0:00

By the way, in case you haven't heard, my brand

0:02

new book, Feel Good Productivity, is now out. It is

0:04

available everywhere books are sold, and it's actually hit the

0:06

New York Times and also the Sunday Times Best Seller

0:08

list. So thank you to everyone who's already got a

0:11

copy of the book. If you've read the book already,

0:13

I would love a review on Amazon. And if you

0:15

haven't yet checked it out, you may like to check

0:17

it out. It's available in physical format and also ebook

0:19

and also audio book everywhere books are sold. Hi,

0:22

my name is Cliff Weitzman. I am 22 years old

0:24

and I'm a student of Brown University and I'm severely

0:26

dyslexic. What you're about to hear

0:28

is an interview between me and Cliff Weitzman.

0:30

He is the founder and CEO of Speechify,

0:32

which is an app that helps basically convert

0:34

the internet into audio books. 24

0:36

million people using the app. That's a lot of people.

0:39

So you're a CEO of a company that employs 100

0:41

people. That's right. That's wild. I

0:43

don't care what your pedigree is. I don't care if you went

0:45

to Brown or Harvard or Stanford. I don't care if you worked

0:47

at Facebook or Google. I care that you learn fast. You have

0:49

fire in the belly for the product, high loyalty to the team,

0:52

and you're able to ship features fast and move metrics.

0:54

And that's all I care about. If I get to

0:56

learn, there's nothing that's in my way

0:58

that stops me from learning. Damn. That's

1:00

really cool. Okay,

1:02

so for a bit of context, where is Speechify now and how long

1:04

have you guys been working on it? I've

1:07

been working there for six and a half years. There

1:10

is 24 million users who

1:12

use it. There's 100 of us working in the company,

1:14

a little more than 100. There's like 72 engineers. And

1:17

people consume almost 10 billion words

1:19

per month. Bloody hell. 24

1:23

million people using the app. Yeah. App

1:26

and Chrome extension. That's a lot of people. It's a

1:28

lot of people. So you're a CEO of a company that employs 100

1:30

people. That's right. That's

1:33

wild. It's fun,

1:35

yeah. Okay, so we

1:38

left off our previous podcast interview in

1:41

around about 2015 when you're at university. And

1:44

one of the things you said was that you

1:46

were kind of struggling because of your dyslexia to do all of

1:48

the reading that was assigned to you in school as well. And

1:51

at some point you found a way to convert text into

1:53

speech. Yes. And how did that progress into

1:56

where we all know. So, right, I moved to the

1:58

US when I was 13. Didn't speak English. a

2:00

summer reading book before high school was Marley and me. My mom

2:02

had to read it to me. And

2:04

I worked really, really hard in high school. I applied

2:06

to 26 schools, and I got rejected for most of

2:09

them. But I got into this one place called Brown

2:11

University in Providence, Rhode Island. And I love

2:13

this place more than anywhere else in the world. My cheeks started hurting

2:15

from smiling so much three weeks in. But

2:17

we had a summer reading book, Some of the Providence. And I

2:19

spent the entire summer before school trying to learn how to read

2:21

this book. And I didn't finish.

2:24

I finished maybe not even half. And

2:27

I didn't want to be the one kid who showed up in school

2:29

and was not having read his book. So my mom read

2:31

it to me. But she worked, and she didn't have time

2:33

to do the entire thing. So we finished in maybe a quarter, and I had

2:35

a quarter left. And

2:38

before I had this assignment, I was trying to place into

2:40

a higher level chemistry class. And I was going through

2:42

the textbook. And my younger brother, Tyler, is a year

2:44

and a half younger than me. He

2:47

was also interested. So we hooked up this text

2:49

speech system on our computer to read out the

2:51

textbook to us. And Tyler helped

2:53

me crack a Kindle version of this book I

2:55

needed to read, because I didn't have an audiobook.

2:58

And we ran it into my iPhone overnight.

3:00

And it worked, and I listened to it on the plane. And

3:02

that's how I finished the book. And I was like, wow, text

3:04

speech is amazing. And Tyler is

3:07

very talented. He

3:09

started coding when he was in third grade, building Dragon

3:11

Ball Z websites, in fifth grade

3:14

Thomas Assembly, and started hacking video games, MapleStory.

3:16

He'd get four and a half years of math in high school

3:18

at Exeter. And then he studied math as an undergrad at Stanford.

3:21

became president of his fraternity as a second year

3:23

student, dropped out to run a cybersecurity company that

3:25

he founded full time. And he came back to

3:27

finish his math degree, to finish his master's in

3:29

AI, and focused on text-to-speech

3:31

and natural language processing. While

3:34

he was doing that, I was in

3:36

college. I built 36 products, everything from 3D

3:39

printed skateboard breaks, iPhone apps, websites, payments companies.

3:42

And I still

3:45

had trouble reading. And the solution

3:47

was I built an app for my phone that could

3:50

take pictures of physical books and handouts. And it would

3:52

OCR them, optical character recognition. And then it would read

3:54

them to me. And then I built a parser for

3:56

PDFs, where I could upload a PDF and it would.

