and do a good job doing that. Here’s what I need to be successful in that,” then we’re like, “Great, do it.”

(01:01:39):

I mean candidly, I’ve told people like, “I think you’re done here. I’m not firing you, but I don’t see a way to get out of this spin. So what do you want to do? How can I help you with that acknowledgement?” So I think that’s the first thing. And then if you’re at a point where it is a conversation like, “You’re done here,” then I think often the conversation we’ll have with them, if there’s space and nothing has happened that’s too egregious, there’s a conversation about like, “How do we want to share this? What’s useful to the team with a learning agenda?”

(01:02:09):

So the idea is like, “Hey, this is done. We’re going to take really good care of you. You’re going to move on. You’re going to have a better fit somewhere else anyway. And here’s nine pieces of evidence that has always happened.” And the fact that this didn’t work out as a learning opportunity for us, so what can we share with the team that will help everybody to learn about what happened here about you and us, and treating that really honestly. And then there might be a parting note from them or a parting note from us about them, or both, that would be like, “Hey, here’s what happened. Here’s what we learned.” Sanitized for HR preferences, but still getting into the meat of like, “Hey, here’s what we did. We bit off more than we could chew. We put this person in three roles worth of work instead of one, and it totally worked their year,” or, “We’re a place that really seems to prize reliability, and now we know that, and this was the situation where this person was wildly creative but not able to be reliable.”

(01:03:04):

I think just even a few nuggets of taking accountability on both sides, and then ultimately what’s the learning for the organization, that’s where I like to get off, if possible. And I’d say, “Do we pull that off every time?” Hell no. But when we do, I’m like, “That was clean, that was perfect.” And then you find out three months later, that person is a director at wherever and they’re doing great.

Sam Corcos (01:03:25):

Sure.

Aaron Dignan (01:03:26):

You know?

Sam Corcos (01:03:27):

Yeah. So the problem is the global problem, not the local problem.

Aaron Dignan (01:03:34):

Yep, yep.

Sam Corcos (01:03:35):

I’ve only been made aware of this recently within the last few weeks, that people who were on the teams knew what was going on.

Aaron Dignan (01:03:46):

Mm-hmm. Put everyone else in the dark.

Sam Corcos (01:03:50):

It’s like, it’s not just that they’re in the dark. In the service of being kind to people and allowing them to save face, protect their reputation, these are a lot of the words that are being used internally, in order to do that, we have created a tremendous amount of anxiety for the majority of the team who feel like they could be next.

Aaron Dignan (01:04:15):

At any moment.

Sam Corcos (01:04:16):

At any moment. We have no idea why these things happened. Why is all this stuff happening?

Aaron Dignan (01:04:22):

Do you have a written agreement about how people are asked to leave?

Sam Corcos (01:04:26):

No. And I think we should. Actually, I sent a note about that very recently to the team.

Aaron Dignan (01:04:30):

It’s a huge boon because then, that uncertainty goes away. If you basically stand up as the CEO and you’re like, “This is the agreement that the leadership team came up with for how people are asked to step away. We follow this every time to the letter. It has caveats for harassment and instant firing and all those things that need to be in there. But if you’re not doing that, this is how it works,” then everybody can breathe and be like, “Well, nobody’s put me through step one of that.” And now that’s one piece of the puzzle for sure. I think that that adds a lot of value. And then the other piece is figuring out how do we communicate what we learned without making it about the person.

(01:05:06):

I mean, you’ve heard me barking about roles for a long time now, but it’s another place where I think roles are really helpful. Where it’s like, in this role, this role was unable to do this, or this person was unable to inhabit this role successfully because of X, Y, and Z, as opposed to saying, “Phil’s a low performer,” because Phil’s not a low performer. In the right situation, in the right context with the right instrument in his hand, he’s a high performer, but in this role, it wasn’t working and here’s why. And then people can be like, “Oh, shit. I see that that’s what’s expected in that kind of role and why there was maybe some issues there.” If I want to know more, I can go find out from the team, but just something about that gives it a little distance where it’s no longer about the person, it’s about the context. And that’s what really matters anyway. It’s like, “This was not working out and now we need to find someone who can inhabit this role and kick ass.”

Sam Corcos (01:05:56):

Yeah. It’s also role over time because the role changes.

Aaron Dignan (01:06:04):

Absolutely. Yes.

