⬅️ Back to 10 🧠 Knowledge

Summary

Ben Horowitz's The Hard Thing About Hard Things is a field guide to the parts of running a company that have no playbook - the decisions that are emotional rather than analytical, made under existential pressure, with no good moves available. Its throughline is that hard things are hard precisely because there is no formula: leadership is the willingness to sit in that ambiguity, make the least-bad call, and own the struggle rather than wish it away. The struggle, in Horowitz's framing, is where greatness comes from.

A large share of the notes is about people and organization as the real product of a CEO. Horowitz treats communication architecture as the CEO's most important operational job - one-on-ones, reporting lines, and deliberate process design are how information and trust scale as the company grows and the easy things (communication, decision-making, common knowledge) get hard. He is blunt about the human cost of leadership: being a good company matters most when things go badly, layoffs must be done by direct managers, and political behavior always starts at the top. Hiring is for strengths not absence of weakness, training is among the highest-leverage things a manager can do, and the right kind of executive ambition optimizes for the organization, not the self.

The closing thread is identity and conviction. Strategy and story are the same thing - a company's "why" must answer why anyone should join, buy, or care - and great CEOs build internal intelligence to keep making good decisions as context shifts. The final lesson is personal: embrace the struggle, and on top of that embrace your own background, weirdness, and instinct.

The Hard Thing About Hard Things

Leadership & the CEO's job

the hard thing about hard things is that there's no formula to sort them out
leadership is the skill to get someone to follow you even out of curiosity
needs always trump wants in acquisitions and mergers
innovation requires a combination of knowledge, skill and courage
all decisions are objective until the first line of code is written - after that, all decisions are emotional
when you're building a company, you must believe there is an answer to your problems and don't focus on your odds of solving it
the secret of being a CEO - the ability to make the best move in situations when there are no good moves ^hthat-no-good-moves
most important decisions test the courage more than the intelligence ^hthat-courage
take care of people, product, profit exactly in that order ^hthat-people-product-profit
3 aspects that make a person worth following:
1. ability to articulate vision
2. the right kind of ambition
3. the ability to achieve the vision
truly great leaders create an environment where their employees have the experience that they matter more than the leader itself ^hthat-leaders-matter

People: hiring, training & layoffs

if you don't have a world-class talent, you won't be a world-class company
people are most valuable assets even in technological companies
training is one of the highest leverage activities a manager can perform ^hthat-training-leverage
functional training focuses on finding the most important skills to train your colleagues so that they can do their job well
there are only two ways how to improve employee's performance - motivation and training
an equivalent of being too busy to train is to be too hungry to eat ^hthat-too-busy
the only good reason to hire a senior person to the team is to acquire knowledge and experience on your part you can't fill with existing talent in your organization and that's why often promoting technical roles to executive roles works better than e.g. in sales as knowing internal tech stack is way more important for the role versus knowing client behavior/environment needed for a good sales person can be more easily acquired from the outside
it's important to hire for strengths rather for the lack of weaknesses ^hthat-hire-strengths
most layoffs happen because the company did not meet its goals / hit its strategy / it is more rare that it's a personal issue and it is important to admit company failure to prevent losing trust of those that remain
people won't remember everything working for your organization but they will remember the day of being laid off - that's why it's important that their direct managers lay them off and not HR or something similar as they don't have that strong relationships with those being laid off ^hthat-layoffs

Product management

in enterprise software companies - two most important roles are VP of sales and VP of engineering
good product managers crisply define the target, the "what" and manage the delivery of the "what", they don't give direction informally but they gather information informally ^hthat-good-pm
good product managers anticipate product flaws and build real solutions, bad PMs firefight all day ^hthat-pm-anticipate
bad product managers feel the best when they figure out "the how"
good product managers decompose problems, bad ones mold problems into one ^hthat-pm-decompose
vanity metrics might decently work for attraction but not for retention as retention metrics don't describe the problem of retention deeply enough
measurement debt is similar to technical debt and can harm the org from different perspective - occurs when overcapacity to counter-offer an employee that wants to quit or with no performance management being afraid of feeling "like a big company"
sales methodology is an important intellectual property of each company
technological founders are the best to run technological companies

Communication

sharing setbacks & issues with broader team empowers them to contribute to fixing what's most worth working on right
in any human interaction - the amount of required communication is inversely proportional to the level of trust - the ability to build this trust is the difference between an organization that works well and the one that is slow ^hthat-trust-comm
the more brains working on the hard problem, the better - "Given enough eyeballs, all bugs are shallow"
as companies grow bigger, their biggest problem is efficient communication
CEO's most important operational goal is to implement communication architecture in the company ^hthat-comm-architecture
one on ones are the best mechanism how to propagate information higher up in the company ^hthat-one-on-ones
one on ones are employee's meeting and the speaking ratio should be ~90:10 employee:manager

