So I watched Sam Altman’s “How to Start a Startup” in a single day

Sam Altman’s “How to Start a Startup” - MashupCorner

This article is authored by Chris Markl  who serves as a Social Entrepreneur in Residence at Florida State University. This article originally appeared on medium.com (hyper link to the story on medium)
In the fall of 2015, Sam Altman President of Y-Combinator, taught a class at Stanford entitled “How to Start a Startup”. All lectures are online.

I’m on faculty as an Entrepreneur in Residence at Florida State University. I, like other professionals in the field, struggle with finding time to consume content. So created an event open to all Florida State University students where I screened the entire semester of “How to Start a Startup” in a single day from 7am-11:50pm, 16 hours straight.

Below is a live commentary.

7:00AM Lecture 1 — How to Start a Startup

Alright let’s do this!

Helping startups is hands on and customized to a startup.

30% is pretty standard and is what is taught in this course.

Information in this course is geared towards hyper growth. Honestly who wouldn’t want hyper growth?

Four determinants of startup success: Idea, Product, Team and Execution.

1) Idea

Long term thinking is a huge advantage in startups. The best companies are almost always mission oriented. If you don’t love what you are doing, you’ll likely give up along the way. Good startups, take 10 years, not 2–3 years. Starting with a massive idea, inspires stakeholders and motivates work harder.

The best idea usually looks terrible at the beginning. You want an idea that can turn into a monopoly. Great companies start by taking over one small market and growing quickly. Investors think about the growth of company, but also think about the growth of the market. Customers are usually desperate for a solution. You can’t force a market, thus it is important to conduct some research. Good startup ideas can be explained in a sentence. Any company is a clone with a small differentiator, usually fails.

First student just walked in to join me! Its 7:30 in the morning and there is no one else in the building!

2) Product

Build products, talk to users, eat, sleep, and nothing else. A founder’s goal is to build something that users love! It’s better to build something that a small number of users love, instead of something a large number of users like. If you make something people love, they will tell their friends about it. Over the long run, great products win! Start something simple.

Talk to users, don’t use google ad words, just go talk to people around you. Get user manually and realize that the goal is to get a few users to love the product. You have to do sales and customer service yourself. Live on growth.

Dustin Moskovitz — is now talking that being a founder is incredibly difficult, stressful, and requires 24 hour focus. Being a founder isn’t glamorous. Being a CEO, you aren’t the boss, everyone else is your boss.

Second student has just arrived, actually impressed, especially because its before 8am.

I think Dustin’s honesty about entrepreneurship is important. But I also think that a normal job in an uninspiring company is perhaps decently miserable. I, like other entrepreneurs, love the startup environment, problem solving and commitment. I also Dustin thinks of flexibility wrong, while yes as an entrepreneur I can work 24 hours a day, having the flexibility to choose what I work on, I think is incredibly rewarding. I also love the idea of betting on one’s self.

So why become an entrepreneur? The world needs it and you are passionate about the idea enough and are going to do it anyways!

Lecture 2 Team & Execution

Choosing a founder who you aren’t friends with is just a bad idea. The track record for founders who don’t know each other is just not good. Solo founder is better than a random founder. Co-founders should be relentlessly resourceful, like James Bond. 2–3 co-founders is a right amount. Try not to hire, you are judged by the number of employees. It sucks to have employees, high burn rates, tension, slow decision making. The cost of getting an early higher wrong is incredibly high.

An organization, any organization, is a group of people, who create products. A startup is such a small team, thus the productivity, commitment and joy that each person brings is incredibly correlated with productivity. This is why Sam Altman is correct about getting the right people to work for one’s company.

It takes forever to find the best people. Founders consistently under-estimate how long it takes find someone as a co-founder or as a hire. Hiring should spend either 0% or 25% hiring. A mediocre hire will kill a startup. The best source to find someone to hire, is your network. When you work for Facebook in the first 2–3 weeks, a HR basically forces you to reveal each smart person you know. Sam Altman’s three determinants of hiring: smart, get things done, do I want to spend time with them. Try to work on a project instead of conducting an interview. You want people who will just get stuff done. Mark Zuckerburg, hires people he will spend time with socially and also who he would feel comfortable reporting to.

Provide early employees with large amounts of equity. The first 10 employees should receive 10% and they have to earn it over the first four years. Be as generous with equity to employees, and be as stingy as possible with investors. Retaining employees is incredibly important and difficult. Three things that provide an employee with pleasure and meaning: autonomy, mastery and purpose. Fire fast, its best for everyone. Fire people who are bad at their job, create office politics, and are persistently negative.

Co-founder split should be near equal. If there is a non-equal split it is likely a reason to step back and think about this relationship. Vesting of four years is the industry norm, with a one year cliff.

– Execution

Sam Altman’s “How to Start a Startup” - MashupCorner

8:52am Lecture 3 Paul Graham, Founder, Y Combinator

Starting a startup is counter-intuitive. But one’s intuition about people is usually spot on. Work with people you respect and likely working with, and you have known for a while. You don’t need to be an expert in startups to be a successful founder. What you need is expertise in your users. The best way to convince an investor to invest, is to demonstrate that a startup is growing. The way to create a startup that grows, is to create a product that users love. You are unable to game the system in startups. There is no boss to trick, there are only users. Users will only judge if they want and love. You can somewhat trick investors, for one or two rounds, if you know what you are talking about, but you are really tricking yourself because your equity is likely to ultimately be worth not much when your company fails. As a company grows, the total volume of worry and concern also grows.

