How to Start a Startup – Ready to Grow Breakout Session

Alex Schultz & Naomi Gleit

  • You need to grow fast. Moving fast is the core to what has worked for Facebook and many other companies
  • Growth Funnel
    • Acquisition- important to get this, but worthless without the other three
    • Activation- how many people have used the site in the last 30 days?
    • Engagement
      • Engaged users are what drives your virality
    • Virality
  • 1,2,3 approach
    • Understand
      • you need to understand your product and issues that confront your users
        • Hive (analytics site)
        • Facebook analytics
    • Identify
      • A good understanding of your product is nothing if you cannot derive actionable insights.
      • Identify the opportunities in the data
    • Execute quickly
      • Based on what you identify, you should be using rapid prototyping based on your hypothesis
  • Acquisition, Activation, Engagement, Virality
    • Acquisition- Registration
      • Facebook spent an entire month on growth analysis. Did a registration waterfall analysis: how many people came in, how many people went out, how many did we lose along the way. How many registrations
      • Realized they needed to simplify the conversion rate to make it simpler and less intimidating (removed all fields that weren’t completely critical
        • Put items into the ‘next page’
      • Facebook Connect can be used to optimize this as a company
      • They are able to support 80 languages since Facebook users were the ones that translated the site. Example- French was translated in 24 hours. They had to invest in the infrastructure to allow users to do the translating.
    • Activation – Facebook had 3 steps
      • Find friends
        • This is critical to the functionality of Facebook. FB needs to find friends for the user immediately or else the user will not see the value in it.
      • Fill out info
      • Add a picture
    • Engagement-Getting users to come back to engage and produce content
      • Send to mail transfer agent ->open, click, conversion
      • Optimize your emails – FB was able to improve click through rate by switching from text to HTML with a picture of your friend that requested you
      • Go beyond the letter of the law for spam laws
    • Virality – Invites
      • Focus on retention
        • The exponential growth curves are all about retained users, the ‘new users’ are a very small portion of that growth curve
          • ‘Don’t pour water into a leaky bucket’
  • You need to figure out what matters…


Lecture 6: How to Start a Startup – Growth

Presenter – Adam Schultz

  • Retention is the biggest key to growth. Look at a retention curve of % monthly active vs days from acquisition. If your Y axis ends up parallel to the X axis metric, you are able to
  • What % of your users on day 31 were monthly active?
  • Ask: does this curve flatten out? If it does not, focus on product/market fit.
    • People usually don’t have product/market fit when they think they do
  • Different verticals have different retention rate requirements
    • Social media sites should be high: 70-80%
    • More specialized could be 20-30%
  • Startups should not have growth teams
    • It should be the whole team
    • CEO should set the direction for where you need to go
      • Mark Zuckerberg used “monthly active users” vs total registered
      • AirBnB – Nights booked
      • Yam – Messages sent
      • The second you have more than one person controlling things, you lose direction on what you should be doing
      • Each company should have its own single ‘North Star’
  • How do you drive towards the ‘magic moment’
    • There are two videos linked in the lecture notes
      • Alex Schultz and Naomi talking about it
      • Danny Farante on growth
  • When we think about growth, the biggest mistake is that we optimize for ourself
    • “I am getting too many notifications, we should optimize notifications”
      • Instead you should focus on the marginal users. Example is Facebook users that don’t have very many friends – making it easier for them to find their friends.
      • Don’t think about yourself, your issues. You are a power user, you should worry about keeping the marginal user than someone who is already firmly in the product.
  • Tactic
    • “If we build it, they will come” is an easy trap to fall into, but the worst trap to fall into.
    • Facebook was late to expanding internationally, but they were able to scale quickly because they were able to focus on the languages.
      • Built a scalable translation infrastructure that allowed Facebook users to translate themselves
    • More tactics
      • Virality –
        • Viral loop – Adam Penenberg  is a good book
        • Payload – how many people can you hit in one blast
        • Frequency – how many times you can hit them
        • Conversion rate – %
        • Virality example is Hotmail
          • “link at bottom of email saying ‘sign up for free email here’. Individual emails didn’t have much payload, but they had very high frequency (multiple emails a day) and high conversion since people were stuck paying their ISP for email service
        • Paypal example
          • Ebay was their vehicle for viral growth- conversion rate was extremely high with ‘sign up for PayPal and you will get $10’ but their payload and frequency were very low
        • You should go through this scenario in your business. Look at how your business drives the three legs of virality.
      • K-Factor – multiply out all of the % below
        • Import
        • Send
        • How many people? 100 people
        • How many click? 10% = 10
        • How many signup? 50% signup = 5
        • 10-20% import, which means your K-Factor is between .5-1.0 and if it is less than 1 you will not be viral
        • Don’t bother with this until you have a large retention to begin with
      • SEO
        • Keyword research – do your research first before you set your keywords
          • How valuable is it for you
          • How many people are searching for it
          • How many people are supplying it (competition)
          • Google Adwords
          • Get valuable links from popular sites
      • ESPN
        • Email is dead for people under 25
        • Be a high class citizen
        • You care about email rate and click-through rate
        • The most effective email is a notification
          • For Facebook, they only sent notifications for marginal users. Turned it off when people got a lot more friends so it didn’t become spam
      • SEM
      • Affiliates/Referral programs
  • “A good plan violently executed today is much better than a perfect plan executed next week” George S. Patton



