8 People: Gardening Your Teams
This chapter is currently in draft form.
People management sits third in the hierarchy of leverage, but don’t mistake that for low importance. While strategy and structure create the framework for delivery, people execute within that framework. Get the people lever wrong, and even perfect strategy and structure won’t save you.
Before diving into hiring, development, and performance management, let me start with the most practical advice I can give you: cultivate a relationship with HR. I don’t care what you think about HR personally. We in technology have this extraordinarily stupid idea that HR is the enemy. They’re not. They’re not going away. And that relationship drives what you can accomplish everywhere else in people management. Without it, you’ll be hindered at every turn. With it, you’ll be supported when you need to make difficult decisions.
8.1 The Garden Metaphor
My general philosophy for people management is simple: think of your teams as a garden. You’re constantly growing, fertilizing, watering, and also pruning. Both activities are necessary for a healthy garden, and both are necessary for a healthy team.
You grow the people you can. You prune the people you can’t. This is a constant process, not an annual event tied to performance review cycles.
Which side you lean toward depends on your organizational context. If you’re leading a team of five people in a startup and your people are performing well, you won’t be pruning very often. If you’re managing 150 people in a larger organization, you’ll be pruning regularly. But you should always be growing.
The garden metaphor matters because it reframes difficult decisions. Pruning isn’t punishment; it’s maintenance. A gardener doesn’t prune a branch because they hate it. Some branches take resources without contributing to growth. Others grow in directions that harm the whole plant.
8.2 Hiring: Choosing What to Plant
Before you grow anyone or prune anyone, you choose what to plant. Hiring is the cheapest lever in the garden and the highest-leverage moment you get, because everything downstream, growing people up and growing them out, costs more than choosing well in the first place. A good hire is upstream mercy. It spares you the trembly afternoons later.
I’ll be honest: I don’t have a model for hiring. I have heuristics. After enough years you stop looking for a rubric and start pattern-matching on proxies, and the proxies I trust all point at the same thing.
Hire for demonstrated learning, not current knowledge. The specific stack a candidate knows today is the least durable thing about them; this book spends whole chapters on how fast that ground moves. What survives every shift is the ability to learn, the desire to learn, and a practiced habit of doing it. That is the trait. Everything else is a proxy for it. It is also, not by accident, the same capacity The Bones of Growth says organizations run on. That capacity starts at the hire.
Most of my green flags are just different windows onto that one trait:
- Waterloo graduates. The co-op program means they’ve shipped real work in real organizations before they graduated. They arrive already de-risked on whether they can operate.
- Electrical engineers who write software. They trained under hard physical constraints and learned to think in systems. Software’s abstractions feel forgiving by comparison, and they reason about the whole system, not just the code.
- Career-transitioners. The doctor who became an engineer, the musician, the anyone-else. They chose this deliberately as adults and proved they could learn an entire field from zero. That’s the trait, demonstrated in the loudest possible way.
- The self-taught who’ve succeeded at the higher levels. Nobody made them do it. They show you ability, desire, and practice directly, with no institution in between. I’ll own the bias: that description also fits me, and I have to correct for how much I like seeing myself in a résumé.
The red flag is the negative of the same lens: the one-tool career. A developer who has only ever worked in a single language, a single stack, a single way of thinking has never once proven they can teach themselves a genuinely different one. I can’t think of a single strong engineer in my entire career whose résumé showed only one. This isn’t snobbery about any particular tool; the person who’s ranged widely is fine wherever they happen to have landed. It’s the mono-tooled anything that’s the tell.
Then there’s a move most leaders skip: filter before the funnel. Candidate pools have wildly different concentrations of quality. The largest, most accessible pools, the technologies everyone lists because they’re everywhere and easy to start in, spread quality thin, so the cost of filtering inside them, in your time and attention, the scarcest things you have, is enormous. The smaller pools, the ones whose barrier to entry is that they take real effort to learn at all, are a fraction of the size and a multiple of the concentration. And it’s the same lens again: nobody drifts into the hard corners of the craft, or into a mid-career transition. Those pools come pre-selected for the appetite you’re screening for, so they do most of the sorting for free.