Opened my first mountain dew, which is a bit troubling. I really want to avoid caffeine to make it through the day. Totally disagree with Paul Graham that starting a company in college is a bad idea. There is so much learning involved in both creating a product and acquiring a customer. Learning that is very complimentary to one’s curriculum. College students have a lot of time and are less worried about finances. Infact, I can’t think of a better time to start a business other than college. Ok Paul Graham, nice way of turning this around, and focusing on domain expertise to guide one’s interest.

Learn how to unconsciously think about startup ideas, learn a lot about things that matter, work on ideas that interest you, and work with people you like and respect.

Paul Graham states the business school teaches management. If you want to be an entrepreneur, he advises individuals to learn about product. The best way to learn about product is actually just to start a company.

You will know a side project becomes a startup when it takes on an alarming amount of your time.

9:42 Lecture 4: Adora Cheung, Founder, Homejoy — Building Product, Talking to Users, and Growing

Down to 2 students here… Curious why I decided to do this, but my significant other has packed me fantastic food. While the top layer is unhealthy, the good stuff is on the bottom, including amazing home made soup!

Sam Altman’s “How to Start a Startup” - MashupCorner

Ok to the next lecture: Lame approach to startups: Build product in secret, exclusive press launch, wait for users, buy users, and give up…

I see this all the time. Basically each day, a student comes to me with this exact plan. Literally my last meeting yesterday I advise against it, sometimes an entrepreneur listens, most times they don’t.

Become an expert, identify customer segments, and storyboard ideal user experience.

To become an expert, become obsessed with your industry and competitors. Seriously read the nerdiest information relevant to your industry.

Minimal Viable Product (MVP). Must be viable. Simplicity is key.

First few users, should be you, your social network, friends, family, online and local communities, influencers, cold calls and emails, and press.

While the list above is simple, entrepreneurs sometimes skip all of these things.

Customer Feedback — have a way for customers to contact you, ideally a phone number with voice mail. Ultimately you should be getting out from behind your desk and talk with customers. Talking with a customer is not a laboratory, you should focus on getting to know the customer, take them out for drinks or coffee. For a paid product, the less a founder knows, the more honest the customer feedback. For a free product, this may not be true, for free products, where users who you don’t know might not care enough to give you honest feedback.

When building a tech product, make features manual in the beginning that will be automated over time.

This manual process reduces the technical requirement to launch a product and also facilitates learning.

Features can hide the root problem, to understand the root problem talk with customers. S is for stealth and stupid, just launch already…

Just heated up an awesome breakfast burrito!!

Viral growth experience is about an amazing 1) experience plus 2) referrals.

To create a robust referral service 1) collect customer touch points (sign up, use product, etc), 2) program mechanisms (receive $25 when a user you refers spends $25), 3) conversion mechanics (maximizing the environment to convert once a potential user is sent to the site).

Have a growth plan when you start. What is an optimistic and realistic way to grow this business? Example: Week 1 get that first user. Week 2, 2 users. Early on it should be easy to reach these goals if you created a product that consumers love. If you are failing to grow for 3–4 weeks, you need to step back and consider a pivot. Anticipate uneven growth.

10:34am — Lecture 5: Competition is For Losers: Peter Thiel ,Founder, Paypal, Palantir, and Founders Fund

Always want to aim to become a monopoly! You have to create value and capture the market. Airlines vs google. Airlines is a huge market but profit margin is so incredibly small compared to the profit margin of google.

This is why Tech companies are so attractive, because of the relatively small amount of money to scale. I think instagram was a 13 person team when it was acquired by Facebook.

If you are in a perfectly competitive industry, you will tell a lie that the industry is less competitive. Non-monopolies will describe a super small industry, as a niche “We are the only british food restaurant in Palo Alto”. Does this intersection make sense? It is incredibly hard to make money in this situation.

On the other side, the people who have monopolies, pretend they do not have a monopoly. Two lies, but there is actually massive differences in the market structure and value between these companies. Google has 66% market share but never describes it self as a search engine, but instead as an advertising company or a technology company. Amazon, Microsoft, Google, and Apple each have lots of cash 10–100b and huge gross margins.

Start with a small market. If you think the market may be too big, it most certainly is. Most tech companies dominate one market and then expand. These companies often start with markets that many people think is too small. Think about Facebook market which consists of 10,000 students at Harvard, reached 60% penetration in a few months.

You want to be the only company in a market who can lead. Larger markets equal more competitors.

Characteristics: Propriety technology (order of magnitude better than the next best thing), network effects, economies of scale (tech is great at economies of scale, as one platform can work for one user or one billion users), branding.

Its not enough to have a monopoly for a day, you want to be a monopoly over time. You want to be the last mover, not the first. Most of the value of the value in a company exists far in the future. Whenever you do the math on a tech company, 85% of cash flow values are atleast 10 years out. Ask, why will this be a leading company 10–15 years out?

