Sam Altman - How to Succeed with a Startup

0:00 okay today I'm going to talk about how to succeed

0:03 with a startup obviously more than can be said here in 20 minutes

0:06 but I will do the best I can the most important thing

0:10 the number one lesson we try to teach startups is that the degree

0:14 to which you're successful approximates the degree to which you build

0:18 a product that is so good people spontaneously tell their friends about

0:21 it startups always ask us for the secret to success they always

0:25 want to believe it's something other than this because this is really

0:28 hard to do but this is it if you can build

0:31 a product that is so good people spontaneously tell their friends about it

0:34 you have done 80% of the work that you need to be

0:37 a really successful startup if you think about the most successful companies

0:41 you know Google Facebook whatever you probably found out about them because

0:46 a friend of yours said you got to try this it's great

0:48 so this is the bar something that people love so much they

0:52 tell their friends about it one important indicator for a product like

0:56 that is a product it's simple to explain and easy to understand

1:00 if you can't explain in a few words what you do

1:03 and if people don't if at least some people don't say oh

1:05 that's pretty interesting that's usually a mistake

1:07 it's usually a sign of unclear

1:09 thinking or a need that is not big enough another thing

1:14 that startups need to look for is a market that is either

1:17 started to undergo or is soon going to under grow exponential growth

1:20 I think this is actually related to one of the biggest mistakes

1:24 investors make when evaluating startups investors

1:27 always say well what's your growth

1:28 rate we care about the growth rate investors will forgive smallish revenue

1:32 today if it's growing quickly for some reason people don't think about

1:35 markets this way but if you think about the most important startups

1:39 they are the ones that I start in small markets that are

1:42 growing very very quickly 11 years ago the market for iPhone apps

1:46 is zero dollars it's not huge and I think if you only

1:50 think about the TAM today you'll make a big mistake what you

1:53 really want to do is identify a market that's going to grow

1:56 every year and be able to ride that up elevator a really

2:01 important thing to figure figuring this out is learning how to differentiate

2:05 between real trends and fake trends a real trend is something

2:09 that's actually going to happen in the fake trend is not

2:10 or at least not yet and before you make a big bet

2:14 on a new platform you want to make sure it's real now

2:17 there's an easy trick for this which I'll share now real trends are

2:20 ones where a new technology platform comes along and the early

2:24 adopters use it obsessively and tell the friends how much they love

2:27 it a fake trend is one where people may buy the product

2:31 but don't use it or at least not enough so an example

2:34 of a real trend I already mentioned the iPhone I'll mention

2:37 that again when the iPhone first came out many people were dismissive because

2:41 they only sold a million or two million that year and they

2:44 said well this just doesn't matter but for the people that had

2:47 an iPhone they used it for hours every day it became

2:49 central to their lives they loved it they told their friends you've

2:52 got to get one and I think it was obvious then do

2:55 people pay an attention that something had fundamentally shifted and we had

2:58 a new a new computing platform that was gonna spawn huge businesses

3:02 and it was a good time to bet on mobile apps a fake

3:06 trend or at least a fake trend as of august 2018 I

3:10 would say as VR I do believe VR will be big someday

3:13 but today most people that I know that own a VR

3:17 headset use it never or very rarely and so although a lot

3:20 of people talk about it and maybe even a lot of people

3:22 buy them there's not the intense usage per user among the early adopters

3:27 that I think you want to see before you make a big

3:29 bet another thing that startups need at least one evangelical founder usually

3:35 the CEO someone at the startup has got to be the person

3:39 that is going to recruit sell the product talk to the press

3:44 raise money and this requires someone who can infect with enthusiasm

3:49 the whole world about what the company is trying to do and someone

3:54 who becomes the chief evangelist for the company it's very hard

3:58 to succeed wildly without that it's very hard to build a team

4:02 at all without being able to do that one thing that helps

4:06 for this is having an ambitious vision you never want to be

4:10 grandiose that turns people off but you want to let yourself grow

4:13 more ambitious over time and as long as you do that organically

4:16 people will respond ambitious visions are exciting they're fun to work

4:20 on in fact I think in 2018 at least in Silicon Valley it's

4:25 easier to start a hard startup than it is to start

4:28 an easy startup now this sounds

4:30 paradoxical but ambitious projects are interesting you

4:33 know in the current environment it may be relatively easy to raise

4:37 capital but it's really hard to do everything else there are so

4:39 many startups it's so easy to start one they all sound so

4:42 promising that that bringing together enough

4:46 talent in one organization is really

4:48 hard to do and if you're working on a problem that you

4:51 know maybe modestly successful it's it's kind of easy to get

4:55 the first two people to join you can give them a lot

4:57 of equity but then it gets really hard why is employee xx

5:01 gonna join why is this why does this matter to the world

5:03 why should someone work on your startup versus any of the other

5:07 things they could do and picking something that matters if you're successful