Hold all of this loosely. These are concentration bets, not verdicts on human beings. There are brilliant self-taught, single-language, state-school developers, and these heuristics will miss some of them; that’s the price of any filter. They tell you where to start looking, not who’s allowed through the door. They’re lenses, with all the false negatives lenses carry. Use them to aim your attention, then actually look at the person in front of you.
Underneath every heuristic sits one requirement: your bar has to be high, and you have to hold it when holding it hurts. A high bar is what earns you the right to be this picky, to accept all those false negatives, to leave a seat empty rather than fill it with someone who’ll tax the whole team. The bar is only real if it survives the pressure to lower it: the open role, the impatient recruiter, your own manager asking why it’s taking so long. Anyone holds a high bar when candidates are plentiful. The discipline is holding it when they’re not.
And sometimes you won’t. Sometimes you have to put a body in a seat, because you’re growing too fast to wait, or an empty org carries its own cost, or the work simply won’t hold until the right candidate appears. I don’t like it, and I try not to do it. But it’s real, and pretending otherwise is its own dishonesty. You use the leverage you have.
The discipline in that moment comes in two parts. First, name the compromise, to yourself and to whoever you owe honesty: this hire is below the bar, here’s why I’m making it anyway, here’s the debt I’m taking on. A compromise you’ve named is a decision; a compromise you’ve hidden from yourself quietly becomes your new bar. Second, and this is the part leaders forget: the moment they start, they’re a person, not a discount. The bar didn’t move. You did. Hiring someone below it isn’t permission to expect less of them; it’s a commitment to close the gap yourself. If you don’t honestly believe you can grow them to the bar, you haven’t found a pragmatic compromise, you’ve just made a bad hire with a story attached. And if you do believe it, the work starts the day they do, which is where the rest of this chapter begins.
Hiring managers is the same lens held higher. The bar rises because the blast radius does: a weak engineer costs you their own output, a weak manager costs you their whole team’s. But you’re still looking for the same thing, the appetite to keep learning a job that keeps changing underneath them.
8.3 Growing People: Situational Leadership in Practice
For CTOs, growing people means growing managers. This happens through delegation: giving your direct reports ownership of meaningful work and coaching them through the experience using radical candor. Real growth comes from progressively handing over responsibility for systems, teams, and decisions that matter to the business, then mentoring them through delivery with frank, unambiguous feedback.
Delegation serves dual purposes: it develops your managers while returning bandwidth to you. Every responsibility you successfully delegate is capacity you regain to focus on higher-leverage work. But delegation without coaching is abdication. You must stay engaged through honest, direct conversations about what’s working and what isn’t.
8.3.1 Matching Your Approach to the Person
Effective delegation requires matching your approach to where someone sits developmentally. Hersey and Blanchard’s situational leadership model (Hersey and Blanchard, 1977) captures this: leadership style should vary based on the person’s competence and commitment for the specific task at hand.
A new manager facing an unfamiliar challenge needs direction: clear expectations, frequent check-ins, explicit guidance. You might delegate implementation planning but retain architecture decisions. As their competence grows, you shift to coaching: challenge directly, care personally, tell them when they’re making mistakes, ask probing questions, provide context. They might own team performance conversations with your active support.
Once they’re competent but still building confidence, you step back further. You’re available when requested, encouraging their decision-making, but no longer driving. And eventually, you reach true delegation: they own the outcome completely and come to you only for major decisions.
The mistake most CTOs make is treating delegation as binary: you either own something completely or hand it off completely. Effective delegation is graduated, moving through these stages based on the person’s development level for each specific responsibility.
Development opportunities emerge constantly: new teams need leadership, critical systems need ownership, major projects require coordination. Match the opportunity to the person’s development needs. The real leverage comes when your managers adopt the same approach with their teams. Make your managers effective at situational leadership, and that capability cascades through the organization.
8.4 Pruning: The Necessary Discipline
Firing isn’t optional in healthy organizations. It’s necessary maintenance. If you’ve never had to let someone go, either you’re incredibly lucky with hiring, or you’re not maintaining the garden properly.
This is where that HR relationship becomes critical. When you identify performance issues (and you will if you’re coaching actively), bring HR into the conversation early. Not to start a termination process, but to create a development plan with clear expectations and timelines. Your job is to decide whether to grow someone up or grow them out.
