The leadership books worth finishing
The leadership shelf is the worst shelf in Waterstones. Most of what's on it is one decent idea stretched to 280 pages, wrapped in a cover with a lighthouse on it. I can say this with some authority because I owned about fifteen of them. I've read maybe four. The ones that would sit on the shelf behind you on video calls, subtly trying to make you look ambitious.
So the filter for this list is simple: a book only counts if it changes what you actually do on Monday morning. Feeling inspired on the train doesn't count. I've felt inspired on the train plenty of times, usually while listening to an audiobook. It tends to wear off somewhere around Milton Keynes.
High Output Management, by Andy Grove
Grove ran Intel through its ugliest years, and the book reads like it: no lighthouse, no parables, just a man explaining how to run things with the same seriousness he'd apply to a factory floor. Because for him it mostly was one.
His big idea is that a manager's output is their team's output, plus the teams they influence. That's it. Your own to-do list barely matters. What matters is how much each hour multiplies. An hour writing a clear one-page brief that saves six people a week of confusion is a brilliant hour. An hour in a status meeting reading out what everyone already knows is a wasted one for the whole room, and Grove would tell you to cancel the meeting. He is right, and I say that as someone who has run that exact meeting more than once.
The bit you can use tomorrow is what he calls task-relevant maturity: how closely you supervise someone should depend on their experience with this specific task, not how senior they are or how much you like them. Your best engineer doing something for the first time needs more structure than your newest hire doing something for the fiftieth. Most micromanagement, and most of its neglectful opposite, comes from getting that one dial wrong. I have personally managed to get it wrong in both directions with the same person in the same month, which I'm told is impressive. Written in 1983. Still sharper than anything published this decade.
Turn the Ship Around!, by L. David Marquet
Marquet took command of the USS Santa Fe, officially the worst submarine in the American fleet, after spending a year training on a completely different submarine. His technical knowledge was suddenly useless, which made his orders suddenly dangerous. So he stopped giving them.
What makes the book worth your time is that he replaced orders with a mechanism, and mechanisms travel. His crew stopped asking "permission to submerge?" and started saying "Captain, I intend to submerge the ship", followed by their reasoning. The person closest to the information does the thinking. The leader's job shrinks to catching dodgy reasoning before it becomes a dodgy action. The worst boat in the fleet became the best one, with the same crew.
If you're a project manager, this should land harder than most leadership writing, because you've never had real authority anyway. You can't order a developer to do anything. You can't order a stakeholder to do anything. They've mostly just been humouring you, which is a sobering thought to have on a Tuesday. Marquet's approach doesn't need authority to work, which makes it one of the few leadership models built for the job you actually have. And you can test it this week: next time someone brings you a decision, ask them what they intend to do instead. The first few answers will be wobbly. That wobble is the amount of thinking you've been quietly doing for other people.
Crucial Conversations, by Patterson, Grenny, McMillan and Switzler
Fair warning: this one is extremely American. There are acronyms. There are worked examples featuring people called Chuck. You will wince, and you should read it anyway, because it deals with where leadership actually falls over. Projects rarely die of bad strategy. They die because somebody dodged a conversation for six weeks: the underperformer nobody mentioned, the deadline everyone privately knew was fantasy, the requirement that made no sense and was never questioned.
The idea worth the cover price is separating facts from the story you've built on top of them. "You sent the report late twice" is a fact. "You don't respect this project" is a story. Open with the story and a performance conversation becomes a row. The book trains you to lead with what you can actually see and hold your interpretation loosely, which sounds obvious, and under pressure almost nobody does it. I once opened with the story and then spent the next forty-five minutes trying to reverse out of it, so I can confirm the failure mode is available to everyone.
It slots in neatly behind the other two. Grove tells you what the job is, Marquet tells you how to hand it out, and this one covers the moments where both of those collapse into two people disagreeing in a small room.
The one to skip: Start With Why, by Simon Sinek
Sinek's TED talk is eighteen minutes long and genuinely good. The book is that talk stretched to two hundred pages, and the stretching shows you the problem: the whole argument runs backwards from its examples. Apple succeeded, Apple has a "why", so the why must have done it. The companies with equally lovely whys that went bust don't get a chapter, because they'd ruin the story.
And it fails the Monday test completely. Finish it and try to write down what you'll now do differently. You'll get "communicate purpose more", which you already believed, said better in the talk, which was free.
Read fewer, apply harder
Three books is plenty. That's the actual recommendation. The people getting through thirty leadership books a year are usually avoiding the harder job, which is picking one idea and running it against a live team for long enough to find out if it survives. Choose Grove's hour-multiplying maths, or Marquet's intent statements, or facts-before-story, and use it for a full quarter before buying anything else. The shelf can wait. It's very good at waiting. Mine had fifteen books' worth of proof before I gave them to a charity shop.