I’m feeling good right now. Learning a lot. Hope this continues for the next 13 hours!

The first one in a market rarely makes money. This connects back to the previous lecture that there is no reason to be in stealth mode with one’s startup. Launch, have a go at it, acquire customers, learn, and scale.

Look at Tesla, what is impressive is that it looks like an old school monopoly, solving something incredibly complex. Compared to a software company, great product + fast adoption rate + capital efficient business can become a monopoly, e.g. airbnb.

We find ourselves driven towards competition, if we see someone else doing something we want to do it. Competition is a form of validation. But its not that there is wisdom in crowds. The fact that there are so many people doing something, it may be a sign of insanity. Why fight when the stakes are so small? When its hard to differentiate yourself from others, you have to compete, on metrics that may be more imaginary than real.

Remember it is important to reflect on what important for you.

11:30 Lecture 6: Growth: Alex Schultz, VP Growth, Facebook

Retention is a main determinant of growth.

This explains why its important to create a product that consumers love.

Focus on active users in the last 30 days. This figure will show retention rates. If retention rate is not good, need to think hard about product market fit. Different verticals need different retention rates, e-commerce 20–30% of re-occurring users is completely successful, if this was a social networking site, this rate is awful. If you are a startup, the whole team is the growth team. The CEO needs to set the north star of where the company will go. Mark Zuckerburg focused on monthly active users, this is the north star he created. Airbnb is nights booked! Ebay is growth merchandise volume. Each company needs a different north star. This north star will guide company’s decisions, especially employees.

At Facebook, focus is on helping a new user reach 10 friends in 14 days. The magic moment is when you sign up, have zero friends, and your friend’s face pops up. On ebay, the magic moment is when you find the old pez dispenser. Airbnb, the magic moment is when you find an incredible rental house, or when you host that you actually get paid. Finding this magic moment, is important because it has a significant positive impact on retention.

To create growth, focus on the marginal user.

Viral Loop by Adam L Penenberg — I’ll likely buy this book on Viral Marketing. Three components to viral marketing: payload (how many people can you hit), frequency (how many times can you hit them), and conversion rate (% they convert).

At the bottom of each email with hotmail the tagline is “Sent by Hotmail”, low payload because each email’s payload was the recipient, frequency is high, and conversion high. Paypal offered free money with low payload and frequency with high conversion rate.

SEO: human research, see what people are searching for thats related to your site, how many people are searching for this term, how valuable is this search, etc.

Email: Dead for individuals under 25. Email, SMS, and Push all act the same and need to be delivered. After delivery it is about conversion. The most important email is a notification. Don’t treat a user who signed up yesterday as the same as a person who has been a user for years. What notifications should you be sending? How can you create great trigger marketing campaigns?

12:13pm Lecture 7: How to Build Products Users Love: Kevin Hale, Founder, Wufoo and Partner, Y Combinator

Ok I’m starting to fade a bit.

Focus on emulating experiences from the real world, treat new users like you are dating, and treat current customers like you are married.

Oh Kevin Hale — employing an example from Dan Gilbert’s first TED Talk regarding picking one’s nose on a first date vs in marriage. I think what he is trying to say is that a bad early impression will dissuade a potential user from adopting or becoming a regular user. He’s talking through some examples about interesting web copy to create a unique experience. I feel like this is an important thought, but for me, I focus on clarity. Clarity in the message I want to convey in a brand. Is this what the audience is hearing? I try to get everything else out of the way. I see come across many examples where a company is trying to be cute or interesting and in reality this ‘design feature’ is actually a distraction.

Customer support occurs in each step in a customer conversion. Making everyone conduct customer support does two things: it creates the right incentives for everyone including designers and developers to understand and solve customer problems.

This talk is basically a user experience talk, that is example driven. The take away message is treat each real person and understand her goals and concerns. Connecting high level individuals at your startup with consumers empowers helps your startup to solve both of these problems. He’s also talking about the importance of evolving the user experience by increasing information clarity, creating a more engaging environment, and letting a customer know that you appreciate her.

12:50 Lecture 8: Doing Things That Don’t Scale, PR, How to Get Started: Walker Williams, Founder, Teespring, Justin Kan, Founder, Twitch and Partner, Y Combinator , Stanley Tang, Founder, DoorDash

Ok I’m starting to zone out a bit. I can’t tell if it was the last talk, or just a general feeling. More students have shown up, which is nice. Its gross outside, kind of the weather where you just want to be at home. The idea of sitting in this room for the next 11.5 hours is not that much fun.

Stanley Tang, Founder, DoorDash

Door Dash, launched in an hour with a landing page, pdf menus, and a phone number. Someone called and the founders delivered the order. At the beginning founders should focus on testing one’s idea and seeing if the product is something that consumers want.

The founders of Door Dash created a technology company but everything was done manually by the founders, emails, delivery, google docs for orders. This facilitated maximum learning for the founders to become experts in the field. Scaling doesn’t matter in the beginning. Learn then scale.

Test your hypothesis, launch fast, and do things that don’t scale.