Lecture 5 – How to Start a Startup: Competition is for Losers

Presenter – Peter Thiel

Capturing Value

  • A business creates X dollars of value and captures Y% of X. X and are independent variables
    • The key here is that X and Y are completely dependent. You need to capture some of the value you have created.
  • A big piece of  a small pie
  • Perfect competition
    • Pros
      • Easy to model
      • efficient in a static world
      • Politically salable
    • Cons
      • Psychologically unhealthy
      • Irrelevant in a dynamic world
      • preempts question of value
  • Monopoly
    • Pros
      • incentive to innovate
      • stable, long-term planing
      • deeper project financing
      • symptomatic of creation
    • Cons
      • lower output, higher prices
      • price discrimination
      • stifle innovation
      • tying
  • Thiel argues that there is little between the perfect competition monopoly businesses.

The Lies People Tell (for more see Zero to One)

  • Monopolies pretend not to have monopolies, because they do not want to get hit from antitrust laws
    • “We’re in a huge market”
  • Perfectly competitive companies want to pretend that they have something unique about them.
    • “We’re in a narrow market”
  • There are very powerful incentives to not be seen as a monopoly

How To Build a Monopoly

  • The right size
    • Start small and monopolize
      • It is easier to dominate a small market than a big one
      • DO NOT go after a large market from day 1
      • A lot of these companies start with markets so small, you would’t realize it was even significant to look at.
      • Grow the small market concentrically
    • Clean tech bubble
      • Started in a huge market – minnow in a vast market and got killed.

Last Mover Advantage

  • Characteristics of a monopoly
    • Proprietary technology
      • you want a technology that is an order of magnitude better than the alternative
        • PayPal could clear checks more than 10X as fast
    • network effects
      • Difficult to get started, but need to make it valuable to the first person to do something (think Facebook)
    • economies of scale
      • High fixed costs, low variable costs – easy to scale
      • Software has incredible economies of scale
    • branding
  • It is not enough to have a monopoly for the moment- it has to last the test of time: MSFT, Google, Facebook, etc.
  • Realize that most of the value exists far into the future – PayPal had about 2/3 of the value coming from about 10 years out – even at a 30% discount.
  • One thing we overvalue is growth rates, and undervalue the durability. Growth rate is much more quantitative, while durability is much more qualitative. Put the time dimension to all of the elements of the monopoly
    • be able to explain why yours will be the last breakthrough for a long time, or why you will be able to innovate further

History of Innovation

  • Historically, scientists never make any money since they do not capture the value
    • Einstein
    • Wright Brothers
    • etc.
  • The first industrial revolution was steam power, textiles, etc. 60-70 years of tremendous improvement, but the competition prevented people from making any money in these areas.
  • Where people did make money
    • Vertically integrated complex monopolies – capital intensive, complicated,
      • Standard Oil
      • Ford
      • SpaceX is a new version of this (without a 10X improvement)
    • Software
      • Economies of scale
      • Low marginal costs
      • The world of bits vs the world of atoms which can create a fast adoption rate