4:00

run natural language processing on it, and I had to figure

4:02

out how to read that thing to me. And the key

4:04

was I loved listening fast. So

4:07

most people might listen to podcasts like this one at

4:09

2x speed. So I would listen to

4:11

audiobooks first at 1. Well, in the beginning, I didn't

4:13

speak English. So I listened to 0.75, and then 1x, and then 1.25, and

4:15

then 1.5, and then 2x, and then 2.5, and

4:17

then 3x, and then 3.5. So

4:21

I got really good at listening fast.

4:24

Highly recommend the Chrome extension video speed

4:26

controller, because I would 4x speed YouTube videos

4:28

too by 0.1 increments. So I built that for speech on

4:30

my computer. I had a Mac app that I could highlight

4:32

anything, and it would read it to me at 750 words

4:35

per minute. And if

4:37

I was reading through a physics textbook that was pretty

4:39

dense, I dynamically slowed it down to 350 words per

4:41

minute. And the second I got through

4:43

the area that was complicated, I'd go back to 750. And

4:46

if I was reading a Facebook post, I'd listen up like 900 words

4:48

per minute. And so the fact that I

4:51

could dynamically change the speed with a keyboard

4:53

shortcut was life-changing to me. And

4:55

it made it so that I continuously got faster and

4:57

faster and faster at my ability to listen. And

5:00

I wanted to do that for everything. And so

5:02

every time I couldn't listen to something, I would

5:04

code to figure out how to make that thing listenable.

5:07

And my experience in college was I came in, I

5:09

studied renewable energy engineering. So it was a mix of

5:11

physics, engineering, computer science. Brown has a very open

5:13

curriculum. So I took a lot of design classes

5:15

at RISD down the street, the Rhode

5:17

Island School of Design. And I

5:19

did eight hackathons where

5:22

I didn't know how to code. I would jump on a table,

5:24

do a backflip, convince people to work with me. And then I

5:26

would pitch and I won with my team half

5:28

of those hackathons. The problem is after the hackathon, no one

5:31

would keep working on the product. And I

5:33

was like, OK, I got to learn how to code, otherwise I'm not going

5:35

to make any progress. So my second year at

5:37

Brown, I took my first computer science class.

5:39

And I had, you know, it

5:41

was tough. It took me about 13 hours

5:43

to finish the first assignment. Most people took like three

5:46

or five. And the problem was I

5:48

would always misspell variables. So if you are dyslexic

5:50

and you're very bad at spelling and the variables

5:52

don't have the same name, the program crashes. So

5:55

I just brute forced it. Every day I'd come to the computer

5:58

science lab, maybe at 8 AM. like

6:00

eight peanut butter sandwiches, and I'd work there until midnight, sometimes

6:02

two in the morning, and I would do all my other

6:04

homework there. And I did this for maybe

6:06

a month or two, and by the end of that, I

6:08

started to be able to tell when I'd make

6:11

a mistake as a result of a spelling mistake,

6:13

versus a mistake versus an issue with coding. I

6:15

became really good at debugging. And

6:18

then I took two other computer science

6:21

courses, and then I took one on Udemy by

6:23

Rob Percival called the Complete Web Developer Course, and

6:25

another one on iOS App, by Rob Percival called

6:27

the Complete iOS Developer Course, where I built 19

6:29

apps, 24 hours of video, and

6:33

like mini clones of like Instagram, and

6:35

Tinder, and Snapchat, and Google Maps. And

6:37

so I just had a repository on

6:39

my computer of 19 apps

6:41

that I built, and like 15 websites that I built. And

6:43

you're building these as part of this God Udemy course. Yeah,

6:46

that's pretty good. Yeah, and I just did this, instead

6:48

of going home for winter break, I would just sit and I

6:50

would do these courses. And over the summer, I'd do an internship,

6:52

and I would do these courses. And so

6:54

every hackathon I did after, I did maybe 42 in

6:56

all. I

6:59

could just full code that I'd built for other apps, and I'm like, cool,

7:01

you made it when you make an app that does this? Great, let me

7:03

just like Frankenstein something together.

7:07

Towards the time I was graduating, I

7:09

ended up building this tool called findmeskoliships.com.

7:12

And in the US, school is 66K a year.