Walker Williams, Founder, Teespring

Doing things that don’t scale is an advantage! By don’t scale: doing things that are financially unsustainable, likely because they take way too much time.

When you first launch, you are going to be bad at selling it, you don’t have success stories, you don’t know pain points. So it is your responsibility to spend personal time to acquire first customers. Think about pushing a giant rock, the beginning is the toughest! It is impossible to make money on early users, but don’t give away your product for free. Need to turn these first users into champions, through an amazing experience. Do this by talking with your users. You are never going to get a better sense of your company and product then by talking with your customers.

Three ways to talk to customers: run customer service as long as possible, proactively reach out to current and churned customers, and social media/communities.

Problems are inevitable, do whatever it takes to make them right. Again message is treat people like people.

Product-market fit. Product to scale is unlikely the product that you launch. Focus about speed.

Only worry about growing your business to the next order of magnitude. Don’t worry about a 1 million users, when you have a 100 users.

The moment you stop doing things that don’t scale, is the moment you give up one of your biggest advantages.

Justin Kan, Founder, Twitch and Partner, Y Combinator

Getting press! Types of stories: product launches, fundraising, milestones, business overviews, stunts, hiring announcements, and contributed articles.

Think of a story, get introduced, set a date, reach out, pitch, follow up, and launch your news.

A lot of this advice is back to spending thoughtful time to think through your story and prepare and engage with reporters.

Public relations firms, are likely a waste of time or too expensive for a startup.

Make sure getting press is worth it, press does not mean that you are successful, and its completely not scalable.

This is smart commentary, I think some press is always a good signal to stakeholders that the company is legitimate. Press has a wave, where getting the first article helps a startup get a second article, etc, but at some point this wave will pass. And I’ve found its hard to transition from chasing press to other sales channels. I do believe press can be a bit of a distraction.

1:55PM Lecture 9: How to Raise Money:

Marc Andreessen, Founder, Andreessen Horowitz and Founder, Netscape Ron Conway, Founder, SV Angel, Parker Conrad, Founder, Zenefits

When evaluating a founder, evaluate: is this person a leader? Is this person obsessed with this product? Does this person possess strong communication skills?

Venture capital is a business of outliers. A lot of startups that check all boxes, don’t hit a huge return. Investing in a team that has extreme strengths is a good strategy, as weaknesses can perhaps be fixed.

Sam Altman’s “How to Start a Startup” - MashupCorner

Ok sorry this is a bit grainy, but I didn’t feel like standing up to take a picture of it. In a quick glance I love it. It provides a difficult but simple path for entrepreneurs to follow while building a business.

You are almost always better off in making your business better and not making your pitch better. Almost everything you’ll ever do as a founder is harder than raising money. Raising money perhaps not a milestone or an accomplishment, it just empowers a founder to do the other things, which are harder and are important.

Startups have multiple risks, launch, product, customer, etc. As a company grows, it achieves milestones, and reduces different layers of risk.

If an investor makes a commitment, get in your car, type an email to confirm details.

To raise a series A, a seed raise is required, unless it is a successful founder. This series A, is usually sold at 20–30%. It is important to pick a good seed investor, who has connections with venture capitalists. But it is quite difficult to figure out who is a good VC. 40% in an Angel round would demotivate a founder and the team.

I think I’m 10 hours away from being finished with this marathon and close to 5,000 words written for the day.

When you found a company, you have to find someone better than you to be a co-founder. The anomaly is Mark Zuckerburg as a solo co-founder.

If your company goes great, it doesn’t matter who your investors are. But each large company, faces large storms, and through these storms it matters who are your co-founders, investors, etc. Think of an investment in a startup as an investment for life. Does a VC get your business? Do you respect their team?

This gets back to a much earlier point about trusting your intuition about people.

Each investment in a company, eliminates a company in that category, do to conflict policies. You don’t know which companies will come up next. The second issue isn’t the loss of funds, but the opportunity cost in human capital. Each open board seat that a firm takes, closes this possibility to take on a different board seat. A partner can take on 8–12 companies at a time. Given these two metrics, it is a high opportunity for one investment.

If a startup goes well, the founder has control. If a startup is going poorly, the investors have control.

2:45pm — Lecture 10: Culture

Alfred Lin, Former COO, Zappos and Partner, Sequoia Capital

What is culture?

Assumptions beliefs, core values, behaviors, actions that pursue the company’s mission.

Culture is about creating an environment that increases performance. The benefits of the right culture, focuses on employees and create a guide for how to act. Creating a fantastic culture will empower an organization to attract and retain top talent.

One way to develop core values is to think about what you don’t want in an employee, and focus on creating the opposite. This is only step one though as teamwork, honesty, service are difficult to create.

Sam Altman’s “How to Start a Startup” - MashupCorner

Brian Chesky, Founder, Airbnb

You go from building a product at first, and at some point you switch to growing a company. As a founder, you are the parent of a company. As a parent you want your child to out live you. So you have to create the right environment to maximize your child’s life.

Reaching back to a few hours ago. Startups are about people. To build a billion dollar company, you need amazing people. Amazing people don’t want to work for an organization with a crappy culture.