Psychology of Competition

  • It is an intellectual and psychological blindspot that we want to compete.
    • See competition is a form of validation
  • When the objective differences are small, you need to compete more to stand out. This is true for highly competitive things.
  • Don’t let your identity get caught up in these competitions. Competing will make you better, but make sure what you are competing for is what is truly important.
    • “Don’t go through the tiny door that everyone is trying to rush through, go around the corner and go through the vast gate that nobody is taking”


  • Don’t wait long enough to figure out what your doing is something people want. You will be at risk of losing any lead you had
  • The tendency to see competition as validation is very deep, but the starting point is to never underestimate how much you are effected by it.


How to Start a Startup: Lecture 4 – Building Product, Talking to users, and Growing

Presenter – Adora Cheung

  • The novice approach:
    • Build product in secret
    • Have an exclusive press launch (sidenote – you should definitely launch on Techcrunch)
    • Wait for users
    • Buy users
    • Give up
  • Problem statement:
    • What is it
    • How does it relate to you – it should be a problem you have
    • Verify others have it too
  • Where to start
    • Learn a lot, become an expert
      • Learn the industry (Learn how to professionally clean, become a waiter, etc.)
      • Find all potential competitor intelligence/industry expert
      • People should trust you when you are building the product
    • Identify customer segments
      • You want to corner off a portion of the customer base- focus on their specific needs.
    • Storyboard ideal user experience: what is the perfect user experience?
      • How the customer finds out about you
      • When they get the service what is that expectation
      • After the service, comments/review
  • What is V1?
    • Minimal viable product
      • The smallest feature-set required to solve the problem you are trying to solve. You need to talk to users, follow the ideal storyboard
      • Be able to clearly tell the users what they are going to get out of it
    • Simple product positioning
      • “clean home for $20 per hour”
  • First few users
    • You
    • Parents
    • Friends, co-workers
    • Local communities
    • Online communities
    • Niche influences (mommy bloggers)
    • Cold calls + emails
    • Press
    • What HomeJoy did was went to a fair and got people to sign up by giving out free cold water bottles on a hot day. Kind of a guilt trip, but most of them kept their appointments.
  • Customer feedback
    • is very important
    • Surveys: OK// interviews are better- make it into a conversation, make them feel comfortable (take users out for drinks, coffee, etc.)
    • Quantitative: retention, ratings, NPS (how likely are they to recommend to a friend?). Know if you are trending in the right direction and need to change things.
    • Qualitative: ask why
    • Beware of honesty curve – if it is free, your friends are a pretty good source for information and random people really don’t care enough. If they had to pay for it, your mom will not tell you the truth of whether your product sucks, your friends will be a lot more helpful, but random people will be the most brutally honest.
    • Use leading indicators
  • V1 Feature creep
    • Build fast, but optimize for now
    • manual before automation (things that don’t scale). You will learn a lot about what the manual things are and better how to scale
    • Temporary brokenness > permanent paralysis
    • Beware of Frankenstein – don’t implement all of the features the day after talking to the users. If someone tells you to build a feature, get to the bottom of why they want to build the feature. See it as them telling you what the problems are instead of their features. They usually will not have the best ideas.
  • S is for stealth, and stupid
    • Someone will steal your idea
    • There is a first mover advantage
    • Just launch it already. Don’t delay your launch because you are trying to make it perfect. There is really no point to waiting to launch your product.
  • Ready for a lot of users?
    • Learn one channel at a time – execute on one channel really well until it caps out.
    • Iterate working channels – Channels are dynamic: Facebook ads, Google ads, etc. should be iterated to optimized.
    • Revisit failed channels – eventually go back to the channels that failed at first
    • The key is creativity – you need to find the little thing that nobody is doing and do that to the extreme
  • Types of growth
    • Sticky growth – getting your existing users to keep buying stuff.
      • Good experience wins
      • clv + retention cohort analysis
        • aka ltv: how much revenue a customer will give you over 12 months
      • repeat users buy more and more
      • Graphic used to show sticky growth and how you want to improve the retention curveIMG_4570
    • Viral growth
      • WOW experience + good referral programs (i.e. PayPal with referral bonus, clymb, etc.)
        • Customer touch points
          • leave behind
          • after signing up
          • after using the service
        • Program mechanics
          • $10 for you, $10 for a friend
        • Referral conversion flow
    • Paid growth: risk putting money out there  in order to get
      • if you are doing paid growth, you should know if your clv>can. Know conversion rates and determine whether they are good or not.
        • sem
        • display ads
        • direct mailer
      • Beware of cash flow!!!!
  • The Art of Pivoting
    • Bad growth
      • if you are still early and not seeing any growth for several weeks, you need to rethink what you are doing since something NEEDS to change
    • Bad retention
    • Bad economics



