The Live Oak Playbook
Achieving enduring growth by embracing a scarcity mindset
If you hike in Northern California there are two types of trees that you will see most often: eucalyptus and live oak. They have very different growth strategies.
Eucalyptus trees grow as tall as they can, quickly, until they become towers teetering over their surroundings. They are a high-growth, invasive species and in places that they are transplanted — like California — they quickly crowd out other species.
Contrast that with the live oak, a tree that grows deliberately and slowly, broad and strong.
The eucalyptus strategy seems unbeatable: take advantage of good conditions to grow as quickly as possible and dominate the landscape. But that high-growth comes at a cost: eucalyptus trees are brittle. If the wind is too strong, they topple. If a forest fire gets anywhere near them, they burst into flame.
The live oak tree, in contrast, is a tank. Its breadth and deep roots mean it can handle even severe wind. Its hard wood and thick bark mean it can easily survive forest fires.
There is no such thing as a dominant strategy; different strategies succeed in different contexts. But sometimes a context that has been stable swings wildly into a totally different equilibrium and all of a sudden previously powerful strategies become dangerous liabilities.
A strange new era
For years in Silicon Valley we honed the one true approach to technical innovation, what might be called the high-growth playbook. Tech was the king of the world.
But now conditions have changed. Things that used to feel like the most natural thing in the world feel wrong — a bit off. Weird things are happening. What’s going on?
Imagine being locked in a windowless room for a decade. All of your experiences in that room assume that the force of gravity pulls in a specific direction. But now imagine the room tilts on a hinge so gravity pulls at an angle. Everything still looks the same, but things feel a bit different. Things that used to stay fixed firmly in place roll around aimlessly, and some things even fly off the table and smash against the wall. All of your intuition in this environment was formed under conditions that no longer exist, but superficially still appear to. Trusting your gut only works when your gut is calibrated. When the conditions change fundamentally but subtly, trusting your gut is actively dangerous.
Friends, the reason we felt like geniuses is not because we had figured out something fundamental that everyone else missed. The reason we felt like geniuses is because we were playing on easy mode.
A lot of the tactics that we learned — head down, go go go, no time to think, scale like your life depends on it — were, to some degree, a Zero-Interest-Rate Policy (ZIRP) phenomena. In such an environment capital desperately seeks a place to go, meaning anyone with a vision and a business model that looks superficially tech-like can be inundated with opportunity. Tech businesses are defined by marginal costs that approach zero and winner-take-all dynamics. In an environment flush with capital, it makes sense to invest in anything that looks even superficially tech-like just in case it turns out to be one of the massive successes.
But we are no longer in that environment. In this hangover it suddenly dawns on us that maybe we need a new approach. In this harsher environment a more resilient technique is necessary. This resilient technique was always a good idea — it’s just that in a harsh environment like today that it becomes an inescapably necessary idea.
Much like training at altitude makes you a beast at sea level too, mastering this new technique will serve you well in any environment. Previously, our environment was cushy enough that we didn’t have to worry about a resilient playbook. Now understanding it and mastering it is a matter of life and death.
Silicon Valley has been like the eucalyptus tree — a technique that works only in the most favorable of environments. As our environment gets significantly more harsh, it is time for our industry to adopt an approach more inspired by the mighty live oak.
The good news is that although the approach will feel a bit foreign, and even in places a bit backwards, it’s not complicated. There are just a handful of precepts to follow:
- Survive, then thrive
- Keep your metabolic rate low
- Think as long-term as your runway allows
- Don’t be a hero
- Take a step back to go strong
- Stay within your iterative adjacent possible
- Be rigorous by letting go of excess precision
- Be curious about disconfirming evidence
- Take the sting out of failure
- Clear the bar of good enough
- Continuously improve what’s working
- Focus on things with compounding returns
- Don’t hold out for any one miracle, but have exposure to many
- Create space for innovation
- Embrace combinatorial innovation with a scavenger mindset
- Invest in trust
- Create high-trust teams with a diversity of perspectives
- Reduce the need for coordination
Let’s dig into each one.
Survive, then thrive
The core goal, at the root of everything, is to first survive, and only then turn your attention to thriving.
First, cap downside risk. If you don’t survive in the next round, then it’s game over — there are no more rounds to play. No matter what you do, you must live to play another round. But survival is a bar to clear, not something to maximize.
Once you survive, you want to create as much value as you can. This is the term to maximize once you’ve cleared the survival bar.