7:15

I was trying to figure out how to pay for it. And

7:18

I was running all these companies, 3D

7:21

printed skateboard breaks, whatever it might be, to try

7:23

and pay for school. And I realized actually I

7:26

can also apply for scholarships. And most scholarships that

7:28

were very good matches for me, I would win,

7:31

right? Imagine a scholarship

7:33

specifically for people studying renewable energy, or who

7:35

are good at math, or who are Jewish

7:37

from Marin County. And

7:39

I realized that that was the key, find scholarships that

7:42

are unique to you. So then I was like, well,

7:44

it's tough to find good scholarships online. Let

7:46

me hire someone to help me. And so I

7:48

ended up hiring, in total, I

7:50

think I hired about 40 people on

7:52

Upwork from the Philippines, but I whittled it down to

7:54

like 10 who were really good. So I had 10

7:56

freelancers in the Philippines, whose full-time job was to find

7:59

and apply a scholarship. I just made a Google Sheet,

8:01

numbered it 1 to 100, and I was

8:03

like, cool, every scholarship has to be more than

8:05

$5,000, and I uploaded every essay I'd ever written,

8:07

and we matched the right essay, and it

8:10

worked really well. So then I opened it so other people could use

8:12

it too. And I kept

8:14

building these projects, and then

8:16

I graduated. And all my

8:18

friends were getting jobs at Palantir, and Goldman

8:21

Sachs, and McKinsey, and Google. And I was like, I

8:23

don't want to do this. I don't want to work at any of

8:25

these companies. I want to do my own thing. But

8:29

none of the things that I had worked on up until that

8:31

point did, I think, warranted

8:34

the level of ambition that I felt towards

8:36

what I was going to do with my life. OK. So

8:39

you're not thinking that, hell, I can't wait to work on

8:41

findscholarships.com for the next 10 years. So

8:44

Find Me Scholarships, the beautiful part about it is

8:46

I got to know all the people who I

8:48

was working with. And so for example, Nadine from

8:51

the Philippines, her

8:53

day job didn't pay her that much, and

8:55

she really didn't like it, but she's an artist.

8:57

She loves painting on a tablet. And

9:00

I was like, well, why don't you learn graphic design? And she's like,

9:02

oh, I don't know what to do, whatever. So

9:04

I got her an account for Figma. And

9:06

I bought her a course on how to do Figma, and

9:08

she started learning. And then I helped her find some jobs

9:10

as a graphic designer. And there were some, and also I

9:12

taught them how to do auto layout for iOS. And someone

9:15

else I taught how to scrape a JavaScript. And I was

9:17

like, you know what? All these people outside of the United

9:19

States, they are smarter people in the

9:21

United States. They just don't have access to these jobs.

9:23

And they don't have access to the one or two

9:25

things that you need to learn how to do. And

9:27

if you knew how to do it, you'd crush it.

9:29

So you have a Skillshare course, I think, for how

9:31

to use Premiere. Amazing. Anyone who learns how to use

9:33

Premiere from that course can go immediately make $25 an

9:35

hour being a video editor. Let me teach

9:37

people the basic skills and then help them get jobs.

9:39

So that I was excited about, but

9:42

it was so much work to manage so many people

9:44

remotely. It was a big, big headache. And

9:49

I wrote a 30-page paper about my world views

9:52

and started thinking and figured out the 25 things that

9:55

I believe that most other people don't believe. And I

9:57

read a ton of biographies. And I started reading S1

9:59

filings. S1 Finance, what are those?

10:30

S1 Finance, what are those? S1

11:00

Finance, what are those? S1

11:30

Finance, what are those? And

12:00

it started to be the case to this

12:03

day, 10 to 15% of our

12:05

reviews on the App Store are people who say they

12:07

genuinely started crying when they use speechify because it

12:09

solved such a big problem for them. And then I

12:11

started even before that visiting schools for kids with learning

12:13

differences. And so that cohort

12:16

loved the product. And as the

12:18

product got better, we started having not only people

12:20

with dyslexia or vision issues or ADHD, but

12:23

people who had autism, or they

12:25

were a second language learner of English, or they had a

12:27

concussion. And then we started

12:29

having normal professionals who didn't have any

12:31

learning difference whatsoever, doctors, lawyers, accountants, people

12:33

in the military, executives, people in finance.

12:36

And it's just because the product

12:38

kept getting better and better and better and better. And

12:42

so with time, the next step was building

12:44

the theme. So that was building the product, right?

12:47

And two and a half years in, I still struggled

12:49

with acquiring users and getting

12:51

people to use it enough. But

12:53

if you grind in one direction, so this is the

12:55

thing I learned about building all these products in college,

12:57

right? I built maybe in total about 36, and I

13:00

took classes in every department you can imagine, right? Biotechnology

13:02

and medicine, you

13:04

know, electricity and magnetism, and

13:07

I tried different fields. You

13:11

don't make progress like this going in a bunch of

13:13

different directions. You make progress like this going in one

13:15

direction. And when you hit a brick wall and you

13:17

crumple, you pick yourself back up, you go back, you

13:20

don't go in, you go again against the same brick

13:22

wall over and over and over and

13:24

over again until it cracks. And then you keep running

13:26

like Mario, and there's another wall, and you

13:28

crash through it again. Because

13:30

those brick walls you're willing to run

13:33

through, other people are not willing to run through them. And

13:36

the thing is, text-to-speech has been around since the

13:38

1960s. It predates the internet, but it's always socked.

13:41

And I was like, I'm just going to make it really good.

13:43

And cool. So when we passed that, it was

13:45

a matter of how can we get the best people in the world to

13:47

work at Speechify. And so when we

13:49

were 25, 28 people, 18 of

13:51

the folks in the company previously were

13:54

either CTO, CEO of VP of Engineering at the company's

13:56

app beforehand. And you meant a bunch of them. Okay.