How to Start a Startup: Lecture 3 – Counterintuitive Parts of Startups, and How to Have Ideas


Instructor – Paul Graham

Startups are counter-intuitive. You cannot always trust your instinct. What you don’t need is expertise in startups. You don’t need to know all of the steps of a startup, raising capital, etc.

  • You should trust your instincts about people though. If something seems off about someone, but smart it, trust your instincts and don’t
  • Stop looking for tricks in startups – it is more important to make something that people love. Don’t try to ‘game’ the system.
  • If you start a startup, it will take over your life for more time than you can imagine.
    • It is like having kids, it can change your life irrevocably
  • What you need to know
    • The needs of your own users
    • Do not start a startup in college
      • 20 is not the optimal time to do this
      • Success takes a lot of the serendipity out of life
      • You are more likely to succeed later on
    • Should you do it at any age?
      • Is it too hard? It is hard to tell
      • Starting a startup is unlike anything you have ever done. If you are merely unsure of whether or not you want to do it the only way to find out is to try.
  • The way to get startup ideas is not to try to make startup ideas (you will think of ideas that are bad and plausible sounding)
    • What you should do is take a step back and try to set it into the unconscious level: side projects which normally get rejected as hobbies/toys/etc.
      • Work on things that interest you
      • Learn a lot about things that matter (doesn’t have to be technology)
      • With people that you respect (to get co-founders at the same time)
    • How do you know?
      • Real problems are interesting
      • Gratifying your interest energetically is the best way to set yourself up for finding these ideas
      • If you think of technology as a fractal stain, every point on the leading edge represents a problem
        •  Get yourself to the leading edge of technology
        • Ideas will seem obvious to you
      • The best thing to do in college is LEARN POWERFUL THINGS. If you have curiosity, you will follow those inclinations. You need to have domain expertise: Larry Page was the expert on search engines.
        • Summarized: Just Learn


How can a non-technical founder most contribute to a startup? A. if it is a purely technical startup, the co-founder should focus on sales. If it is a hybrid like Uber, they should focus on recruiting drivers or something like that while the programmer works on the app.

Is there any value in business school if you are interested in entrepreneurship? – No. you need to develop products, but frankly the best way to learn things is just to do them. Business school was made to train the officer corp for the corporate world.

Managing the first hires? Ideally you are successful before you make the first hires. When you do, the first hires should be as close to founders you can get. You shouldn’t have to manage them much.

What would you learn in college right now? Physics – because it is interesting to him. You should do things that are interesting to you.

What do you do to make yourself efficient? He gets things done 2 ways: doing essays comes naturally and he gets excited about it. Setting hard deadlines also helps a lot in his experience with Y Combinator and especially on things that you don’t want to do.

When is a good time to turn a side project into a startup?  You will know when it takes up an alarming portion of your life.

When a startup is growing slowing but not much? Read “do things that don’t scale” essay

What kind of startups should not go through YC? Not many – a lot of the problems startups have are the same

If you hire people you like you can create a mono-culture, what do you do with blind spots?  The benefits of hiring people you are friends with greatly outweighs any downside of hiring a mono-culture workforce. There will be much greater problems that arise and you will have to deal with them.



How to Get Startup Ideas

This is a summary of an article by Paul Graham about generating potential startup ideas. Of all of the things I have read about the topic this article is by far the best.