If you don’t survive, then the value you can create is severely capped. If you don’t thrive, then there’s no point in living. You need both. Survive, then thrive.
Being a viable business and a valuable business are different. Survival is necessary, but not sufficient.
Note that the position of funders and founders is necessarily different; a VC can survive the death of any company in their portfolio and needs only one of them to be a blockbuster to make it all worthwhile, meaning they will want each company to focus most on thriving. A founder, on the other hand, is lashed to the mast of the company they are currently running — they have only one bet going on at once and need it to survive. These incentives will pull investors and founders in different directions.
Keep your metabolic rate low
Your energy can go into surviving or thriving. Your basal metabolic rate is the amount of energy you need to invest just to survive. If your basal metabolic rate is high, then you need to spend a lot of energy just to keep the lights on, and then there’s very little energy left for thriving. If the conditions become harder you might find that there’s not enough energy to sustain your metabolic rate anymore, and you die.
Once you increase your metabolic rate, it’s hard to reduce it significantly — every decision you made so far implicitly locked in and assumed a given metabolic rate. This means your metabolic rate tends to ratchet up. Note that increasing your burn rate also doesn’t necessarily lead to significantly better results. If you increase headcount, for example, you’ll get sub-linear returns because coordination costs will tend to grow super-linearly.
A lower metabolic rate significantly extends your runway and increases your horizon. It allows you to spend more of your energy on thriving and less on surviving.
Think as long-term as your runway allows
When you take an action, there are direct effects, immediate and obvious. But there are also indirect effects, rippling out in space and time. Over long enough time horizons, the indirect effects will dominate the direct effects.
Direct effects are hard to miss: shiny, loud, obvious. Indirect effects are massive but diffuse, making them hard to spot without careful study. In the heat of the moment they’re easy to miss when direct effects are blinding you.
If you think only over a short time horizon you might erroneously conclude that an action is beneficial when actually over long-enough time horizons it is significantly destructive.
As an example, the compounding effects of many SaaS-shaped businesses come from an accumulating shingling of successive cohorts that individually grow over time. That means that the vast majority of in-year revenue is driven by actions and happenstance from years in the past. It’s tempting to take optimizing actions for in-year revenue with the direct levers you have at hand, but be wary of trading off small short-term gains for massive long-term losses.
Things get increasingly uncertain the further into the future they get which means the level of certainty you can have about indirect effects gets ever cloudier the farther out you think. Still, the massive weight of their impact means that getting an order-of-magnitude level understanding of their effects is worth it. And a number of forces of gravity — subtle but omnipresent — make it easy to guess the direction and order of magnitude of impact.
You only want to optimize on a horizon the length of your actual runway because if you’re about to die then nothing else matters. Once you increase revenues to get to the magical point of zero net burn, your runway starts to approach infinite, allowing you to take a significantly longer time horizon under consideration.
Don’t be a hero
In a growth context, it’s aways go, go, go with no time to think. But now there are game-over conditions everywhere.
When you’re in easy mode being a hero is easy. Even when you fail the tailwind is so strong you still tumble forward and win. But when you’re not in easy mode, being a hero is often deadly. Survive, then thrive.
Heroics often take the form of firefighting. Firefighting focuses on the direct value — putting out the fire in front of you — but often with extreme indirect cost, sometimes even creating the conditions for future fires. Fires are an order of magnitude cheaper to prevent than to fight.
If you’re a manger, heroics sometimes takes the form of doing everything yourself. This prevents people on your team from growing, and means that the team can’t scale its output very far, because everyone will be more timid about doing things themselves, instead waiting for you to weigh in or even do it yourself. Overtime, this lack of learning and growth on the team makes it even worse. Delegating is not optional.
See fire fighting — especially when it has to be done in the normal course of business — as not a thing to be lauded but something to treat as a failure, a bug to diagnose.
Take a step back to go strong
In the fog of war there’s always an overwhelming number of short-term, fast-twitch actions you could do. That means that if you aren’t careful you’ll keep doing fast-twitch actions that over time build you to a sub-linear result or execute yourself into a wall. Remember that the indirect effects matter the most.
Going fast feels strong, but it’s actually weak. The true way to go strong is to be patient, to take a step back to think just a bit more before you move. If all you’re doing is greedily chasing short-term wins then you’ll lose sight of the compounding potential of clever ideas and get stuck on an inadvertently-sub-linear path. If you don’t take a step back every so often you’ll be accumulating a series of kneejerk reactions to problems. Running around in circles is an enormous waste of valuable time and effort.