14:00

So that's been a whistle-stop tour of how we kind

14:02

of got to that point. I'd like to rewind the

14:04

fuck back a bit. So these... Okay,

14:07

firstly, learning to code. To what extent is

14:09

it useful today for someone to learn how

14:11

to code? If you want to be a

14:13

founder in 2022, you have to learn how to code. You

14:17

don't have an option not to. Here's the things that

14:19

are important to consider. Number one, you do not need to

14:21

be good at math to be good at coding. Number two,

14:23

you don't need to be good at spelling to learn how

14:25

to code. Number three, you don't need to know English or

14:27

be good at English to know how to code. Given

14:30

that, there's no reason for you to learn how to do

14:32

it. You don't need to be the best, but

14:35

you need to understand it. Okay. So

14:38

if you want to manage engineers or have engineers

14:40

work with you, you have to be able to

14:43

know what TypeScript is, what Node.js is,

14:45

what React is, right? What do you

14:47

use Erlang for? Whatever. Memory

14:49

management, you just need to be able to use these words, number

14:51

one. Number two, there's

14:54

so many good mental models that come

14:57

from coding, right? First in, first out,

14:59

you know, topography, math, map,

15:01

like find the highest point. It

15:05

helps so much in decision-making and strategy. And

15:09

people can just bullshit you about how

15:11

long something will take or what if you don't know

15:14

how to ask the right questions. So

15:16

if I interview to be an engineer at Speechify,

15:18

I get rejected. There's no way I pass the

15:20

programming test. But it's so

15:22

important that I understand what's happening. And

15:26

again, all you need to do is

15:28

take a course or two that'll take no

15:30

more than 30 hours from your life to just

15:32

understand what's going on. And that's

15:34

it. And do it with a friend, you know, find a

15:36

way of committing. You don't need to have a degree in

15:38

it, but you need to play around with it. Yeah. Yeah,

15:40

I find that that's still, you know, I still get emails

15:42

every day from people being like, how do

15:44

I make money on the internet? And it's so hard

15:47

to say anything other than like, look, learning how

15:49

to code is a really, really good first step.

15:51

There are a lot of people that try go

15:53

down the creator route and sure, that's fine. But

15:55

I think the expected value of learning how to

15:57

code is just way bigger, way, way, you know,

15:59

the. The payoff for everything in

16:01

your life is just way higher than let me

16:03

try and make some gaming YouTube videos Which is

16:05

an area that a lot of people can go down and

16:07

competing with like a zillion other people doing the same thing

16:09

Even being a creator I'll give you

16:12

an example like when I was a kid I used to

16:14

make parkour videos for YouTube all the time So I got

16:16

really good at iMovie and you know some other editing software

16:19

But I don't use tiktok that much as a consumer but

16:21

I want to make tiktok videos because I think that it would

16:24

be a great advertisement for speechify and And

16:27

So I don't want to use tiktok on my phone, but cap

16:29

cut is an amazing editing software for tiktok I haven't learned how

16:31

to use cap cut yet And I've had it

16:33

on my phone for at least a month and I'm actually genuinely

16:35

frustrated at myself that I haven't taken The time to sit down

16:37

for an hour and just learn how to use cap cut So

16:40

that's probably gonna happen in the next like month

16:42

Yeah And here's when you know that I'm finally

16:45

gonna sit down and Sacrifice not

16:47

talking to the team not hiring more people

16:49

not spending more on ads whatever not Learning

16:52

something new that I haven't It's what

16:54

I wanted enough And so if you want

16:56

to make money on the internet and you haven't taken the 30 hours to learn

16:58

how to code You just don't want it enough and

17:01

this is coming from a person who is literally dyslexic Is my

17:03

first language is not English right? I do not have a

17:06

high aptitude for programming to begin with You

17:08

just got to learn how to do it And so the

17:10

same thing is true for learning any skill on the internet

17:12

using figma using Photoshop Whatever it might be you got to

17:14

first learn how to use the tool and then implement it

17:16

to your advantage You mentioned

17:18

that learning how to code is basically a requirement for people who

17:20

want to be founders Are there any other categories of people that

17:23

you think learning how to code would be would be helpful for?

17:26

It's helpful for everybody. It's

17:28

not required for everybody as a

17:30

founder You have to know it because technology

17:32

is so core to building an exponential growth

17:34

product And you're gonna have

17:36

engineers who work for your company and so you just you

17:39

need to understand what's going on Computer

17:43

science is just incredible because it has

17:45

so many tool sets for how to

17:47

think So when I was

17:49

in high school, I was really

17:52

interested in economics, right? Keynesian economics

17:54

Hayek, Mises, Malthus I

17:57

was so I read so many books about

17:59

it and I I read so many books, listen to

18:01

so many books, about

18:04

philosophy and business and theology

18:07

and all this kind of stuff. They

18:10

taught me things like gift and goods,

18:12

right? Diminishing marginal returns.

18:15

What happens when the PPF curve shifts to the right? All

18:17

this stuff is key to how my brain just operates on

18:19

a daily basis. And it lets me explain

18:21

things in my head to myself that otherwise I would

18:24

not be able to explain. Computer science

18:26

gave me like an equal amount of

18:29

data structures and algorithms for how to think. And

18:33

that is key for making better decisions and being

18:35

able to use more information. So if you wanna

18:37

think about your brain as

18:40

a machine, right? It has a certain amount of

18:42

processing power. And so you can either upgrade the

18:44

hardware or you can upgrade the software. So I'd

18:46

say that I have an okay brain, right? The

18:48

place where, you know, people have, you know, IQ,

18:50

intelligent quotients, EQ, emotional intelligence, AQ,

18:52

adversity quotient. The place where I break the

18:55

scale is adversity quotient. I like never give

18:57

up in places where other people would. My

19:00

IQ is good, it's not great. I

19:04

would argue. But you can upgrade your software. So

19:07

that's one way of upgrading the software of your

19:09

brain. It just teaches you new methods of thinking.

