Hey there! This is a Book Edition 📚 — where I publish our review + summary of a famous non-fiction book in our space.
It is made possible by the book club in the community, where every two months we decide on a non-fiction book to read, and we review it together in a live event at the end of the period.
Over the last two months we have read An Elegant Puzzle: Systems of Engineering Management, by Will Larson, who is simply one of my favorite authors.
Will is CTO at Carta, ex-CTO at Calm, ex-Stripe and Uber — but most of all, for me, he is the author of countless, fantastic essays on Lethain.com, which I have often quoted and referenced on Refactoring.
An Elegant Puzzle itself is a collection of such articles, loosely organized into five main sections. The anthological nature of the book is important, because it makes it extremely different from other works you may have read in this space.
In fact, An Elegant Puzzle is probably, at once, the most interesting book about engineering management I have ever read, and also the hardest to read — for reasons I will explain.
It is also equally hard to summarize.
In most business books you can find a few, seminal ideas, which you can explain at a basic level in just a few sentences, and which are continuously explored throughout the book in increasing detail. Think of:
You can’t do this with An Elegant Puzzle.
This is an extraordinarily dense work, where basically nothing is ever repeated twice. If you had to find a common thread, you won’t find it in a specific topic, but more in the spirit of how such a wide range of problems are addressed.
In this review, I will try to capture this spirit; I will give you an overview of the book's main themes, and will comment on a few that were particularly interesting to me.
Here is our agenda for today:
🧩 Systems thinking — the true heart of the book.
📗 Book structure — how the work is organized, and what you will find in it.
🚀 Growth & hypergrowth — a both visible and invisible thread that runs through most of Will’s thinking.
🏢 Big vs small tech — reflecting on the ideal recipients of the book.
📌 Bottom line — my parting thoughts.
Let’s dive in!
🧩 Systems thinking
The second chapter of the book is called Tools, and it opens with a small essay about systems thinking. In my opinion this is the true heart of An Elegant Puzzle, and, if you ask me, it should have opened with it.
Systems thinking is a wildly overused term, but Will gives a very precise definition for it. Systems are combinations of stocks and flows:
🧱 Stocks — are accumulations of resources.
🔀 Flows — are streams that make stocks increase (inflows) or decrease (outflows).
This definition looks extremely simple and, in fact, through the use of diagrams, it turns systems into equally simple boxes and arrows.
Such apparent simplicity is part of the reason why, I suspect, many people feel somewhat authorized to give systems thinking a loose interpretation. Simple, though, doesn’t mean vague — on the contrary, stocks and flows are precise ideas that, applied well, can model infinitely sophisticated situations.
One of the examples provided is about DORA metrics. The four metrics are flows that impact their respective stocks, and together model the software delivery process as a clear, predictable system.
Now, thinking of your dev process as a system is still intuitive, or at least it should be, to managers and engineers who have been in the industry for more than a few years.
An Elegant Puzzle, though, applies this line of thinking to everything.
It addresses otherwise nebulous themes like team organization, culture, hiring, or hypergrowth, by uncovering their inner logic and creating mini frameworks that always feel grounded and actionable.
This is extraordinarily useful. In Will’s own words 👇
Once you start thinking about systems, you’ll find it’s hard to stop. Pretty much any difficult problem is worth trying to represent as a system, and even without numbers plugged in I find them powerful thinking aids.
So, “An Elegant Puzzle” may look like an obscure title for an engineering management book, but it all makes sense when you understand Will’s approach to problems. Engineering management’s challenges are explained as a series of puzzles which you have to solve by creating systems for them.
📗 Book structure
As anticipated, the book is a collection of essays. For the most part, each essay focuses on a single problem and works well in isolation, similarly to a Refactoring article (I hope!).
The essays are organized into five sections:
🏢 Organizations — explores organization design topics, like sizing teams, addressing hypergrowth, and the various stages of performance.
🔧 Tools — management tools across a wide variety of scenarios. From metrics, to migrations, to product management frameworks, and more.
📖 Approaches — more high-level and less prescriptive advice about circumstances where you may need to adjust your approach. It explores policies, saying no, management philosophies, and more.
✨ Culture — practical advice about nurturing an inclusive + effective culture. It dives into topics like creating opportunities for your teammates, dealing with hero culture, and managing freedoms.
🪜 Careers — finally, the book explores careers lifecycle topics, like hiring, performance management, and career frameworks.
My personal opinion is that, while individual essays are often invaluable, sections—as wholes—are not equally useful. For two reasons: 1) they often overlap, and 2) they are not exhaustive.