At least 20% of your plans should be reserved as slack. This slack allows you to absorb surprises without getting swamped. It also allows you space to follow hunches and plant a few cheap dandelion seeds. But most importantly, it gives you the space to take a step back and synthesize.
It takes time to synthesize: to incorporate disparate examples, to look at the problem from different angles, to find patterns. But when you do you’ll often find obvious no-brainer ideas whose indirect effects create huge amounts of value. Taking a step back will help you peer further out into your time horizon and better intuit the indirect effects of various actions.
For your work schedule, this might look like reserving four days a week for the hustle and bustle of executing — scurrying between meetings, commenting on docs, slinging slack messages. But reserve one day a week where you don’t take meetings and disable notifications, allowing you to take a step back and think expansively. Think about what ideas, had you implemented them earlier this week, would have saved you significant effort or created better outcomes. Think about the cheap things that your gut tells you will have large positive indirect effects, even if they’re hard to explain to others. Then do them!
When you’re building a product or a platform, this takes the form of not just implementing exactly what the customer asked for: a quick but over-fit solution. Instead, think just a bit more broadly, and overbuild by just a little. Make the feature just a bit more general, or just a bit more cleanly implemented.
When you take a step back, guide yourself with the simple “Which action am I almost certain to not regret in the future, and have some possibility of actively loving?” Survive, then thrive.
Stay within your iterative adjacent possible
Your adjacent possible is the set of actions that you could take that are almost certain to work: the actions that are within your grasp.
When you’re aiming for hyper growth it’s tempting to aim for outcomes that are significantly beyond your reach: to swing for the fences with big bold bets.
But these plans almost never actually come to fruition. More often they get increasingly difficult as you execute on them, and as it becomes increasingly clear that the plan isn’t working you get stuck in a spiral of increasingly short-term fixes that inadvertently take you further and further from the goal until everything grinds to a halt or explodes.
Aiming at the edge of your adjacent possible, but still well within it, will allow you to rack up accumulating wins, and give you more of a chance to tweak your approach as you learn more, responding to surprising new information. Push your boundaries, but always stay within the adjacent possible.
If you plan for 10% lower output, you will often get 10x the actual returns. This is because the adjacent possible is like a cliff. If you aim just a bit beyond off the edge you’ll get significantly less return than if you aim just a bit shy of the cliff edge.
Remember that your adjacent possible is not a one-time thing; as you take more actions, new adjacent possibilities open up. If you have a clear goal that you sight off of and iteratively take no-brainer, safe actions in your adjacent possible, over time you can arc to wildly better outcomes with minimal risk. Instead of pushing beyond your adjacent possible, increase how quickly you can iterate through small slices within your adjacent possible.
Be rigorous by letting go of excess precision
Everyone wants to be rigorous whenever they can. But in our haste, we end up doing a grotesque performative version of rigor: lots of made-up numbers, lots of unnecessary details, lots of footnotes.
The mark of true rigor is that the closer any audience looks at the argument the more convincing they find it.
Precision is inordinately expensive to create. In many cases it’s fundamentally impossible — the further you peer into the future, the more uncertainty there is, increasing super-linearly. In addition, the cost of precision goes up super-linearly for increasing amounts of precision. Trying to get high levels of precision far into the future is an impossibility. In many cases you will spend extraordinary amounts of time on increasingly misleading faux precision: a dangerous waste of time.
Often you need far less precision than you think to make decisions. One option has super-linear returns whereas one has linear returns? Go with the former because over sufficient amounts of time, no matter the growth rate, it will win. You’d do the same action no matter if the number came in on the lower end of the plausible range or the higher end? Then skip the analysis!
Be willing to say “I don’t know,” and be OK with calibrated bets. If the decision-maker has skin in the game and is on the hook for results over the long-term, then they have an interest in making calibrated high quality bets.
Be curious about disconfirming evidence
Being rigorous means seeking out disconfirming evidence and seeing how to incorporate it into a holistic view — which sometimes means that the idea you were excited about turns out to not be worth developing after all. Be open to new information, especially if it’s unlike the information you already know. If you find surprising information, be curious and seek to find a way to synthesize it into your worldview. It takes only one bit of disconfirming evidence to invalidate an otherwise great plan; if you find it early you can save huge amounts of wasted effort.