Don’t sit around and try to think of startup ideas, what you really want to do is look for problems – preferably problems you have yourself because it will give you the passion you need to dedicate a large portion of your life to this problem. The very best ideas have  3 things him common:

  1. They are things the founders themselves want
  2. Things the founders can build
  3. Things that few people realize are worth doing

It is so important to work on one of your own problems because it verifies that the problem really does exist. It is actually really easy to work on solving a problem that nobody actually has – especially when founders are starting trying to think of startup ideas instead of problems. The thing that is especially dangerous about this is that it is often these bad ideas that aren’t ‘horrible’ but sound good enough for you to start working on them. At YC they call these “sitcom ideas” or an idea they would make up on a TV show- like a social network specifically for pet owners. It sounds like an idea with a lot of potential, and it might not be a terrible idea until you start digging into it and realize nobody wants to pay for, there is  little that is better than other social network sources, etc. The real danger here is that you will see is your friends will not say “I would never use this” instead they would say “yeah, maybe I could see myself using something like that.” Don’t imagine people wanting to use it, find people that REALLY DO want to use it.

Need. When you are launching a startup, you need at least some users that really need what your are making and want it urgently. You will realize you need to compromise in one direction: build something everyone will like a little, or build something a small group will love.  You need to choose the latter because a lot of people would be mildly interested in a social network for pet owners; but it will not gain any traction (especially in the early phase) until there is a core group that loves the product. When you do come up with a startup idea, ask yourself: who wants this right now- even if it is a crappy version? The first version will be just that – crappy, but if you cannot answer that question your idea is probably not good enough. It is usually a good sign if your idea will appeal strongly to a specific group of people. While you want to start in a small segment of a market, you want to have a fast path out of it (without much additional product change). In Facebook’s example, you had Harvard, then Ivy league schools, then colleges in general – now it dominates the web. Facebook had a fast path out of its initial small market. The problem here is that you probably won’t know right away if your idea will have a fast path out, so what you need to focus on is just making the product great. 

Self. Graham gets a bit philosophical quoting a passage from Zen and the Art of Motorcycle Maintenance (page 3 of the essay) but the point is that the way to have good startup ideas is to become the sort of person that has them. One great piece of advice is to go to the front lines of emerging technologies and immerse yourself in the problems (IoT, Solar, VR, etc.) you can be at the leading edge as a user (not just an innovator) and you will see the challenges/opportunities that rise up from the next wave. In this context, he combines two quotes to come up with: “Live in the future, then build what’s missing” Most successful startups are the result of some external stimulus against a prepared mind that was ready to receive it. You need to be prepared to notice the things that are inconvenient or can be done better and NOTICE the opportunities they represent.  The key word in that sentence was the word NOTICE, founders don’t “think up” startup  ideas, they NOTICE them. If you are not at the leading edge of some rapidly changing field you can get to one pretty easily. Graham argues you could get to the “edge” of programming  (building mobile apps) in a year. A year really isn’t a bad investment if you consider you will be spending at least 3-5 years of your life at a startup. It also gives you the ability to put things into code – rapid prototyping. So when you do have a an idea you can say “that’s an interesting idea, I will try building an initial version tonight”

Noticing. Once you are living at the edge of the future, you need to start turning off filters. Not only the “whats missing” filter, but also the “could this be a company” filter. These filters could filter out a lot of good ideas that might not look viable at the surface level – and it can also cause you to focus on the bad ideas. The other filter you need to turn off is the “taking things for granted” filter. Our minds automatically filter the things we don’t need to notice just to keep us sane and on task. Taking this in another direction, you should also focus on things that chafe you. When you take the status quo for granted, you not only make life more efficient, but it also makes life more tolerable. “When something annoys you it could be because you’re living in the future.” When you do find the right sort of problem, you should probably be able to describe it as obvious- at least to you. Don’t sit around and try to think of ideas – instead work on hard problems that are driven mainly by curiosity, and take note of gaps and anomalies. Another good way to trick yourself into noticing ideas is to work on projects that seem like they’d be cool. – you will naturally tend to build the things that are missing. Graham gives an altered version of his quote above: “live in the future and build what seems interesting”

School. Entrepreneurship is something best learned by doing it. When you are in college you should be spending your time learning the waves of the future. It is also good to study multiple domains: if you start in programming , then learn another field like accounting, physics, marketing, etc. you will see the opportunities you can help solve in those fields since you have tools in your programming toolbox.