Integrity and honesty aren’t core values, they are what everyone should have. Airbnb only hires people who champion the mission, bring people together. Hire people who feel that working for airbnb is there calling.

Culture does not pay off in the short term. Culture is an investment, that takes time. Stage 1 interview at Airbnb: if you could hire anyone in the world, would you hire the person that you are interviewing. Stage 2 does this person fit into our culture.

Branding and culture are tied together. Culture is what happens inside the company. Brand is the promise outside the company. Your brand evangelists are your employees.

The way to build a company is through vision, strategy, and through finding the right people. The way to win is to talk about what you value.

Talking about Airbnb’s hosts reinforcing the company’s values. Brian discusses that hosts are partners. I think all interactions in business are partnerships, suppliers, customers, employees, etc. It’s incredibly difficult to locate partners that share these values, and convey why these values are important, but this struggle is similar to user experience. It has to be constantly reinforced and improved.

Its better to have 100 people who love you as opposed to 1,000,000 people who like you. To create 100 users who love you, do things that don’t scale. Focus on creating a perfect experience for one person. Then extract information how this can be scaled. But the perfect experience for a small amount of customers must occur first.

3:35pm — Lecture 11: Hiring and Culture, Part II

Patrick Collison, Co-Founder, Stripe
John Collison, Co-Founder, Stripe
Ben Silbermann, Founder & CEO, Pinterest

Half way there!! Again unsure if this is good or no One thing I’ve noticed is that I feel now, I have a decent understanding of what Sam Altman wears.

You can’t be involved in every decision of your business. Culture is geared to be a guide for other stakeholders for you as your company grows. The first 10 hires are the most important. Its difficult to sell top people, because top people can go to established startups. Also aside from talent, these individuals determine a lot of the trajectory environment of the company. You want people who will care so much, that even a small detail is a great pain, that each person on a startup team wants to fix. For the first 10 people work with them as much as possible prior to hiring on a full time basis. Like on a project for a week, its hard to fake it for a week.

If you are hiring someone outside of your field, talk with leaders in your field how to determine whether someone is ‘World Class’. You want to learn what is good or bad prior to the interview process.

Sam Altman’s “How to Start a Startup” - MashupCorner

The note above and corresponding homemade soup is from Kaelya and is definitely the highlight of my day!

A startup is a company that doesn’t have so many principal agent problems that are prevalent at large companies. So you can push 100% transparency. Stripe used (and may still) to bcc each other employee on every outgoing email, to create cohesion and share information.

The advantage of working on a startup, is that no large company is crazy enough to give a relevantly new hire a massive responsibility. This responsibility is exciting, but just because an employee may fail, doesn’t mean that the employee is a failure.

I like this thought a lot. Failure isn’t such a bad thing. Consistent failure obviously is a problem. But, to create something meaningful, one should be allowed to swing for the fences, and if they are working in a thoughtful and authentic manner, this effort should be championed. Creating an environment where employees are scared to experiment and fail, would be detrimental to a company’s growth. This freedom to fail or lack there of, is likely correlated with trust between an employee and employer.

4:23pm Lecture 12: Building for the Enterprise

Aaron Levie, Founder, Box

Look for trends in the changing technology factors. See Uber, the future isn’t cabs, but driverless cars.

Goal of this talk is to convince someone to create an enterprise company.

In the consumer space, you are always battling convincing someone to pay for your product or gain advertising. Paid apps 35 billion vs digital advertising $135 billion vs 3.7 trillion spent in the enterprise IT industry.

Value equation is quite different for enterprise vs consumer. Consumers try to minimize cost or find the best deal. Enterprise space focus is productivity, things that are very high value, thus corresponding price points are higher.

Enterprise products and sales appear complex. But everything about the enterprise market has changed since 2007. First, all companies are moving to the cloud. This has standardized platforms all companies employ. This reduces barriers to reach out to large companies. Second a company can now sell IT products to very small companies, which previously couldn’t afford to build or employ the infrastructure.

Uber represents a shift in the entire transportation industry. Every industry is gong through a similar shift. And each company will need technology companies to facilitate this shift. Think about a multi-channel commerce. People want to buy goods at any place, at any time, with better information.

Practical advice:

  1. Look for new technologies that are disruptive to how things have been done and what can be done.
  2. Intentionally start small, a sliver of a problem, then expand over time. If someone says its a toy, you are on to something. Industry leaders don’t go after small but important pains. Startups can exploit this point, to build a fantastic product, then move up market to larger problems.
  3. Find things incumbents can’t or won’t do because its currently not feasible.
  4. Find customers who are already employing technology to move forward, and pitch to them.
  5. Listen to customers, but analyze requests and build what they need. Think at a higher level than a customer would. But employ customer feedback to see what works and what doesn’t. Keep the consumer at the center of the product.
  6. Product should sell itself, but you need sales to navigate customers and close deals!

5:14pm Lecture 13: How To Be A Great Founder

Reid Hoffman, Partner, Greylock Ventures and Founder, LinkedIn

Founders are thought of super heroes. But a single person never builds a business. There is such a diversity of skills is required to create a business, that a person needs help. Choosing good cofounders means choosing high achieving individuals who possess complimentary skills.