In a hard-charging environment, any information that throws into question the viability of the big bold bet will be treated as subversive. If everyone believes the impossible thing is true, then maybe it will come true. But if no one believes, then everyone will sandbag and it definitely won’t happen. This approach is viable only in a forgiving environment. In a harsh one, it is deadly.
If you see disconfirming evidence — anything that is surprising and possibly relevant — don’t be shy about it. Dig in, be curious. Uncovering new constraints that you had missed before will help you tweak your plans to make them more likely to actually work.
Don’t be shy about digging into disconfirming evidence when you find it, but also don’t be shy about sharing it. You don’t have to be a jerk about it, but do raise it respectfully and collaboratively with the working group, understanding that it is just one piece of the puzzle and that you may be mistaken, or other factors might balance it out. The only way to have true rigor is to actually have the ideas withstand the tests of the real-world. It may seem like you’re being a team player by pulling your punches during discussion, but you’re doing a disservice to the idea by allowing it to be brittle before making contact with the harsh real world.
Take the sting out of failure
Disconfirming evidence is how you learn and improve your mental models about the world, allowing you to act with increased confidence.
Failure is a more general form of disconfirming evidence; it shows that what you implicitly thought was true was not. Failure is not something to avoid at all costs. Failure is inevitable, because it’s not possible to have perfect clarity about the world into the future.
The trick is not so much to minimize how much you fail, but to maximize how much you learn from that failure. Failure can make you stronger if you learn from it.
Failure provides an opportunity to learn, but it also has a real danger of harming you and possibly knocking you out of the game — a possibility that grows the larger the failure. Smaller failures maintain the learning possibility, while significantly minimizing the chance of a game over.
You should structure your effort so you are actively likely to fail when it’s fast and cheap. Set up the shortest feedback loops you can, kick the tires of the biggest looming failure points early, be on the lookout for evidence it’s not working, and don’t be shy about calling it out when you find it.
Focus on experimenting by sending out small, cheap probes, and then ensuring that you actively learn from them by being curious when they fail.
In an organization it’s easy to fall in the trap where no one wants to show they have failed. From the leaders on down, everyone should acknowledge and celebrate the discovery of smaller failures and model the behavior as seeing them as not only inevitable but actively good.
If you aren’t failing constantly in the small then it’s likely those micro-failures, instead of helping make you stronger, are accumulating behind the scenes into the potential for an increasingly dangerous macro-failure.
If you aren’t having micro-failures often, then you’re playing it too safe, and you aren’t learning enough. Failures help you probe the environment and get a sense for how conditions are changing and where their limits are.
Clear the bar of good enough
One of the ways to get micro-failures early is to get your solution exposure to the real world as early as possible.
It’s tempting to make everything we do as polished as it can possibly be. But the amount of effort it takes to increase the fidelity of something has quickly diminishing returns and a clear asymptote. Perfection is infinitely expensive.
If you had infinite time then, sure, invest as much time as you could. But in the real world, the most scarce resource is time. If you keep on perfecting the thing in a cave, what you finally ship might land with an unsatisfying clunk — or worse, you might die before you ship anything.
When something is out in the real world getting used, you get significantly better signal on how to make it better, from real users using it for real things. In addition, you’re creating real value, which might help reduce or eliminate your burn rate.
Often, the bar for “good enough” is lower than you think it is — especially for a self-selecting set of early adopters. If someone is drowning, they won’t mind if there’s a bit of a stain on the life preserver you throw them.
Clearing a “good enough” bar before getting in the hands of users is not a shrug: it is an active choice, acknowledging that you don’t have all of the answers, and getting all-important feedback loop.
Aim to “minimize the number of people who have such an actively bad experience they’ll never use it again, secondarily maximizing the number of people who have a great experience.” Survive, then thrive. One way uou can do this by having the product be cheap to use — a lower expected cost means it requires a lower expected value to clear the bar of useful.
Another powerful technique is to roll out a feature in an obscure location to start, for example buried deep in your developer docs. Only the most motivated — and thus resilient — users will find the feature, and their bar for “good enough” will be inherently lower than the mass market. That gives you the ability to get the all-important real-world feedback with decreased downside.
Note that “good enough” typically requires something that is not only a useful solution for real users, but also that has a viable business model that at the very least pays for itself as part of the larger system. It’s easy to think you’re a business genius when really you’re just giving away dollar bills for ninety cents. This may sound obvious but it’s extremely easy to miss the true cost of offering a service if you don’t take a broad enough view, including customer acquisition cost, the cost of service and support, and others.