5:11 is a great way to do that and so I think it's

5:14 really important to think about when you're starting a company how is

5:18 this going to evolve into a vision that a lot of people

5:21 want to help with but a lot of people want to be

5:23 associated with because I think getting talent and getting mindshare it is

5:28 really hard in the current environment and people are interested in startups

5:32 that matter another thing that we've noticed among our best Saunders again

5:38 and again and again is that they are have a confident

5:42 and definite view of the future they may be wrong and so we

5:47 say it's good to be confident and flexible but this idea

5:50 that you are confident in definite this is what I think is going

5:53 to happen or this is what is going to happen and being

5:56 relatively sure of that having courage of your convictions being a clear

6:02 leader saying we're gonna do this and that's why even

6:05 in the face of a lot of doubt that seems to really correlate

6:08 with success and this comes back to having an ambitious vision

6:14 but the entire startup ecosystem is best set up to support companies that have

6:19 a low chance of success but are huge if they work

6:22 and I think going for something that is huge if it works

6:26 will attract the best people not going to talk too much about

6:31 the team there are a lot of obvious things I could say

6:34 that have been said many times by many people about you know

6:38 you need smart people who want to work really hard and who

6:41 communicate well these are all really true but I'd like to mention

6:44 a few non-obvious things that we've noticed that we don't hear people

6:47 say as much about the team you need to assemble the note

6:51 Khosla says that the team you build is the company you

6:54 build and I really think that's true I've still met only

6:58 a handful of founders I think that spend enough time on recruiting like

7:01 Mark Zuckerberg is famously one of them but building a great team

7:04 I think other than picking the right market and building a great

7:08 product is the most important thing you do all founders go through

7:12 a transition all successful founders where you switch from building a product

7:16 to building a company and building a company really is about

7:19 the team so you need to mist the whole world will be telling

7:26 you why you're gonna fail as a start-up if you don't have

7:28 that internal fire of belief if you don't have people who say

7:31 you know what we are gonna do this and it doesn't matter

7:34 what the haters say we're gonna figure this out and there is

7:37 you know this problem it must be solvable if you don't have

7:40 the spirit of optimism on the team it's very hard to succeed

7:44 when the world continues to punch you in the face you

7:48 need at least some idea generators there are a handful of people

7:51 in any company that has gone on to be really successful

7:54 that I that I've been able to work with who are just really

7:58 good at coming up with lots of ideas you don't too many

8:01 these people because that's more ideas than a company can follow through

8:04 on but having some people within a company that are just constantly

8:09 throwing out new ideas most of which will be bad it turns

8:13 out to be super important to have on the team this spirit

8:18 of we'll figure it out is my favorite thing to hear

8:22 among early startup team members a lot of things go wrong

8:26 the situations that startups win in tend to be incredibly dynamic and so

8:31 this idea that even if I'm not qualified on paper even if

8:34 I haven't solved this problem before even if this problem feels like

8:37 it's gonna kill the company which many problems will feel that way

8:40 this spirit among the team of you know what we've got the people

8:45 we need we're gonna figure this out we're gonna get this done

8:47 that's super important another thing that I love to hear from early

8:53 team members is I've got it so you know you hear

8:56 in big companies a lot of people say that's not my department

8:58 someone else is gonna do that yeah or you know who

9:01 this is really bad this is gonna hurt us and you want people

9:04 who just step up and say I'll do it I've got it

9:06 don't worry about it you want people to have a bias towards

9:11 action startups especially in their early days often win by moving very

9:16 quickly you never get as much data as you'd like you never

9:20 have as much time to deliberate as you'd like and you want

9:22 people who will are willing to act with much less data than

9:28 they'd like to have with much less certainty and then if they

9:31 act then it doesn't work they adapt really quickly and try something

9:33 else we talk also about the blessing of inexperience we have seen

9:40 many of our startups do incredible things because no one told them

9:44 it was hard no one told them they can't do it um

9:46 there's a great quote from Steve Wozniak about how all the best

9:49 things he ever did came from having no experience whatsoever and having

9:52 no money and you know that obviously is not always true

9:57 but there is a magic thing that happens with startups especially

10:00 in their early days before they've learned they're not supposed to be able

10:03 to do certain things and so I think as a start-up again

10:07 doesn't work to have everybody being experienced but you can take more

10:10 bets than you normally would on inexperienced but super high potential

10:14 people that's the end of a topic on on team one

10:20 of the most important jobs you have as a founder is to never

10:23 lose momentum and this is a little bit depressing because it means

10:26 for the first few years you never get to take your foot

10:28 off of the gas you never get to really rest we try

10:32 to be honest about this that startups are not the best choice

10:35 for work-life balance at all but especially in the early days startups survive

