For years, I tried to lead like other people. I’d read about a leader I admired, adopt their approach, and wonder why it felt forced. It took me longer than I’d like to admit to realise that leadership style isn’t something you copy, it’s something you discover through practice, reflection, and a fair amount of getting it wrong.

The good news is that there’s no single correct way to lead an engineering team. The bad news is that your natural style, whatever it is, won’t be sufficient for every situation. The best leaders I’ve worked with aren’t the ones with the most charismatic style, they’re the ones who can shift between styles depending on what the moment requires.

Three Core Styles

Will Larson identifies three leadership styles that I’ve found genuinely useful as a framework, because they’re practical rather than theoretical:

Leading with policy. This means creating documented, predictable decision-making frameworks for recurring decisions. Instead of making the same type of decision over and over, you create a policy that handles it. “All production changes require a review from at least one other engineer.” “We don’t take on new framework dependencies without a written evaluation.” Policy-based leadership scales well because it doesn’t require your personal involvement in every decision.

Leading from consensus. This means pulling stakeholders together when a decision requires distributed context. No single person has all the information, so you facilitate a process where the right people contribute their perspectives and the group arrives at a decision together. This works well for complex, cross-cutting decisions where buy-in matters as much as the decision itself.

Leading with conviction. This means personally making a definitive decision when no clear proposal exists, when stakeholders are deeply at odds, or when the team is stuck. Sometimes someone just needs to say “we’re going this way” and commit. This requires confidence in your judgement and willingness to be wrong.

Your Default and Its Blind Spots

Most leaders have a default style, the one they reach for instinctively. Mine was conviction. Coming from an IC background where I was used to making technical decisions and defending them, my instinct was to assess the situation, form an opinion, and drive toward it. That worked well when I was right and the team needed direction. It worked badly when I was wrong, when the team had better information than I did, or when buy-in mattered more than speed.

If your default is policy, your blind spot is probably situations that don’t fit neatly into rules, novel problems, ambiguous contexts, or moments where the team needs a human decision rather than a process. Policy-heavy leaders can feel bureaucratic when flexibility is needed.

If your default is consensus, your blind spot is probably speed. Consensus takes time, and there are moments when the cost of delay exceeds the cost of an imperfect decision. Consensus-heavy leaders can also struggle with decisions where there’s no consensus to be found, sometimes the group genuinely disagrees, and someone needs to break the tie.

If your default is conviction, your blind spot is probably listening. Conviction-heavy leaders can steamroll good ideas because they’ve already made up their mind. They can also create teams that stop offering input because they’ve learned it won’t change anything.

Developing Range

Larson suggests a monthly practice that I’ve adopted: identify a problem that could be solved with a style you’re less comfortable with, and deliberately use that style. If you default to conviction, find a decision where you facilitate consensus instead. If you default to consensus, find one where you make the call yourself.

This feels awkward at first. Using an unfamiliar style is like writing with your non-dominant hand, it’s slower, less precise, and you’re very aware of the effort. But over time, each style becomes more natural, and you develop the ability to read a situation and choose the right approach rather than defaulting to the one that’s most comfortable.

The other thing I’ve found helpful is reviewing decisions after the fact with someone skilled in the style you used. If you tried leading from consensus and it felt clunky, talk to someone who does it naturally. They’ll often spot things you missed, a stakeholder you should have included, a question you should have asked, a moment where you accidentally reverted to your default.

Context Drives Style

Osmani describes several leadership styles, transformational, democratic, servant, situational, and makes the case that the best leaders combine them based on context. I think Larson’s three-style framework is more practical for day-to-day use, but the underlying principle is the same: context should drive your approach, not habit.

A start-up in crisis needs conviction. A mature team working on a well-understood problem benefits from policy. A cross-functional initiative with competing priorities requires consensus. The same leader might use all three in a single week.

The environment matters too. I’ve found that start-ups generally reward conviction more, decisions need to be fast, and the cost of being wrong is lower than the cost of being slow. Large corporates tend to reward consensus, the cost of alienating stakeholders is high, and decisions need broad support to stick. Neither environment is better; they just demand different defaults.

Authenticity Within Range

There’s a tension between developing range and being authentic. If you’re naturally quiet and thoughtful, trying to lead with loud conviction will feel fake, and your team will sense it. If you’re naturally decisive and direct, trying to facilitate a long consensus process will test your patience and everyone else’s.

The goal isn’t to become someone you’re not. It’s to expand the range of tools available to you while staying true to your core. A quiet leader can still lead with conviction, they just do it differently from a loud one. A decisive leader can still facilitate consensus, they just need to be more deliberate about holding back their own opinion until others have spoken.

What I’ve found is that authenticity comes from intention, not from style. If your team knows that you’re genuinely trying to make the best decision for the right reasons, they’ll follow you regardless of which style you’re using. What they won’t follow is inconsistency without explanation, switching styles randomly without helping them understand why.

The Style You Haven’t Tried

If you’ve been leading for a while and you’ve never felt uncomfortable with your approach, that’s probably a sign that you haven’t stretched enough. The growth is in the discomfort. The style that feels most unnatural is usually the one you need most.

For me, that was consensus. Learning to sit with ambiguity, to let a conversation run longer than I wanted, to resist the urge to just make the call, that was genuinely difficult. But it made me a better leader, because it meant I could access a wider range of approaches depending on what the situation actually needed, rather than what my instincts defaulted to.

Find your uncomfortable style. Practice it. You won’t regret it.