19:11

So anyone who needs to make decisions on a

19:13

regular basis, highly recommend at least spending 20, 30

19:15

hours studying the basics

19:17

of computer science. Fantastic. This

19:19

episode of Deep Dive is very kindly being

19:21

sponsored by Hostinger. Now, if you're listening to

19:23

start a business or develop your personal brand in 2024, then

19:26

you're gonna need a website. But the question of

19:28

where to start is a question that I get

19:30

asked all the time. So if you've ever wanted

19:32

to set up a website, but you've had a

19:35

question of where to begin because there's all these

19:37

different options out there, then Hostinger has literally everything

19:39

you'll need. Hostinger is a top global website hosting

19:41

service with servers all around the world. It's fast

19:43

and reliable and with over 2 million users, it's

19:45

becoming one of the fastest growing web hosting services

19:47

out there. We've recently moved our website over to

19:49

Hostinger to host my personal website. And not only

19:51

was the transition completely seamless, but the tools

19:53

were actually really simple to use as well. And I really wish

19:55

something like this had been available when I was first starting my

19:57

business like 12 years ago. If you're new to web design. and

20:00

then they've got everything you need to make a

20:02

professional looking website. One of their really interesting features

20:04

is their AI website builder, which can help you

20:06

make a custom website in literally seconds, and a

20:08

whole suite of other really useful AI tools, including

20:10

a logo creator, an image generator, and a heat

20:12

map tool as well. It's super easy to use

20:14

with a drag and drop editor for simple customization,

20:16

and you don't need any coding or technical knowledge

20:18

at all. Hostinger comes out to less than $3

20:20

a month, which includes a

20:22

free domain name, so it's super affordable. And if

20:24

you use the link in the video description or

20:26

in the show notes, which is hostinger.com/Ali Adal, and

20:29

if you use the code Ali Adal in all caps at

20:31

checkout, then you will also get 10% off. So

20:34

thank you again to Hostinger for sponsoring this episode. And

20:37

so these products that you made, these 36 different products,

20:39

was that kind of post learning how to code, or

20:41

were you trying to build stuff on the internet pre-learning

20:43

how to code? Pre-learning how to

20:45

code. So the first thing that I

20:47

built in high school was a company

20:49

called Cliff's Coupons, and then I built

20:52

a pressurized air cannon, it's like DIY

20:54

type thing, and then I

20:56

built a 3D printed skateboard brake, and then I built a

20:58

mini Faraday cage that you iron onto your pocket to

21:01

block cell phone radiation from affecting reproductive organs. And

21:03

then I built a

21:05

iPhone app called Starter

21:08

Pack, and that's the point where I started

21:10

using coding. What got you into building all

21:12

this? Because most kids wouldn't be

21:14

building stuff in high school. Like what was it about

21:16

you that made you go down that route? So when

21:18

I was young, my

21:21

dad would come home after work, and

21:24

maybe we'd eaten dinner or something, but we'd just

21:26

play math games, right? And

21:28

he'd ask me and Tyler, five plus three times

21:31

20 divided by four, whatever. And we'd play chess,

21:33

and he'd say, let's go on a walk. And

21:35

we'd be like, great, we'll go on a walk.

21:37

And on the walks, we'd ask him questions, and

21:40

he'd just teach us about the world. How

21:42

does the stock market work? What's the difference between a mutual fund and

21:45

a hedge fund? Why is the sky blue? And

21:47

I really admired my dad growing up, and I still

21:49

very much admire my dad. But

21:52

I think I got a very deep understanding of

21:55

how the financial world operates as

21:58

a kid. And I

22:01

realized very quickly that if I wanted to

22:03

have freedom, it's good to

22:05

have a source of income. And

22:09

then I read the four hour work, Greek, Pits and Ferris, I

22:11

think when I was 17 or 18. And

22:14

I was like, wow, I would love to build a

22:16

Muse company. So my goal is I'm gonna build something

22:18

that makes me 300k a year, whether I work or

22:20

not. And so I was working towards

22:22

succeeding in doing that. And I started building stuff even

22:24

before I read that book. But that book, again, gave

22:26

me a bunch of mental models and how to think

22:29

about it that were very useful. And

22:32

so all that stuff inspired me to make stuff. And

22:34

then naturally, I'd just like, I'd MacGyver things with bicycles

22:36

when I was a kid. I always like making tinkering.

22:40

And I think that even more

22:42

than entrepreneur, I am an inventor. I love

22:44

inventing things. And

22:47

the word entrepreneurship actually comes from a Latin

22:49

root meaning to elevate economic resources from an

22:51

area of low yield to an area of

22:54

high yield. Effectively, you're creating value where there

22:56

was none before. And if

22:58

you think about Emerson has a great essay about this called

23:00

On Wealth. The person who made the steam

23:02

engine, right, there was a bunch of like scrap metal, Thomas

23:04

Avery, in his garage or whatever

23:06

workshop. And he applied his thought to matter, rearrange

23:08

atoms in a certain way and then elevated those

23:10

resources and multiplied them times a thousand and a

23:12

million. And now you can use the steam engine

23:15

to like pump water out of wells and create

23:17

electricity and move trains across the country. That's

23:20

the closest thing humans can do to magic.