Continuously improve what’s working
Of course, if you just ship “good enough” things and never improve them, you will have only done one half of it. Survive, but then thrive.
Look at how real people are using your products. Invest more to improve them where they’re being used — even if it’s not the part you had thought would be most valuable.
Invest energy in line with how much momentum you see. Continue taking no-brainer, small actions to respond to real-world results. If you ever get to the point where the investment stops yielding returns, don’t invest more — it’s OK, you only stuck your neck out a little bit. But keep an eye open because it’s possible that later the conditions will shift and you’ll start seeing return, at which point you should restart the incremental investments.
Look for the customers who are crawling through broken glass to use your product, because that demonstrates that they’re hyper motivated. When you see that, there are almost certainly other adjacent users who have similar needs but a lower pain tolerance. Reducing the amount of broken glass to crawl through is a no-brainer: you should see your incremental investment unlock incremental returns at a greater rate as you clear the bar of good enough for adjacent users. If you don’t, no sweat, just invest your next bits of effort elsewhere.
Your users are a swarm, telling you where to improve your product and where it’s creating the most value. Don’t tell them what they are allowed to do (modulo security and legal constraints), instead give them freedom and see what they do so you can use their real-world actions to chart your path of improvement.
When you shipped the good-enough first take, you almost certainly took a few shortcuts to get it out the door cheaply. That’s totally fine and often necessary, but those shortcuts are little stochastic time-bombs of uncertainty that might spontaneously combust and produce a dumpster fire or worse. Clean up after yourself and reduce shortcuts over time.
Shortcuts also tend to pile up and make it increasingly difficult to execute on new things — increasing your basal metabolic rate. When you don’t know what else to do, invest some time in cleaning up what you’ve already built. Sometimes that’s the internals, and sometimes it’s exposed to end users. As a heuristic, just make the product work more closely to how users would assume it works. A self-steering optimization metric is, for a given set of functionality, minimize the number of words of help center documentation that would be required to comprehensively document its full behavior.
The amount of effort you spend on paying down debt should scale with the amount of accumulated debt and the length of your runway. The benefit of paying down debt is huge but indirect, which means it’s easy to forget about in the moment. You should almost always do more than you think you should.
Focus on things with compounding returns
The best strategies have some kind of compounding returns: the outcomes keep getting better and better at an accelerating rate for the same cost. Your long-term strategy must have compounding returns of some form or it’s not worth doing. Compounding returns seem magic but they’re reasonably common. Any time you have some kind of preferential attachment effect, you’ll naturally have a compounding loop.
As a heuristic, compounding loops show up often where there’s a swarm of independent actors. If members of the swarm are making decisions partially based on what other members of the swarm are doing, and your product has Product Market Fit (a growth rate larger than 1), you’ll have a compounding loop. Network effects are a classic example, and this is also what causes the compounding loop of ecosystems. But even word-of-mouth for a delightfully great product has compounding returns: the more people that use it, the more people there are actively evangelizing it.
In contrast, many times you think you have linear returns actually have sub-linear returns. Shortcuts tend to accumulate, making progress take an increasing amount of time. Coordination costs grow as headcount grows, creating less output per unit effort. If you think the returns are linear, assume they’re actually sub-linear.
If one option has compounding returns, and one doesn’t, then pick the one with compounding returns, no need for an in-depth analysis. Good enough.
Don’t hold out for any one miracle, but have exposure to many
It’s tempting to hold out for a miracle. “If we just got approval for 20 more heads, we could build this big bold bet.” “If we just landed this one big partnership it would fix everything.” If you’re holding out for any single miracle, then you’re almost certainly over-investing your time in a lottery ticket — a strategy that will almost certainly lead to death.
Assume there is no savior, that that miracle won’t happen. Take matters into your own hands, and do the best with what you’ve got.
Of course, miracles do happen sometimes, in wonderful but unpredictable ways. If you’re clearing the bar for survival, and a friend gives you a few free lottery tickets, it doesn’t hurt to keep them. The likelihood of a miracle at any one time is low. But if you survive long enough, and have a broad enough diversity of cheap seeds planted, the likelihood one of them will miraculously sprout is a near-guarantee.
