The Hard Thing About Hard Things by Ben Horowitz

  • Rating: 4/5
  • Amazon
  • Colin Powell: leadership is ability to get someone to follow you even if only out of curiosity.
  • My father turned to me and said, "Son, do you know what's cheap?", "No, what?" "Flowers. Flowers are really cheap. But do you know what's expensive?", "No, what?", "Divorce." Made suddenly clear continuing course, I might lose my family. By doing everything, I would fail at the most important thing. I always thought about myself first. When you are part of a family or part of a group, that kind of thinking can get you into trouble, and I was in deep trouble.
  • "No matter what we say, we're going to get killed. As soon as we reset guidance, we'll have no credibility with investors, so we might as well take all the pain now, because nobody will believe any positivity in the forecast anyway. If you are going to eat shit, don't nibble." Dave Conte
  • Our business was still cloud business, I gave no indication to rest of staff I had other ideas. Doing so would have instantly doomed the only business we were in, as everyone would want to work on the future and not the past.
  • Product management had allergic reaction to prioritizing potentially good features above features that might beat BladeLogic. "How can we walk away from requirements that we know to be true to pursue something that we think will help?" It turns out that is exactly what product strategy is all about—figuring out the right product is the innovator's job, not the customer's job. The customer only knows what she thinks she wants based on her experience with the current product. The innovator can take into account everything that's possible, but often must go against what she knows to be true. As a result, innovation requires a combination of knowledge, skill, and courage. Sometimes only the founder has the courage to ignore the data; we were running out of time, so I had to step in: "I don't care about any of the existing requirements; I need you to reinvent the product and we need to win."
  • No answer to "the Struggle", but here are things that helped:
    • Don't put it all on your shoulders. Easy to think things that bother you will upset your people more. That's not true. The opposite is true. Nobody takes losses harder than person most responsible. Nobody feels it more than you. You won't be able to share every burden, but share every burden you can. Get maximum number of brains on problems even if problems represent existential threats. At Opsware, losing competitive deals, called all hands, told company we were getting our asses kicked and if we didn't stop bleeding, we were going to die. Nobody blinked. Team rallied, built winning product, and saved my sorry ass.
    • It's not checkers; it's chess. Technology businesses tend to be extremely complex. Underlying technology, competition, market, and people move. There is always a move. You think you have no moves? How about taking your company public with 2 million in trailing revenue and 340 employees, with a plan to do \75 million in revenue the next year? I made that move in 2001, widely regarded as worst time ever for technology company to go public with six weeks of cash left. There is always a move.
    • Play long enough and you might get lucky. In tech, tomorrow looks nothing like today. If you survive long enough to see tomorrow, it may bring you answer that seems so impossible today.
    • Don't take it personally. Probably your fault. You hired the people, made the decisions. Everybody makes mistakes. Evaluating yourself and giving yourself an F doesn't help.
    • Remember this is what separates women from girls. If you want to be great, this is the challenge. If you don't want to be great, then you never should have started company.
    1. Trust. Without trust, communication breaks. More specifically: In any human interaction, required amount of communication is inversely proportional to level of trust. If I trust you completely, I require no explanation because I know what you're doing is in my best interests.
    2. More brains working on hard problems, the better. Total waste to have lots of big brains but not let them work on biggest problems. A brain cannot solve problem it doesn't know about. "Given enough eyeballs, all bugs are shallow."
    3. Good culture is like old RIP routing protocol: Bad news travels fast; good news travels slow. Failed companies, many employees knew about fatal issues. Company culture can discouraged spread of bad news. Healthy culture encourages sharing bad news. Build culture that rewards—not punishes—people for getting problems into the open where they can be solved. Beware management maxims that stop information from flowing freely. For example, "Don't bring me a problem without bringing me a solution." What if employee cannot solve important problem? Do you really want him to bury that information?
  • Lay people off right way
    1. GET YOUR HEAD RIGHT. When you must fire employees that took great time and expense to hire, weighs heavily, difficult to focus on future, because past overwhelms you—but that's exactly what you must do.