23:22

So I like creating magic. And

23:24

so you do that by inventing and then by bringing your

23:26

dreams into reality. And so

23:29

there's a portion which is like being the technologies, technologist,

23:31

and coming up with an idea, making it work. And

23:33

the second one is how do you get people to

23:35

actually use it? For your college,

23:37

you've built a few things in the past

23:39

and then you decide I need to learn

23:41

how to code. And then at that point,

23:43

you take this Udemy course on web development

23:45

and iOS development. And now that really accelerates

23:47

just your ability to make stuff. Yeah. And

23:50

then you put into practice all these hackathons that you're going to. And

23:52

so presumably when you have the idea for like, hey,

23:55

text to speech is interesting. It gets the gears

23:57

grinding in your head in a way that if

23:59

you. who weren't familiar with coding computer science and stuff,

24:02

you'd probably have thought like, most other dyslexic people out there. I'd

24:04

be too scared of it. Yeah, but this is kind of hard.

24:07

But as soon as you have that knowledge and you're like, hang on,

24:09

this is, there is a possibility that

24:11

I could make a text to speak and speech engine and you

24:13

were kind of hacking away at it yourself. You

24:15

start thinking that, okay, could this turn into a company someday?

24:18

Yeah. That's exactly right. And

24:20

so I decided, you know what? I'm not gonna take a job.

24:22

I'm gonna stay at Brown. I'm going to stay

24:24

here as a visiting scholar. I'm

24:27

not gonna get paid any money. And it's

24:29

even like a little bit embarrassing, right? Your parents are like,

24:31

what are you doing? Parents, friends are like,

24:33

so is Cliff unemployed? Like what's he doing? And

24:35

I'm like, even if I fail,

24:37

that's fine because I can always go and try to

24:39

get that job at Google. Let's say my parents are

24:41

like super embarrassed, right? I can't sleep at home. Okay,

24:43

I have probably like 15 friends who let me crash

24:45

on the couch every once in a while. So I'll

24:47

circulate between them. But I would just be sad

24:50

going directly to take one of these jobs. And

24:52

so my thing was, okay, I'm either gonna find a

24:54

company where I really admire the management team and I'm

24:56

gonna work there. And I applied to a couple of

24:58

these companies that just didn't get in at the time.

25:02

Or I'm gonna do my own thing. I said, okay, I'm

25:04

gonna do my own thing. And then I was really excited

25:06

about Naira applications of deep learning. So I

25:08

was like, cool, I'm gonna start building this stuff. And then

25:10

I thought really good at user acquisition, right?

25:13

Running ads on YouTube and Facebook and Instagram

25:15

and building products that people share

25:17

with each other. How did

25:19

you get good at that? Like pre-Speechify? I

25:22

got good at it during Speechify. Okay, right. So

25:24

my thesis is your number one role as a founder is

25:26

to learn. You learn how to code, how to design, how

25:28

to talk to users, how to recruit, how to do whatever.

25:31

As a CEO, you have three roles. Make sure there's enough money

25:33

at the bank, set the vision and put the right people in

25:35

the right seats. But

25:38

at this point and even today, I'm just

25:40

always learning new things. And my system for learning is I

25:42

read literally a hundred books on the topic. And

25:45

then I talked to every expert there is and then I rewrite the

25:47

playbook from scratch. So for ads, what I did is I read all

25:49

the books that I could. And then I

25:51

made a list of the top 100 best performing consumer subscription

25:53

companies in the world. And then

25:55

I flew around the world and I

25:57

met all of them. And I sent them emails.

26:00

I sent some Facebook messages and Instagram messages

26:02

and LinkedIn messages and I just went and

26:04

I got to know the founder of Audible,

26:06

the founder of Grammarly, the founder of Reflectly,

26:08

the founder of Lightrix, the

26:10

founder of Lululemon, Richard Branson, whatever it was.

26:13

And that's the CMO of Airbnb, the CMO

26:15

of Netflix. And I would just spend time

26:17

with them. And in some cases, I would

26:19

literally go into the office, sit behind them

26:22

and saw how they bought ads. And then

26:24

I would go in and do the same

26:26

thing. And then I wanted to get good

26:28

at creating content. So someone happened to introduce

26:30

me to Ammar from Yes Theory at a certain point when

26:33

I was living in LA. And this is

26:35

before I got really into ads. And Ammar and I

26:37

became like instant best friends. We started going on a

26:39

lot of adventures together. We were very similar people. And

26:41

then Ammar introduced me to Logan. And then Logan and

26:44

I started doing a bunch of adventures together. And he

26:46

helped me shoot one of my ads. And

26:48

then we met through

26:51

Valentin. And then I met Eric

26:53

because he was a big user of Speechify. And then Eric

26:55

introduced me to Mr. Beast. And then I went and I

26:57

slept in Timmy's house for like a week and

26:59

saw how James operates and Tyler and the entire production

27:02

team. And at a certain

27:04

point, what I did is I made a list of the

27:06

top 100 best performing ads in history. And

27:08

I literally sat in my house in LA. And I

27:10

remade all the ads. And someone that did really well

27:12

and someone that did really bad. Same

27:15

thing as coding. I just put in the time point. And

27:19

that's it. Now, OK. So someone

27:23

listen to this. I've

27:25

kind of heard bits of your story before. We've

27:28

talked a bunch of times. I'm always just absolutely floored

27:30

by just A, the

27:33

tenacity and just the going overboard

27:35

that you do on everything. And

27:37

B, just like how casually you talk about it. You're like,

27:40

oh, yeah, I wanted to learn about ads. So I just

27:42

like Googled the top 100 best performing consumer companies.