In choosing a location for one’s business, choose a location with the strongest network for your business. Groupon, may not have grown in San Francisco. Groupon needed massive sales forces early on, and silicon valley is opposed to this mentality. Go to where your startup will be most successful.

Ok, I’m trying to follow long, there is a lot of dichotomies, (vision vs data or persistence vs flexibility) with no clear answer. I think the purpose of this talk that a founder has to be thoughtful with each decision.

Thus maybe a higher level thought is that a founder should focus on being aware, in this David Foster Wallace, This is Water sort of way. Specifically:

“learning how to think really means learning how to exercise some control over how and what you think. It means being conscious and aware enough to choose what you pay attention to and to choose how you construct meaning from experience. Because if you cannot exercise this kind of choice in adult life, you will be totally hosed.”

Though this thought feels right, I think it could just because my brain may be a bit tired, still I’m throwing this quote out there, and can reflect on it another day.

There is not one skill set to be a good founder, but you must process, adopt and build a coherent network, to become one.

Ok I’m zoning out during this Question and Answer…

6:04 Lecture 14: How to Operate

Keith Rabois, Partner, Khosla Ventures

Building a company is putting all the irrational people in one room and moving forward. Growing a company is like a drawing that looks clean and neat but in reality it is an engine with duct tape. Ultimately you want to develop a high performing machine that is largely automated.

Role of a leader

Sam Altman’s “How to Start a Startup” - MashupCorner

Early on need to think about what actions are colds and which actions are fatal. To achieve this, need to edit. By editing should

  1. Simplify, this is the goal of a manager. Make everything incredibly simple. You can change the world in 140 characters, you can market massive products with 2 words. Edit everything
  2. Clarify. Ask questions. Eliminate unnecessary processes.
  3. Allocate Resources. This can be top down or bottom up.
  4. Ensure Consistent Voice.
  5. Delegate.

Goal over time is to use less red ink every day. Measure how much red ink you are creating each month, it should trend downwards. When you are unsure and there are low consequences, you should delegate. When it is a high risk decision, you can’t delegate. You need to try to explain why in your decision process.

Important to edit the team. Think about people as gun barrels vs ammunition. Most people are ammunition. A barrel can take an idea through shipping and bring people with them. Its incredibly hard to find barrels.

Test peoples roles and have it expand until it breaks and it will break.

Where to focus on barrels? Insist on focus! A+ problems are difficult to solve, so these problems are easy for people to skip over. Force people to work on A+ problems.

Metrics, analytics & transparency: Create a dashboard, that shows company performance. Measure how many employees employ this dashboard. Share board meeting decks and explain each slide. Share notes of each meeting greater than 2 people.

Be smart with metrics, if you seek to minimize fraud, you may create a bad user experience.

Details matter: If each person executes at a high level, a company will perform at a high level. Detail obsession is required to build a successful company. For example, food matters. If food is awful, people are unhappy, and complain about the food. If food is healthy and fantastic, people will focus on innovating. Same with space, space is a determinant of culture.

Treat customer support like a product. You should have a one on one with each person you manage each week or two week. Agenda should be crafted by the employee.

High levels employees can be evaluated by output divided by number of employees managed.

6:50pm Lecture 15: How to Manage

Ben Horowitz, Founder, Andreessen Horowitz, and Founder, and Opsware

When making a decision you have to be able to understand how this decision will be understood and affect each stakeholder in your company. Especially including individuals who may not be in the room.

Example 1: Demote or Fire — hard worker, well liked, just not World Class. Think from perspective of yourself (CEO), employee, and everyone else.

Example 2: Excellent Employee asks for a raise

Need to create an evaluation process, and not offer raises outside of this process. Processes protect the culture.

Yes ok processes minimize risk, but I think this is perhaps an oversimplification. Creating an incentive structure to reward progress is quite difficult. Creating a reward process which is a bit off can be toxic for a culture and perhaps kill productivity. Infact all of these examples seem to explore the incentive structure and impacts of different decisions. I’m unsure if I can forecast a point to put everything together. But I’m assuming it will be a general idea that one just needs to consider decisions from different stakeholders both in the short term and the long term. There may not be an optimal answer, but you can attempt to explain these constraints to others.

Ok this connection to the slave revolution is a bit difficult for me to follow. I see its connected to culture and the culture you want to create should be driving decisions, but I’m just sort of tuning it out.

Lecture 16: How to Run a User Interview

Emmett Shear, Founder and CEO, Twitch

Who you talk to is just as important as what you ask them. Who would you go ask if you wanted to build a lecture taking note app?

Figure out who is the most important person who would use an app. Beyond college students, you could talk to administrators, parents, and get familiar with the space.

Emmett is walking a student through some basic interview questions. The goal of these questions is just to understand the thought process behind a user. This is incredibly basic, open ended questions, letting the respondent explain in a judgement free way how she currently takes notes. He isn’t pitching his app, but instead just trying to understand the current process. This is smart as people are bad at predicting what would help them or not.

On a side note, I’m officially getting a little tired. I do have two mountain dews left. They aren’t even mountain dews, but instead mountain lightning. I know its insanely bad to drink them but I think I can drink one now and another one at no earlier than 10pm and chance it from there. I’m riding the mountain lightning wave.