    2. DON'T DELAY so word doesn't leak.
    3. BE CLEAR IN YOUR OWN MIND ABOUT WHY YOU ARE LAYING PEOPLE OFF. Board may say, "Gives us opportunity to deal with performance issues and simplify business." May be true, but do not let that cloud thinking or message to company. You are laying people off because company failed to hit its plan. If performance were only issue, you'd be taking different measure. Distinction is critical, because message must be "Company failed and in order to move forward, have to lose excellent people."
    4. TRAIN YOUR MANAGERS. Managers must lay off their own people. People won't remember every day they worked for your company, but they will surely remember day you laid them off, reputations of your company is on line.
      • Prepare them
        1. Briefly explain what happened and that it's company rather than personal failure
        2. Be clear employee is impacted and decision is nonnegotiable
        3. They should be fully prepared…[truncated]
    5. ADDRESS THE ENTIRE COMPANY. Prior to executing layoff, CEO must address company.
    6. BE VISIBLE, BE PRESENT after speech.
  • Firing executive
    1. Be clear on reasons. Don't equivocate or sugarcoat. You owe it to them to be clear about what you think happened.
    2. Use decisive language. Do not leave discussion open-ended. Use phrases like "I have decided" rather than "I think."
    3. Have severance package approved and ready.
    4. Will be keenly interested in how news communicated to company and world, let them decide, "Ben, you cannot let him keep his job, but you absolutely can let him keep his respect."
    5. Quickly update company and staff on change. Correct order is (1) executive's direct reports (2) your staff (3) company. All should happen same day, preferably within couple of hours.
  • Humans, particularly those who build things, only listen to leading indicators of good news. For example, if CEO hears engagement for her application increased an incremental 25% beyond normal growth rate one month, she'll be off to races hiring to keep up with demand. If engagement decreases 25%, equally intense and urgent in explaining it away: "The site was slow that month, there were four holidays, made UI change that caused problems."
  • In competition with BladeLogic, many of smartest people came to me with ideas for avoiding battle: "Let's build lightweight version of product and go down-market.", "Let's acquire company with simpler architecture.", "Let's focus on service providers." All these approaches reinforced to me was that we weren't facing market problem. The customers were buying; they just weren't buying our product. Was not time to pivot. So I said, "There are no silver bullets for this, only lead bullets." We had to build a better product. There was no other way out. Lead bullets.
  • There comes a time in every company's life where it must fight for its life. If you find yourself running when you should be fighting, you need to ask yourself, "If our company isn't good enough to win, then do we need to exist at all?"
  • Parcells: "Al, I am just not sure how we can win without so many of our best players. What should I do?" Davis: "Bill, nobody cares, just coach your team." That might be the best CEO advice ever. Because, you see, nobody cares. When things go wrong in your company, nobody cares. The media don't care, your investors don't care, your board doesn't care, your employees don't care, and even your mama doesn't care.
  • In enterprise software companies, the two most important positions tend to be VP of sales and VP of engineering.
  • Management training is the best place to start setting expectations for your management team. Do you expect them to hold regular one-on-one meetings with their employees? Do you expect them to give performance feedback? Do you expect them to train their people? Do you expect them to agree on objectives with their team? If you do, then you'd better tell them, because the management state of the art in technology companies is extremely poor. Once you've set expectations, the next set of management courses has already been defined; they are the courses that teach your managers how to do the things you expect (how to write a performance review or how to conduct a one-on-one). Keep in mind that there is no investment that you can make that will do more to improve productivity in your company. Therefore, being too busy to train is the moral equivalent of being too hungry to eat. Furthermore, it's not that hard to create basic training courses.
  • Onboarding: require them to bring comprehensive set of questions about everything they heard that day but did not completely understand. Answer those questions in depth; start with first principles. Bring them up to speed fast. If they don't have any questions, consider firing them.
  • Consensus decisions about executives almost always sway the process away from strength and toward lack of weakness.
  • Like technical debt, management debt comes in too many different forms to elaborate entirely, but a few salient examples: 1. Putting two in the box 2. Overcompensating a key employee, because she gets another job offer 3. No performance management or employee feedback process.