27:44

And I just emailed them all and said, can I just hang out with you?

27:46

And I just flew around the world hanging out with them.

27:49

There are so few people that I know who

27:51

would be thinking in those kind of ways. They'd

27:53

be thinking, I need to learn how to get good ads. Let me

27:55

watch a YouTube video. OK, maybe

27:57

there's a course. But it's $300. to

28:00

this course, oh, I don't know if I can, oh, you know what, screw

28:02

it, let me just screw around myself. What is

28:04

it about you that, how did you get to that point where

28:06

you're literally sending 100 emails and going around the world to hang

28:08

out with these people? I can actually answer this question really well.

28:11

Okay. Let's

28:13

take us back to 2019, before 2020. 2020

28:15

is when I emailed all these people. 2019, I did

28:17

not have any money, and

28:19

I met this wonderful woman, Jennifer

28:22

Sindel, who I love dearly to this day, and

28:25

her daughter, Anna Sindel, is also dyslexic, and

28:28

Anna started working with us at Speechify, and

28:30

they offered me that I could live out of their guest house. And

28:33

so I was living in this guest house in Palo Alto,

28:35

paying no rent, trying to make this company work, and Anna

28:37

was using Speechify, and she started helping me work on it,

28:39

et cetera. I started studying

28:41

SEO, studying ads, taking courses,

28:44

figuring it out, and I was not making the

28:46

progress that I wanted. God damn it, Ali, I

28:48

was trying, I was not succeeding, and I was

28:50

so frustrated that I was not succeeding. So what

28:52

I ended up doing is I pulled up Google

28:54

Sheets, and I made a calendar of

28:56

every single hour in every single day, and

28:58

I retroactively, and moving forward, wrote what I spent

29:00

my time on. Was I sleeping? Was I taking

29:02

a walk? Was I driving somewhere? Was I eating?

29:05

Was I studying SEO? Or was I implementing SEO?

29:07

Was I coding? Was I doing whatever? And even

29:09

when I worked on the first versions of the

29:11

app, I committed that I would spend five hours

29:13

a day coding in Xcode, and I measured it with rescue

29:15

time. And if I did not spend five hours a day

29:17

coding in rescue time, by 10 a.m. the next day, I

29:20

had to do 300 pushups and 200 pullups. And

29:23

I had a group chat with Simmermangit, Max Deutsch,

29:25

Valentin Perez, and Tyler Weizmann, and I would tell

29:27

them if I didn't do it, and if I

29:29

didn't do the consequence, I had to run 10

29:31

miles that weekend. And

29:35

I kept not making progress. And

29:37

so I was like, cool, I've taken the courses. I'm still not

29:39

where I wanted to be. Most people maybe at that point just

29:41

stop. I was like, no, I can't stop, I've used

29:43

it. So I messaged people who

29:45

could help me. Sometimes I knew them, sometimes

29:47

I didn't. And so in the case

29:49

of, okay, well, then how did you find the, first

29:52

of all, how did you even find a list, Cliff,

29:54

of the top 50 consumer subscription companies in the world,

29:56

the top 100? It's

29:58

like I found the scholarships, or everything. There's a

30:00

website called Sensor Tower that ranks

30:03

iOS apps by their revenue. And

30:05

so I just hired a bunch of Philippine-based

30:10

virtual assistants to just index it for me. And

30:13

I used maybe seven different sources, and then I

30:15

made a list. And then I asked them, okay,

30:17

find the CEOs and the heads of growth and

30:19

the heads of marketing on Instagram, and

30:21

then put them in the Google Sheet. And then

30:23

I want you to include the link to their

30:26

Instagram, LinkedIn, Facebook, Twitter, and

30:28

their email. And then I made a list by

30:30

each one. So now I have three people from every single company,

30:32

and I have five sources of contact. And

30:35

then for each one of those people, I

30:37

put three rows, reach out number one, number two,

30:39

number three. Did they respond back? And

30:41

that entire sheet is full. Every

30:44

single person got three messages,

30:47

at least, from me. And

30:49

then eventually they'd respond. And I'm okay with

30:51

people not responding to me because I'm always

30:53

polite. I'm honest. I'm direct. I'm concise. I'm

30:55

not sending these messages.

30:57

What's the last cold email that I sent? Let's

31:00

say I'm messaging you. Sure. Be

31:03

like, hey, Ali, just watch your podcast

31:05

on Brandon Sanderson. Kaladin

31:07

is my favorite character I've ever read about,

31:09

period. I really admire the way in

31:12

which you interview your guests. I

31:15

think you're extremely thoughtful

31:20

and well-researched. I'm considering starting

31:22

a podcast. So far,

31:25

I've interviewed the president of the World Bank

31:28

and seven teal fellows.