This is all by design: Will is very upfront about An Elegant Puzzle being a collection of thoughts and ideas, as opposed to a top-down manual about all-things-EM.
Still, I believe it is useful to comment more on this:
1) Sections often overlap
While writing this review, I could remember vividly many specific essays, but I often couldn’t remember which section they belonged to.
For example, “Setting organizational direction” is in Approaches, but, if you ask me, it could equally be in Organizations. Likewise, “Make your peers your first team” is in Culture but, to me, could just as well go into Approaches.
Sections’ boundaries aren’t always clear, and many topics overlap.
2) Sections are not exhaustive
Essays from the same section maintain some loose correlation, but they don’t cover it all. “Organizations” doesn’t have the pretense of telling you everything you should know about organizations, but rather covers Will’s best ideas about them.
This means that, sometimes, essays unfold in an order which is not easy to follow. E.g. “Guiding organizational change with metrics” is followed by “Migrations are the only scalable way to fix tech debt”, which in turn is followed by “Running an engineering reorg”.
🚀 Growth & Hypergrowth
An Elegant Puzzle is many things. An important one is: it’s the best book I have ever read about hypergrowth.
This framing isn’t explicit in the book, and maybe wasn’t even in the author’s intentions, but it is not surprising. When the book was written, Will’s most rewarding experiences had been at Uber and Stripe, where he saw his teams grow multiple folds in a few years 👇
It was only when I got the opportunity to work at Uber, which was growing its engineering team from 200 to 2,000 over two years, and then at Stripe, which was experiencing similar rapid growth, that I had the opportunity to truly refine my approach to management through exposure to an endless variety of challenges. There are few things peaceful about managing in rapidly growing companies, but l've never found anywhere better to learn and to grow.
Hypergrowth is explicitly addressed in one chapter, inside the Organizations section. The chapter itself is a gem — it includes some of my favorite insights in the whole book, like how hiring impacts productivity, or how systems rarely survive more than one magnitude of growth.
Many of these ideas look familiar or ring true intuitively, but Will has the ability to unfold and demonstrate them with the precision of a scientist, like physics theorems.
However, I feel that hypergrowth leaves its mark pretty much everywhere in the book, especially when you look at Will’s approach to topics.
In An Elegant Puzzle, advice is never static. It never deals in absolutes, but rather in ratios, formulas, and relatives. It assumes things are dynamic; it creates systems and modulates them based on team maturity, size, product stage, and more.
This goes hand in hand with the systems thinking approach. In fact, I suspect systems thinking got so wired into Will’s philosophy in part because of hypergrowth.
Systems naturally pay off in environments where you can create them once and then scale and reuse them multiple times. In slow-moving organizations, systems are good, but in fast ones they are indispensable.
🏢 Process & big tech
The flip side to this style is that it tends to create process-heavy solutions. To use the words of Reed Hastings from Netflix, it skews towards Rules & Process, instead of Freedom & Responsibility.
Process optimizes for consistency and reliability, which is not always what you want, especially in small teams.
For example, there is a long chapter in the book about creating a good hiring process. It is great advice, thoughtful and exhaustive. At some point, though, it explicitly recommends not to hire for potential, because potential is subjective: it is hard to capture in rules, and easily leads to bias.
This is one example where process raises your floor, but it also lowers your ceiling. In early stage startups, and small teams in general, you most definitely want to hire for potential. It’s one of the most important ways through which you can win great candidates vs big tech, while also not breaking the bank.
This leads to my next comment: An Elegant Puzzle can be appreciated by anyone in software engineering, but it is the most useful to managers and tech leaders working in big orgs.
That shows both in the choice of topics and in the type of solutions proposed. If you work in a small team, you may at the same time 1) not be interested in some chunks of the book, and 2) not agree with how some problems are addressed.
📌 Bottom line
An Elegant Puzzle is a collection of the best ideas about engineering management from one of the best thinkers in the space. It is incredibly dense and incredibly rich. In fact, I can’t think of any other book that covers so much ground on these topics, while also being so consistently rigorous and thoughtful.
This also makes it not an easy read. The density of information makes it hard to consume from cover to cover, and chances are not all the parts will be useful to you.
So, it works best as a reference manual — the one you keep on your shelf and open up when you have a specific problem, confident that Will has gone through it and created a system for it.
Personally, I loved it and, as a fellow author, was greatly inspired by it.
And that’s it for today! See you next week 👋
Sincerely
Luca