Competition. Don’t be afraid of the competition – it is very rare fro startups to be killed by the competitors unless they have some lock-in feature that keeps customers there. You should be able to do some quick Google search to see what the competition is: who is in it, what are their capabilities compared to what you are trying to do, etc. It is common for inexperienced founder to give their competitors more credit than they deserve. You don’t need to worry about a crowded market, you DO need to know your competition, and what they are missing. A crowded market usually means that nobody has come up with a good enough solution to dominate the market. There are 2 ways to break into market: Enter small market, monopolize, expand; Enter crowded market with a secret weapon.

More Filters. There are two more filters you need to turn off when thinking of ideas: the unsexy filter and the schlep filter.

  • Schlep filter: do things other people don’t want to do – the pain in the ass or tedious projects. They are usually not as bad as they seem (Stripe example of not wanting to deal with payments)
  • Unsexy filter: keeps you from working on the problems you despise rather than the ones you fear. The unsexy filter is not completely worthless since if you ware going to be at the leading edge of a field that is rapidly changing your ideas about what is sexy will usually translate to what is valuable. Also, if an idea is sexy, you will be more likely to work on it more enthusiastically.

Recipes. It is much better to ‘notice’ ideas organically, but if you really need a recipe, you need to be able to filter out the bad ideas much better. Here are some tricks/tips:

  • Look where you have some expertise. Your expertise raises your standards.
  • Look at the things you need
  • In your previous job, did you ever find yourself saying “why doesn’t someone make x?”
  • It is easier for people in their teens and early twenties because they understand their peers and up and coming technology.
  • Try finding gaps in the world: what is tedious or annoying?
  • Talk to people about their work – let it get general, try to get something to spark a thought
  • Seek out ideas that are unsexy or involve schleps
    • Make something unsexy that people will pay you for
  • Replacement tech usually comes in from the side: don’t look for the replacement of X, look for what is there and will later become the replacement for X. (example was journalism – Traditional journalism is dying, what replaces it will probably come from within the industry but will not look like journalism in its form today)
  • Think about what new things will be possible in the new world we’ll have in a few years. Stay on the edge of the future technologies to have a close pulse

Organic. If you are on the edge of some rapidly changing field you don’t have to look for the wave; you are the wave. If you have the right background, good startup ideas will seem obvious to you. Be patient, it will take time to come across situations where you notice something missing. Once again the quote: “live in the future and build what seems interesting. Strange as it is, that’s the real recipe”