10:39 on their own momentum if you have momentum people keep delivering

10:44 results beyond what they think they're capable of if you lose momentum

10:47 it's very difficult to get it back and so continuing to make

10:52 sure that the startup has a cadence that the startup keeps winning

10:55 on a relatively predictively short and predictable and this is really important

11:00 and it's up to the founders to make sure that that you

11:02 don't lose ahold of this another thing that we think startups need

11:09 is a competitive advantage over time now this is something that sounds

11:13 so obvious I hesitated to even put it in this is

11:16 well discussed but we're seeing more and more startups apply to YC

11:22 and when we ask them so what is the long term monopoly

11:25 effect here what is the long term competitive advantage what is you know

11:28 where is the network effect in this business they look at us

11:31 like it's the first time we've ever they've ever heard this question

11:34 all of the really great businesses I know have an answer

11:38 to this question and in fact the better they are the more they

11:41 pretend not to but this is something that you you want

11:45 to have a plan for another thing you want to have a plan

11:47 for is at least a sensible business model you don't have

11:51 to have it all figured out at the beginning but when we

11:54 ask founders so how are you ever going to make money

11:56 and they look at us like it's the first time they've ever

11:58 been asked that question which happens more often than you would think

12:00 recently that's a bad sign too again this one is so common

12:06 that I hesitate to put it in here but when we

12:08 ask a start-up how they're gonna grow how they're gonna get users

12:12 and they look at us like it's the first time they've ever

12:14 heard that question bad sign so some sensible idea to try here first

12:21 Paul Buchheit once spent a bunch of time looking one

12:25 of the YC partners spent a bunch of time looking at the traits

12:28 of our best founders and tried to distill down what they were

12:32 and he came up with a frugality focus obsession and love I actually

12:36 I think that is really good I don't know how much

12:38 to add but I think these are things that you should be able

12:42 to say about what you're doing and you as a founder

12:45 and finally I want to talk about why startups get to be big

12:50 companies we've there are lots of reasons I'm going to talk about

12:54 a few common ones here I think these trends are valuable enough

12:59 that as you evaluate startup ideas you might do it's worth thinking

13:02 about if you fit into these because it's really hard for startups

13:06 to beat big companies most of the time and here are

13:09 some areas where we see it happening repeatedly so I think one

13:15 difference is if you are a product manager at a big company

13:19 and you want to do something that sounds sounds like a bad

13:22 idea but is a good idea you have to get everybody

13:26 from your boss sometimes all the way up to the CEO to say

13:29 yes so one no can kill you if you're a start-up you

13:33 can go to I see demo day any number of the thousands

13:37 of investors can say yes and you get to have a crack

13:39 at it so it's a very different mindset and for those ideas

13:43 that sound bad but are good because of this phenomenon this one

13:47 no versus one yes startups can win in fact startups usually

13:51 do beat big companies in that category of ideas so look look

13:55 for ideas that sound bad but are good and where you are

13:59 much more likely to get one yes then someone in a big

14:03 company is to get all yeses another area where startups usually be

14:10 big companies aren't very fast changing markets startups great speed great

14:14 advantage is an agility and speed the more a market is changing

14:17 the higher the number of decisions you get to make and the higher

14:22 number of tweaks to your product and your strategy you get

14:25 to make and you want to optimize the number of those decisions

14:28 that someone has to make to compete with you because a big

14:31 company will make them on average worse than certainly much slower than

14:34 you do so the speed of market evolution gives you a lot more

14:39 chances to compound your advantage over a big company and then

14:45 finally startups usually win on big platform shifts many people have observed

14:50 that startups companies clusters those clusters

14:53 usually come after a big platform

14:55 shift we'll stick with the iPhone example here after mobile apps became

15:00 a thing many new companies got started that are now quite valuable

15:05 and one of the reasons here is you know most large companies

15:09 work on sort of an annual cadence at least and when there's

15:13 a huge platform shift they are not good at making a big enough

15:17 strategic pivot the Battleship just turns too slowly whereas a start-up

15:21 can say wow woke up this morning the world is fundamentally different

15:25 than it was six months ago we're gonna go all

15:27 in on this new direction and so that's a way that startups usually win

15:30 or that that's an area where startups usually win there there are

15:34 many others but thinking about these three I think is is directionally

15:39 a good thing to do alright that's all I've got for today

15:43 Jeff thank you very much for having me and for the class you