Sam Corcos (01:06:06):

Actually, we’ve had this arced recently with them. My co-founder David, who started out doing really well on product, didn’t have the experience to lead the entire product org. We’ve moved him into a different role and he’s now doing extremely, well and we’re adding more and more responsibility onto him. So just finding where people’s sweet spot is at any given time. Sometimes the function can grow faster than you realize, and then you end up-

Aaron Dignan (01:06:41):

Ideally. I mean, that’s the ideal scenario. Right?

Sam Corcos (01:06:43):

Right.

Aaron Dignan (01:06:43):

That’s what you want to happen, is you want the organization to outgrow the people because you’re so successful in each of the stages. And that’s one of the reasons also why we pride ourselves so much on a role mix rather than a role singular. Because it may be that it’s like, “You know what? We’re pulling you out of this role because you’re not performing there. You’re really performing in this role and we’re going to build another one like it that we can put you in, or we want you to go part-time, or any number of other possibilities.” But suddenly, we have more tools in the toolbox to play with to get maximum performance. And that’s why when people ask like, “How do you achieve the levels of profitability that you do? Or how do you achieve what you do with the number of hours you put in?” I’m like, “We just get more out of ourselves because we don’t treat ourselves single action figures in the box.”

(01:07:29):

Every skill that we have is deployed against a mix of roles over time. As we get better, the mix changes and the rules change, to your point. It’s absolutely the case that roles are going to outgrow you or morph out of you. That’s good. As long as you’re a player on the team who knows how to find and create new roles for yourself that add value, and in our system, you can literally propose them. I could be looking at a role I’m failing in and be like, “Man, I’m going to get fired.” I’m going to propose a role that I can beat the out of. And then suddenly, I’m in a role where everyone’s like, “Wow, you’re delivering and we’re loving it.” That’s the ideal scenario. So obviously, there are some people that are not in a place in their life or with their skills where they can hang and they do have to go, or that they are disingenuous about commitments, they have to go. But a lot more people are just accidentally in UX when they should be in UI, or they’re doing testing when they should be doing engineering or something.

Sam Corcos (01:08:23):

Cool. Well, this has been very helpful food for thought. I’m going to be doing a lot of writing over the next few weeks to try to sort this out for ourselves.

Aaron Dignan (01:08:34):

Cool.

Sam Corcos (01:08:35):

But yeah, I really appreciate you making the time. This is a helpful perspective.

Aaron Dignan (01:08:40):

Likewise. Yeah. It’s useful for me too because as we build Murmur, I’m having to learn how to communicate and build around these different perspectives and compromises and priorities and trade-offs. It’s like, doesn’t do anybody any good to have a tool that is so principled that nobody can use it. At the same time, I want to offer some of the juice for the squeeze here to the broadest number of people. So every conversation, this is really useful to me.

Sam Corcos (01:09:07):

The language that you use around an operating system is very interesting. I think sometimes just knowing what agreements you should have is really helpful.

Aaron Dignan (01:09:24):

Yeah.

Sam Corcos (01:09:24):

My friend Zach Goldberg is writing a book on running technology and engineering teams as a CTL. He wants to be like a… What’s that guy’s name who wrote The Great CEO Within?

Aaron Dignan (01:09:39):

Yes.

Sam Corcos (01:09:40):

Matt Mochary, I think, is his name. Something like that. He wants to make the same book but for CTOs.

Aaron Dignan (01:09:46):

Okay. Yeah.

Sam Corcos (01:09:48):

Yeah. One of the things that I really got out of Zach’s book was saying that, making it clear what the process is for firing somebody actually reduces anxiety.

Aaron Dignan (01:10:00):

Yeah.

Sam Corcos (01:10:00):

Because people know, it’s like, “Well, if I don’t have a pip in front of me, it means I’m not going to suddenly be fired without warning.” So just having that at least lets people know that they don’t have to be on edge all the time. And I hadn’t really thought of that before, but you mentioned something very similar and that makes me think that we really need to do something similar.

Aaron Dignan (01:10:21):

Yeah, I can’t say that enough, which is clarity, clarity, clarity, clarity, clarity. What is the shortest, simplest way to have clarity about anything that is debatable, emotional, scary, frustrating, any of that, that is the name of the game. The shorter and sweeter they are, the better. But then, people are just like, “I know. I don’t have to guess. I don’t have to read Sam’s face in the meeting.” It’s just like, this is how we play the game. So yeah, I think at the end of the day, most of what I do is just teach people monopoly. Got to have a rule book.