Create space for innovation
One way to create more exposure to miracles is to create more space. Your employees are a swarm of creativity and intention. If you crack the whip and force them to be totally over-subscribed on crank-turning execution they will have no time to take a step back and find clever and potentially game changing ideas. But if you give them just a little bit of space, they’ll plant seeds, one of which might turn out to be a miracle.
Think of it like a field of dandelions. Each seed is extremely cheap. With a broad enough field of diverse seeds it’s a near certainty that at least some of them take root.
When times are tight it can be tempting to tamp down on extra capacity, to “put more wood behind fewer arrows”. But your dandelion fields are your source of miracles, the novel upside. Planting seeds is not a ZIRP phenomena — it is a fundamental source of innovation in every environment. Planting a diversity of cheap seeds has a tiny cost, but significant (if unpredictable) upside. Your employees will love planting seeds for their own enjoyment, and every so often one of them will grow into a miracle. No matter how tight things get, if you’ve cleared the bar for continued survival then make sure you have some space for dandelion fields. In companies this might take the form of hackathons, or celebrating delightful, weird little experiments people try out.
Some dandelion fields will be clever little weird ideas the specific employee is extremely interested in for some reason. There’s also a class of improvements that have practically no downside and are easier to do than to explain. For those, giving a bit of space will allow them to just be done… and often they’ll have huge positive indirect effects.
Embrace combinatorial innovation with a scavenger mindset
When you’re in a scarce environment, it feels like you have to make do with linear, obvious solutions. But actually a scarce environment is just as conducive to game-changing innovation.
The model of innovation that we’re most familiar with is what might be called focused innovation. You come up with an idea and then you put your head down and invest capital and time into making it a reality. Most of the time what you build is just OK, and a lot of the time you give up before you actually ship anything. The quality of the idea and the amount of effort you can sink into it are the primary competitive differentiators. It’s a vertical approach. Hopefully the idea ends up being a worthwhile one!
It turns out that the vast majority of ideas that can be conceived turn out to not be viable. Holding out for a given focused innovation investment is like holding out for a miracle.
Another model of innovation might be called combinatorial innovation. You take a scavenger mindset, looking for off-the-shelf components that can be duct-taped into novel combinations. The diversity of the components and the number of combinations you can try are the primary competitors differentiators. The cheaper the combination, the better: use duct tape and don’t overthink it, just particle-collide ideas together until you find something that works. Good enough! It’s a horizontal approach.
It might seem like the most game-changing innovation comes from focused innovation, but in practice the seeds of most game-changing innovation comes from combinatorial innovation. The farther afield the ideas are, the more likely they are to not have been tried before, which means if you find a powerful combination it’s likely to be game-changing.
When there are a lot of components waiting to be combined, combinatorial innovation is extremely powerful. With the excess capacity of today’s LLMs, combined with decades of under-utilized research published by large tech companies, there are a multitude of powerful components just waiting to be combined even in this harsh new era.
Invest in trust
Trust is the lubricant of all cooperative endeavors. Without trust, every bit of ambiguity is treated as though it harbors a nasty surprise. With trust, ambiguity isn’t a deal breaker. Key to surviving is being able to navigate ambiguity together.
Trust shows up any time humans have to work together: whether as employees in an organization, or between customers and the provider of the product they use.
Trust is one of those things whose effects are felt almost entirely indirectly. That means it’s easy to get short-term wins by burning trust, treating your user or your coworkers transactionally.
Trust is not something you can force someone to do; it must be earned, based on demonstrating, over repeated interactions, that you are worth being trusted.
One of the ways that people build trust is by repeated low-stakes interactions. Socializing at work is one of the best ways to create a web of high-trust connections that span the organization. It’s not only enjoyable for its own sake, it also creates significant indirect business value.
Create high-trust teams with a diversity of perspectives
Trust allows people to feel comfortable raising disconfirming evidence and taking risks together. High-trust teams are significantly more resilient and more likely to have truly great results.
Most important things are impossible to build alone. Unfortunately, working as a team is significantly harder than working individually.
If you want to have predictable, good results, then staff your team full of people who are all alike: who look at problems in a similar way, who have similar skills and experience. They’ll automatically trust each other.
To execute a team needs to have a good idea that they all agree to. A team where everyone is the same will listen to one another and be able to easily agree to the best idea among them… but they’re limited by the variance with a low ceiling.