Sam Altman’s “How to Start a Startup” - MashupCorner

Throw an idea out, and see if sticks.

Personally this is my favorite approach. Find some sort of extremely low cost way to test an idea, similar to Door Dash as described above.

For new products most people are non-users. If you only talk to people who employ your competitors, its difficult to expand the market.

The features people were asking for weren’t that important. What was important, was the frustrations they discussed or what they really cared about. No one really does that, asks a person, what do you need, and then actually building that feature a month or two later. Recording interviews, is a positive.

Ok, so if you back up, talking to users I believe is decently simple. Realize that a user has all the information and you need this information. I personally think to start a real conversation is the best way to begin to get a sense for a user. Really focusing on the process behind a person’s decisions and what frustrates them about the process. I think understanding frustrations is key. Really I’ve tried to empathize with each person I work with, including customers. By the end of user questionnaires, you should be able to explain this process to someone outside of your organization. This explanation should make sense to both someone who has no prior knowledge in this field. I’m sorry, I’m a bit tired, but what I mean is this explanation should be reasonable.

8:28pm Lecture 17: How to Design Hardware Products

Hosain Rahman, Founder, Jawbone

Don’t think about things, think about the user. JawBone focuses on Hardware + Software + Data, which is incredibly complex. Everything is a system.

The process of creation:

Sam Altman’s “How to Start a Startup” - MashupCorner

Sideways funnel of from wide to narrow left to right: exploration, early validation, concept, planning, development, and continued innovation.

Exploration: Its a bit of an angel round science show. Would I give this guy 50,000?

Early Validation: A lot of whys & hows

Concept: Highly resolved whys, differentiation strategy, and road map

Planning: Lots of trade offs, dealing with the reality of product, and financial limitations. At the end of the day, does this cross the value threshold to develop?

Development: Sign off from design, hardware, software…

“What is the user problem that once we solve user can’t live without?” — Love this quote its so good!

I found this talk to be really good. Focusing on a mindset behind the process. This mindset can be applied to any product or any organization.

We can dream a lot faster than we can innovate. It can take years to explore a process. One can skip steps, or go backwards through the steps. Again the focus is on the mindset

When entering a market blow it wide open by a magnitude of X….

9:15pm — Lecture 18: Legal and Accounting Basics for Startups

Kirsty Nathoo, Carolynn Levy, Partners, Y Combinator

Students are still hanging with me, its pretty cool. I think we are the only people in the building.

The easiest place to form a company is Delaware. It saves time & effort.

Equity should be approximately equal. It is a red flag if this isn’t the case. The value of a company is in the long term future of that company. Thus the past is such a short period of time compared to the future that the past shouldn’t matter that much. A startup needs a complete team to execute, equal distribution of equity helps keep a team together.

83b election is incredible important to sign and retain proof.

Standard vesting is 4 years with a 1 year cliff. Meaning, after one year a founder receives 25% of her shares. If she leaves on this date, the company will buy back the remaining shares at the original price. Vesting creates an incentive structure to create founders on the team. Solo founders need vesting as well, it helps create a culture that we are in it for the long haul.

Logistics of raising funds — priced vs non-priced rounds. Priced refers to value of the company.

Seeds rounds are non-priced, which is quicker. Investors spend 100k now, and receive shares at a future value, at a future round. This is usually set at an upper bound, the cap. Imagine investing 2 million at a cap at 6 million cap. Then when the note converts, these investors could own 25%, plus the next round.

Geez, I can’t keep up with typing, this is intense in a good way.

Investors need to be sophisticated, as neighbors, family members, family friends, may not understand risks and that this is a long term investment.

Investor requests

  1. board seats: most cases you want to say no, unless the individuals can add significant value. Investors should help because they are investors, they shouldn’t need extra shares to be an advisor.
  2. pro rata rights:
  3. information rights: be careful with over reach.

Just because the amount and type of financing is right, a founder needs to know everything about finances.

Business expenses are the costs of running your business. Business expenses get deducted from revenue to reduce the tax liability. If you are embarrassed by an investor knowing about an expense, it is not a business expense. But you need to keep all receipts.

Founders are an employee and need to get paid and pay payroll taxes. Set up a payroll service. Don’t go overboard with salaries. Founder breakups get super messy when there are unpaid wages. Because the company could have broken the law and the founder who is forced out can use this leverage for additional shares.

For non-founders, one needs to figure out whether an someoneis a contractor or an employee. For contractors, the individual pays taxes and there are minimal liabilities. It is important to obtain an IP agreement.

Again a payroll provider like zenpayroll can alleviate a lot of these pains.

Firing people is really hard. Fire quick, in person, with a third party. Pay all vacation and unpaid wages, cut off access, and repurchase all unvested stock.

Running a company is knowing and following the rules!

10:04pm Lecture 19: Sales and Marketing, How to Talk to Investors, Investor Meeting Roleplaying

Tyler Bosmeny, Founder and CEO, Clever,

Talking to users is part of sales. As a founder you have unique advantages, knowledge about the product and passion! Build or Sell — nothing else matters!