The mistake most technical leaders make is waiting too long. They hope problems will resolve themselves, or they work around poor performers instead of addressing them directly. This is unfair to everyone: the underperformer doesn’t get the feedback they need, the team carries extra load, and you burn political capital managing around the problem instead of solving it.
When pruning becomes necessary, do it quickly and cleanly. Drawn-out performance improvement plans rarely work and create anxiety for everyone. Help people transition out with dignity and support.
I still haven’t learned to do this well. After years of practice, I’m still trembly and useless after letting someone go. I schedule these conversations for late afternoon and take the rest of the day off because I know I won’t be good for anything else. I do it because it needs to be done, but most of the time I hate it. If you find firing easy, something is wrong with you. If you find it hard, that’s the appropriate response to ending someone’s livelihood.
Your ability to execute this depends on your organizational culture and your relationship with HR. Some organizations support quick, clean transitions. Others require extended documentation processes. Work within your constraints, but understand that prolonged performance management rarely benefits anyone.
8.4.1 When You Can’t Prune: Building Around Constraints
Sometimes, however, you’ll encounter people you simply cannot touch. Leadership may have explicitly or implicitly put guardrails around certain employees for political, historical, or relationship reasons. In these cases, getting angry or continuing to push for termination wastes your political capital and accomplishes nothing. Instead, you must adapt your organizational design to work around the constraint.
I once managed someone who was an excellent individual executor but had limited ability to influence others or scale their impact. Give them a clearly defined task and they would deliver solid work. But as the organization matured and the bar for senior engineers shifted toward influence and team impact, they became less able to meet expectations. They couldn’t mentor junior engineers, couldn’t lead technical discussions, couldn’t drive cross-team initiatives. These were capabilities that had become essential for their level.
Under normal circumstances, this would trigger either intensive coaching to develop those skills or a transition out of the organization. But this person was protected due to their relationship with leadership. The path forward wasn’t to fight that constraint; it was to design around it.
I crafted a specialized role that maximized their execution strengths while minimizing their need to influence others. They became our go-to person for complex, independent technical work that required deep focus but limited collaboration. I restructured team interactions so their work fed into the broader system without requiring them to drive consensus or mentor others.
This approach had real costs. Other engineers noticed the different standards. Some questioned the fairness, and they weren’t wrong to ask. It required difficult conversations about why different expectations applied. There’s no good way to handle those conversations; you handle them as best you can. But fighting an unwinnable political battle would have cost far more while achieving nothing.
The reality is messier than any framework suggests. You’ll recognize these constraints when you hit them. Leadership makes it clear through their actions, if not their words, that certain people are off-limits. Once you recognize that reality, redirect your energy toward what you can control: craft roles that contain the impact, design workflows that minimize dependencies, and focus your development efforts on the people who can grow.
These situations are frustrating. Accepting the constraint doesn’t mean you have to like it. But dwelling on what you can’t change depletes energy you need for the work that actually moves your organization forward.
8.5 The Balance of Growth and Pruning
The right balance depends on context. High-growth startups can carry some underperformers temporarily because roles are fluid and hiring is fast. Mature organizations need higher performance bars because there’s less room for passengers.
Context changes, and your approach must adapt. The engineer who thrived in startup chaos might struggle as the organization matures. The manager who excelled with a small co-located team might falter with distributed teams across time zones. This doesn’t make them bad employees. It makes them people whose fit with the organization has changed. Your job is to optimize for delivery, which means recognizing that what the organization needs evolves constantly.
Your technical organization never fires anyone. Your CTO claims they have excellent hiring judgment and strong team performance.
It’s probably not a sign of exceptional leadership; it’s more likely avoidance of difficult personnel decisions. Technical leaders often avoid firing, hoping problems will resolve themselves or working around underperformers instead of addressing them directly.
Ask your CTO about their approach to performance management. Look for clear processes, regular feedback cycles, and evidence of difficult decisions when necessary. If they can’t point to examples of people they’ve helped transition out, dig deeper. The goal isn’t arbitrary firing, but ensuring your technical leader understands that team health sometimes requires difficult decisions.
The people lever requires more emotional intelligence than the others. You’re dealing with careers, livelihoods, and human dignity. But approached thoughtfully, it becomes one of your most powerful tools for building delivery capability. You’re not just managing individuals. You’re cultivating the human system that executes your strategy within your structure.