A diverse team is more likely to have a great idea in at least one head, given the higher variance. But the problem is sharing it and getting others to agree on it. If you don’t have trust, then you’re limited to the best idea that everyone can agree to, the lowest common denominator. Diverse teams will start out with less trust, so will do worse than non-diverse teams. But if a diverse team can build trust, then they’ll listen to even surprising ideas and might come to realize they’re great and then all agree.
Trust allows people to give the benefit of the doubt: “This idea is surprising and crazy to me but let me sit with it and unpack it — the fact you think it’s a good idea means I should assume it’s good too, just in a way I don’t understand yet”.
To be high-performing, teams must be diverse and also have earned high trust among each other. They need to engage in collaborative debate, using positive-sum thinking to create holistic answers that are more rigorous than what any one of them could have created on their own. Teams must seek not to come to a bland consensus but to synthesize the various perspectives into an opinionated outcome better than what any one member could have created on their own. A good pattern is to have one decider (allowing them to be opinionated and not just pick a bland consensus answer) with everyone else making sure to share diverse disconfirming evidence (making sure the idea is less likely to have lurking problems that make it nonviable).
For groups to have collaborative debate, they must have psychological safety. People sometimes erroneously assume that psychologically safe spaces are ones where unsafe thinking is not allowed. But nothing could be further from the truth — it is only in spaces with psychological safety when bold, unsafe ideas can even be considered.
Trust in diverse teams is where you’ll get the most rigorous thinking. Rigorous thinking is not a nice-to-have in a harsh environment.
A high-performance culture is important. However a naive form of that culture can quickly turn into “people are afraid to share disconfirming evidence,” which creates a dysfunctional and unproductive environment. What you really want is everyone holding themselves to high standards and inspiring others to do so, too. Often that requires leadership to model behaviors that go exactly against the naive form of high-performance, like admitting that you don’t know the answers or acknowledging something isn’t going how we’d hoped.
Reduce the need for coordination
As teams grow, the amount of coordination overhead grows super-linearly. This is an inescapable phenomena. The coordination cost can come to dominate all other investments of energy. Tactics that used to work fine when the group was smaller start becoming excruciatingly inefficient.
Coordination is fundamentally costly. The trick is to reduce the need for coordination.
You can do this by preferring simple solutions that use components that already exist. If you need to rely on another team to build something, then you need coordination — and that coordination will cause it to be an order of magnitude more expensive than it otherwise might be if you could just assemble the solution yourself out of off-the-shelf solutions.
This is one of the reasons that cleaning up things you’ve shipped is such a no-brainer; you can introduce more clean boundaries in your solution, making it easier for other teams to self-serve by using the building blocks you factored out. The more building blocks there are, the cheaper combinatorial innovation becomes.
Another way to reduce the need for coordination is to reduce the rate of pivots. When you have highly interconnected teams, and one of them pivots their plans, they need to re-coordinate with everyone else. A useful mental model is to see periods of coordination as periods of chaos. Imagine that the ideal default state of things for an organization is that everyone agrees on a plan and is executing it. If the plan is invalidated, a new plan must be agreed upon. Agreeing on a new plan can be extraordinarily costly, and gets more costly the more uncertainty there is, and the more bottom-up the org is. The cost goes up super-linearly with the number of people who must say yes. If another team has a dependency on the plan, and they do not have enough excess capacity or slack to devote some time to coordinating on a new plan, they also get put into a chaos state until the new plan is decided. For tightly interrelated teams with minimal slack, this can lead to a cascade through the organization when a plan is invalidated.
This is one of the reasons that teams keeping unallocated capacity in the plans can significantly improve actual output — there is more slack to absorb chaos without transmitting it further in the cascade.
Don’t aim for perfect coordination now but convergence eventually. Premature coordination is death, because it forces alignment of things that might be different for good reason, massively slows down their individual speed of execution (making it more likely that none of them successfully build something), and puts more eggs in one basket.
Taking inspiration from the live oak
That’s all there is to it. These techniques, when combined, create resilient slow growth.
None of these techniques are individually that surprising. However when taken as a whole the playbook might seem foreign, or even backwards.
It will seem like you’re giving up, or losing something you once had. But once you embrace this way of working, you see that it can create vastly more real world value, at significantly lower risk, than the brittle high-growth toolkit. In the current environment, a resilient playbook is an existential necessity.
The time has come when we must let go of our naive reverence of the fast-growth eucalyptus and instead take our inspiration from the enduring growth of the mighty live oak.