Sales funnel: prospecting, conversations, closing and revenue/promised land.

Prospecting — find the innovators 2.5% of the market.

Conferences are the place to go! Get the attendee list, write everyone in advance, set up meeting.

Cold emails should be concise, clear.

Conversations: shut up, listen, keep it simple.

Religious Follow up…

Sam Altman’s “How to Start a Startup” - MashupCorner

You want to get people to a yes or a no quickly! Maybes can be a pipeline of death.

There are times that you want to pass on a customer. If they state they need one more feature.

Instead of a free trail offer, with an annual contract, 30–60 days with the ability to void the contract.

Develop your sales cycle based upon the amount of revenue that a customer will offer. A $10 customer is a much different sales cycle than a $10,000 sales cycle.

This is really talking about the cost to acquire a customer.

Michael Seibel, Partner, Y Combinator, Qasar Younis, Dalton Caldwell, Partners, Y Combinator

Best way to improve your pitch is to improve your company.

30 second pitch

  • What does your company do?
  • How big is the market?
  • How much traction do you have? “We launched in X, we grow at Y, and have Z amount of users/revenue” — convince investors that you are using fast.

2 minute pitch

  • 30 second pitch +
  • unique insight: competitive advantage, tell me something that you don’t know, that the biggest players in the market don’t know. Why are you going to kill the competitors, in 2 sentences. Create aha moment.
  • how you make money: business model
  • team: if they have done something impressive say something, if not, we want to hear how many founders, how many technical founders, how long have you known each other, hopefully more than 6 months, and how you met.
  • ask: convertible note or not, cap, raise amount, equity ask, etc, use jargon here, and be clear.

Raise money when you have traction. How do create a scenario where you are strong compared to investors. Have you created a plan where you can grow with little or no money.

To reach an investor, warm introductions are necessary, think in parallel, meaning schedule all meeting in the same week, “we are building crazy for 2 weeks, can we meet in the third week”, 1 team member should be invested in fundraising full time.

Pitches:

Tell a compelling narrative, make it collaborative, and be clear about numbers. Its obvious when people know their business. Anything other than a check is a NO. Create deal heat. Do diligence on the investors. Know who you are selling your company to. Know when to stop fundraising, because it is easier than building a company. Building a company is the goal not fundraising.

10:50pm Lecture 20: Later-Stage Advice

Sam Altman, President, Y Combinator

Whatttttttttttttt I can’t believe I’ve reached the last lecture. Seriously I am smiling. I can’t believe students are here with me, so cool!

Companies are flat until about 25 employees. When lack of structure fails, it is a disaster. The structure doesn’t need to be a complicated. There needs to be clear and simple employee structure. You want to innovate on your product don’t innovate on the business structure. Seriously, don’t make it complicated.

Failure cases:

  1. Being afraid to hire senior people
  2. Hero mode — working to an extreme level
  3. Bad delegation — “hey we need to do a big thing, do research and come back to me & I’ll make the decision and you delegate” vs “here’s what I think, go do research, make a decision, and let me know, I trust you”
  4. Personal organization: tracking productivity.

Codify how you and the company does things. Every founder wishes they wrote down how and why earlier.

Human Resources can speed a company up! It creates a clear structure, offers performance feedback which should be simple and frequent, compensation bands tied to performance and equate.

Equity be generous: your investors will give you bad advice, give 3–5% a year. Keep up with refresher grants to stay in front of individuals vesting schedules. Think of structures such as 6 years, pyramid vesting, and continuous forward vesting and get a management forward system.

There are a lot of rules that change at 50 employees.

Monitor your team for burnout, the company is in a different stage and stability is key.

Announce every offer internally before externally.

Create a program to onboard new employees.

Productivity per employee goes down as it grows. Single most important word is alignment. Test: ask founders and pull 10 random employs, what are the top three goals for the company. Important to reiterate goals and road map.

Transparency and communication is crucial to creating alignment.

Oh wow, I’m starting to fade fast.

The goal is to build a company that creates enduring value over a long period of time.

Most companies do one great thing that the founder pushes along, but repeated innovation is so rare.

As the company grows, the psychological intensity for founders increases. When you are small, people love you, when you grow haters grow. Ignore them… Take vacation and monitor burnout.

Acquisition offers can tank a company. It is demoralizing if it doesn’t happen, can kill morale.

Startups fail when founders quit.

Founders need to figure out the key messages of the company, and repeat it again and again.

for deals

  1. Build a great product,
  2. develop a personal connection,
  3. have a competitive dynamic,
  4. be persistent, and
  5. ask for what you want.

Sam Altman’s “How to Start a Startup” - MashupCorner

 

Wow, I can’t believe I completed this 16 hour marathon. Such a cool experience. Amazing that FSU students could have watched it at home but they are sitting here with me at 11:30pm as we finish this marathon.

1 day

20 talks

7:10am-11:40pm

9,000+ words!!

30 students came by sometime throughout the day!

If you Enjoyed reading the article please support us by Liking the page and share the article around Cheers

One thought on “So I watched Sam Altman’s “How to Start a Startup” in a single day”

Comments are closed