  • Sometimes an organization doesn't need a solution; it just needs clarity. Once I made it clear that cursing was okay—so long as it wasn't used to intimidate or harass—nobody had a problem with it anymore.
  • Right kind of ambition is ambition for company's success with own success only coming as by-product. Wrong kind is ambition for personal success regardless of the company's outcome.
  • Peter Principle holds that in hierarchy, members are promoted so long as they work competently. Sooner or later they are promoted to position at which they are no longer competent (their "level of incompetence"), and remain there, being unable to earn further promotions. It's unavoidable, no way to know a priori at what level manager incompetent. Another challenge is Law of Crappy People: For any title level in large org, talent eventually converge to crappiest person with title. Employees with lower titles naturally benchmark against crappiest person at next level, demand promotions as soon as they reach his low level of competency.
  • One-on-ones: employee sets agenda sent in advance. Gives them chance to cancel meeting if nothing pressing and makes clear it's their meeting. Manager should do 10% of talking and 90% of listening.
  • If you manage engineers, drawing out issues will be important skill to master. Some questions:
    • If we could improve in any way, how would we do it?
    • What's the number-one problem with our organization? Why?
    • What's not fun about working here?
    • Who is really kicking ass in the company? Whom do you admire?
    • If you were me, what changes would you make?
    • What don't you like about the product?
    • What's the biggest opportunity that we're missing out on?
    • What are we not doing that we should be doing?
    • Are you happy working here?
  • Primary thing any tech startup must do is build product at least ten times better at doing something than current way. Two or three isn't good enough to get people to switch. Second thing they must do is take market. If it's possible to do something ten times better, you aren't only one trying. Must take market before somebody else does.
  • Constantly evaluate all members of your team. However, evaluating people against future company needs based on theoretical view of how they will perform is counterproductive
    • Managing at scale is learned skill rather than a natural ability
    • Nearly impossible to make judgment in advance. Was it obvious Bill Gates would learn how to scale when he was Harvard dropout?
    • Act of judging people in advance will retard their development
    • Hiring scalable execs too early is bad mistake. No such thing as great executive, only great executive for specific company at specific point in time.
    • You still have to make the judgment at the actual point in time when you hit the higher level of scale.
    • Make judgment on relative rather than absolute scale. Ask yourself whether executive is great can be extremely difficult to answer. Better question: For this company at this exact point in time, does there exist executive I can hire who is better?
  • When I used to review executives, I would tell them, "You are doing a great job at your current job, but the plan says that we will have twice as many employees next year as we have right now. Therefore, you will have a new and very different job and I will have to reevaluate you on the basis of that job. If it makes you feel better, that rule goes for everyone on the team, including me." In providing this kind of direction, it's important to point out to the executive that when the company doubles in size, she has a new job. This means that doing things that made her successful in her old job will not necessarily translate to success in the new job. In fact, the number-one way that executives fail is by continuing to do their old job rather than moving on to their new job.
  • Elite at giving feedback, don't use shit sandwich
    • Be authentic. Extremely important you believe in feedback you give
    • Come from right place. Give people feedback because you want them to succeed and not because you want them to fail.
    • Don't get personal. If you decide to fire somebody, fire her. Don't prepare her to get fired. Prepare her to succeed. If she doesn't take the feedback, that's a different conversation.
    • Don't clown people in front of their peers. While it's okay to give certain kinds of feedback in a group setting, you should strive never to embarrass someone in front of their peers.
    • Feedback isn't one-size-fits-all. Everybody is different. Some employees are sensitive to it while others have particularly thick skin. Your tone should match the employee's personality, not your mood.
    • Be direct, but not mean. Don't be obtuse. If you think somebody's presentation sucks, don't say, "It's really good, but could use one more pass to tighten up the conclusion." While it may seem harsh, it's much better to say, "I couldn't follow it and I didn't understand your point and here are the reasons why." Watered-down feedback can be worse than no feedback at all because it's deceptive and confusing to the recipient. But don't beat them up or attempt to show your superiority. Doing so will defeat your purpose because when done properly, feedback is a dialogue, not a monologue.

Stay up to date

Get notified when I publish. Unsubscribe at any time.