Organization design & scaling

running a large organization requires a different skillset than building a company
The Peter principle - in hierarchical organizations people are promoted as long as they are competent at what they do until at some moment they get promoted to a place where they are no longer competent and that place can't be predicted
promotion process ideally reminds of the very best karate dojos (e.g. getting a black belt from brown means winning a fight over black belt to make sure you are on par with black belt) - sadly there is no clear analogue to this in business
when companies grow, things that were easier at the beginning become hard, namely these three:
1. communication
2. decision making
3. common knowledge
when scaling organizations - the challenge is to grow while limiting degradation compared to the team of one "that knows everything about everything"
when scaling organizations - it is important to be able to hand over responsibilities and processes bit by bit - only that's the way how to really scale but it means losing ground slowly by transferring the ownership to someone else
the first scaling technique to implement is specialization to combat the jack-of-all-trades
the first rule of organizational design is that all organizational designs are bad
if you want people to communicate - the best way is to make them report to the same manager
when company grows big - it must make a decision whether to organize by a function (dev, sales, ...) or a mission (client-specific, cross-functional teams, ...) and the most important decision is to choose the least evil of those
six steps plan for org design:
1. figure out what needs to be communicated
2. figure out what needs to be decided regularly
3. prioritize the most important communication and decision paths
4. decide who's going to run each group
5. identify flaws & implications in previous steps
6. figure out a plan to mitigate those
the purpose of process is communication or even better - is a well-structured communication vehicle
who should design processes? those that do those activities in ad hoc manner
it's much easier to add new people to old processes rather than new processes to old people
during org design it's good to anticipate growth but it's bad to overanticipate it
there's no such thing as a great executive - there is only a great executive for a specific company at a specific point in time

Strategy, vision & the company story

when people are buying (not necessarily your product) - it's not a good time for pivot
in good organizations, people believe that if they do their job well it's going to translate for the good for them and for the organization - this is what makes working for such organizations motivating and fulfilling
Andy Grove: "The right kind of ambition is the one that helps the company succeed with their vision with the employee's personal success coming only as a by-product of this victory" (on employee ambition types)
the right kind of executive motivation is to optimize for organization's success rather than a personal one otherwise it will demotivate their subordinates as great company mission motivates more than helping my boss having a good career
in good companies the strategy and the story are the same thing thus every strategy produced there is a story and the story of a company goes beyond quarterly and yearly goals and answers around the question "why" → why should I join the company? why should I be excited to work here? why should I buy its product? why is a world better off thanks to what this company produces? ... ^hthat-strategy-story
the CEO does not need to be the creator of that vision but must be its keeper ^hthat-vision-keeper
an example of a great company story is Jeff Bezos' letter to stockholders from 1997
great CEOs create exceptional strategies that allow them to continuously build internal intelligence to support good decision making ^hthat-great-ceos
the number one reason executives fail is that when they continue to do their old job instead of moving to doing their new job especially when context around is changing

Culture, politics & feedback

in a sign of good culture - good news travel slow and bad news spread fast → what helps is to encourage - not punish - people to put problems in the open ^hthat-good-news
being a good company does not matter when things go well but matters big time when things don't go so well which happens at some point for sure ^hthat-good-company
being a good company is an end in itself and Bill Campbell is good at that
the employee lifecycle - from hire to retire
a great HR department should help to support, measure and help improve your management team
head of HR should be similarly to the head of QA, master of process design
sometimes an organization does not need a decision it needs clarity e.g. in cases of defining org values
political behavior almost always starts at the top from the CEO ^hthat-politics-top
org politics = people advancing their agendas and careers by other means than contribution and merits
build strong processes around possibly political topics like compensation schemes or personal development
keys for successful feedback provision: ^hthat-feedback-keys
1. be authentic from your core and believe fully in what you're saying
2. come from the right place and focus on setting up the receiver for success
3. don't get personal
4. don't clown people in front of their peers
5. feedback is not a one-size-fits-all and different people need different approaches
6. be direct and specific but not mean
7. think of a feedback as a dialogue instead of a monologue

Mindset, wartime & entrepreneurship

the struggle is where greatness comes from ^hthat-struggle-greatness
spend less time on what you could have done and focus that time and attention on what you might do now as there is something you can still influence ^hthat-focus-influence
there's no inertia that is putting company to motion - now should kickstart eight to ten initiatives when creating a startup
WFIO - we're fucked - it's over
running a company is like driving a car in 300 km/h in a turn - if you focus on a wall, you crash; if you focus on a road, you carry on ^hthat-driving
peacetime/wartime - in peacetime a company focuses on extending its customer base and reinforcing the company's strengths; in wartime a company is typically fending off existential crisis
hard things are hard because there are no easy answers and recipes
most important entrepreneurship lesson - embrace the struggle & extra on top of that embrace your background, weirdness and instinct ^hthat-embrace-struggle