How to Start a Startup: Lecture 2 – Team and Execution

Presenter: Sam Altman


  • Cofounders/cofounder relationships
    • Don’t choose a random person you don’t know
    • It should be a person you have a history with (preferably for years)
    • It is better to be a solo founder than have a bad cofounder (and they only founded 1 in 10 solo teams)
    • Relentlessly resourceful: Paul Graham coined the model of James Bond
      • You need someone who behaves like James Bond more than you need someone who is an expert in some particular domain
    • Your cofounder should be
      • Tough
      • Calm
      • Technical
    • 2 or 3 total cofounders (5 is too many)
  • Try not to hire
    • Too many employees creates a high burn rate of cash
    • How much can you get done with a small # of employees – try to stay small as long as you can.
    • Goal should be not to hire
    • AirBnB took 5 months before they hired their first person
      • They had a list of several character traits and values that they believe in
      • Make sure you hire for a culture
      • The first 50 employees at AirBnB feel like they were the ones that founded the company. You want to create that culture/level of ownership.
  • Get the best people
    • You need to convince them that your mission is the one they should be fighting for.
    • When you are hiring it should either be 0% of your time or 25% of your time.
    • Mediocre engineers do not build great companies – Steve Jobs on B-Players
      • You can get away with it in a big company, because they will fall through the cracks. In the early days you cannot screw this up.
    • Sources of candidates: people you already know and personal referrals. Typically for the first 100 employees.
      • Personal referrals are the trick to hiring
    • Experience – matters for some roles and not others
      • A lot of early roles, experience doesn’t matter
      • Large scope management roles, etc. does require prior experience
    • 3 things to ask yourself when interviewing- if you get a yes to all three, you will never regret the hire:
      • Are they smart?
      • Do they get things done?
      • Do I want to spend a lot of time around them?
    • Call references and ask them questions like:
      • What specific projects did they work on
      • Are they in the top 5% of people you have ever hired?
      • Why aren’t your trying to hire them back?
    • More points for hiring
      • Good communication skills
      • Manically determined/should be comfortable with risk
      • Pass the animal test (Paul Graham)
        • you should be able to describe any person as an animal at what they do
        • You need unstoppable people
      • Would feel comfortable reporting to them if the roles were reversed
    • Employee Equity
      • You should aim to give 10% of the company to the first ten employees
        • If they are successful, they will influence the company way more than that
        • Normally founders give too much equity to investors and not enough to employees. Employees are the ones that actually build the company over the years.
      • The YC companies that have been generous with equity to employees typically are the most successful.
  • You’ve hired the best – now keep them around
    • Learning a little bit of management skills goes a long way
    • It is not always natural for the founder to give away praise
    • You need to learn to give your team credit for everything that happens and own all of the bad stuff.
    • Realize you are not going to be a very good manager and try to over-compensate for that
    • 3 things that motivate people to do great work (Dan Pink)
      • Autonomy
      • Mastery
      • Purpose
    • Focus on-one-on ones and clear feedback
  • Fire Fast
    • One of the worst parts of running a company
    • It is better for the company and the employee to fire fast
    • Fire if:
      • People are bad at their job
      • Are persistently negative
      • Create office politics
        • These last two are completely toxic to a company
    • If someone is consistently doing things wrong, they need to go. In practice, there is often never any doubt about this.
    • When should cofounders decide on the equity split?
      • As quickly as possible
      • It should be pretty equal
      • If it is not equal, you need to think hard about whether you want them in your company
    • Cofounder vesting is a must- The normal stance is that it takes 4 years to vest cofounder equity (linear scaling of 25% per year). If you don’t the other founder will be dead weight.
      • You need to talk things through early to try to avoid that
    • Really good people can almost always find a place in the company. Even if their role gets a bit out of the initial scope.
    • It is better to lose some initial customers than to hire the wrong people
    • Co-founders in different locations – don’t do it


  • Everything in the startup becomes modeled after the founders
    • Hard working
    • Execution machine
    • Driven
  • CEO has five jobs
    • Set the vision
    • Raise money
    • Evangelize the company
    • Hire and manage
    • Making sure the entire company executes (setting the execution bar)
  • Can you get it done?
    • Focus
      • What are you spending your time and money on?
      • What are the two or three most important things?
        • Say no. A lot
        • Set overarching goals. Repeat them
        • Communicate – you cannot be focused without good communication
      • You have to work really hard on the RIGHT things
      • Everyone in the company should know the key goals
      • Maintain growth and momentum
        • Don’t get too excited about PR – you need to have a real product to live up to any PR hype
      • Work together in person
    • Intensity
      • Startups only work at a very intense level
        • The secret to startup success is extreme focus and extreme dedication
        • Outwork your competitors – even by a little bit to be successful
      • Relentless operating rhythm
        • “move fast and break things”
      • Obsession with execution quality
      • Bias toward action – you cannot be indecisive.
        • Every time you talk to the founders, they should have gotten things done
        • You need to break the project down into smaller projects
      • The best founders:
        • Quick -email response
        • Do whatever it takes
        • show up – “get on planes in marginal situations”
        • Don’t give up
        • Be courageous
      • Always keep momentum – you want your team to be winning all the time. A winning team feels good and keeps winning, a losing team gets demotivated and keeps losing.
        • If you are not winning, “sales fix everything”
        • Where can you get the small wins
        • When you lose momentum, small fights and disagreement starts to come up. if this happens you need to talk to your users.
        • Acknowledge out loud that things aren’t working, and you need to figure out how to fix them
      • Always keep growing – If you build a good product, it will grow
      • Set an operating rhythm
        • Shipping product
        • Launching new features
        • Reviewing/reporting metrics and milestones
      • “the competitor to be feared is one who never bothers about you at all, but goes on making his own business better all the time.” Henry Ford
        • Don’t worry about the PR from competitors