31:32

The person I most want

31:34

to interview is Derek Sivers.

31:39

Could you be free for a 10-minute Zoom call? Friday

31:42

at 3 p.m. or 4 p.m. or

31:45

Saturday at this time or this time? My

31:49

best, Cliff. Yes. Before this,

31:52

I won Harvard's Hackathon, Stanford's

31:55

Hackathon, and MIT's Hackathon. And

31:57

I'm studying renewable energy engineering at Brown. So

32:01

that email shows that I actually follow

32:04

you. It shows what I admire

32:06

about you genuinely. It's relatively short. It

32:08

still shows that I have credibility and what I'm doing towards

32:10

it. And when you read the email, you're like, okay, he's

32:12

not like just

32:14

some kid on the internet. Like he's working hard.

32:17

And if I didn't have those accomplishments, I would say the same

32:19

things I would say at high school. Hey,

32:22

Ali, my name is Cliff Weissman. I'm a

32:24

junior at Redwood High School. I'm the president

32:26

of the Speech and Debate team. I recently

32:28

won the state competition for Sport

32:30

Debate. I'm the president of

32:32

the Jefferson Awards Community Service Club.

32:36

And you

32:38

changed my life. You

32:41

are the person who made me learn how to

32:43

edit videos. And I have now made 300 YouTube

32:45

videos. I

32:48

only have 500 subscribers, but

32:51

I want to be like you when I grew up. Would you be

32:53

free for a 15-minute call this

32:55

time at this time? Here's

32:58

my two questions. And

33:00

if you just send a message that is well-composed,

33:02

where you show that you've done work, here's the

33:05

thing that people love. And

33:07

I was just talking to my brother about this. We

33:09

have a friend, Nima Bardi, who's the CEO of Atlas.

33:13

I love talking to Nima about his business. He's

33:15

worked on it for maybe two years. I

33:17

give the kid, I give Nima an inch,

33:19

he takes a mile. Like if I give him a

33:21

piece of advice, by God, a week

33:23

later, he's done what I said. And

33:25

even if he didn't originally believe it, he'll

33:28

still do it, and I can see the work. You know

33:30

how satisfying it is to give someone a piece of advice?

33:32

You do, to give someone a piece of advice, and they

33:34

follow it. Any person who's ever

33:36

given me advice in my life who I thought

33:38

their advice, I have implemented what they asked to

33:40

the letter. And then I came back to

33:43

them a week later, I was like, hey, I did this, here's the

33:45

results. It's so fun. And

33:48

then they get validation, either it works, or

33:50

actually this is how the world is today,

33:52

great. It's useful for me that I'm

33:54

talking to Cliff, because he's validating all my ideas in the field. So

33:57

I send these messages, and now I get on the Zoom call.

34:00

in the Zoom call, in K, where in the world are you,

34:02

blah, blah, blah, we get on the call. Part

34:04

of the call, they realized that I've been listening to 100

34:06

audiobooks every single year for the last 12 years. And

34:09

they'd be like, oh, have you ever read whatever?

34:11

And they'll mention some obscure book. And I'd be like, yeah. And

34:13

then we'll start a conversation about that book. And they'll be like,

34:16

wow. And at the end, I'll be like, you know, I'd love

34:18

to come visit you in Copenhagen. And they're like, sweet. Well, if you're

34:21

ever here, let me know. And I'm like, great.

34:23

How about Saturday? And they'd be like, what do

34:25

you mean? I was like, I'll come visit you

34:27

on Saturday. And it's like, okay, cool. Cool. And

34:30

I'd be like, is your office sweet? I'll book the hotel

34:32

nearby. I'll fly to Copenhagen. I'll fly to Israel. I'll fly

34:34

to wherever they are. I

34:36

went to Richard Branson's island, Necker Island. And I spent

34:38

three days with him there. And he like taught me

34:40

a bunch of really good things. We

34:43

talked about Lululemon at the beginning of before we started.

34:45

Chip Wilson, I read his biography. I thought he was

34:47

amazing. It took me four years to get him to

34:49

meet with me. And he said, okay, we'll meet. Great.

34:52

I flew to Vancouver. I was in Miami at the

34:54

time. It takes 20 hours to get from Miami to Vancouver. And

34:57

I went there for 14 hours. And

34:59

we did a like four hour hike. It

35:01

was amazing. And then I flew to San

35:03

Francisco and back to Miami. If I get to learn,

35:06

there's nothing that's in my way that stops me from learning.

35:09

And so, okay, so the thing that I do is I, you know, programmatically

35:11

figure out who I need to reach out to. I reach

35:14

out to them very politely, very

35:16

patiently. And I'm like very respectful of the

35:18

time. And then I suck the marrow

35:20

out of it and do everything that I can from

35:22

learning. And then once I learned

35:24

how to do it really well, I teach someone else on the team how to do

35:26

that. And then I move on to the next thing I need to learn how

35:28

to do. Damn. That

35:31

was really cool. Someone

35:33

might be thinking this and saying, well, easy for you, Cliff. You

35:36

had all this money to fly around the world to meet

35:38

people. I'm stuck in my house in my parents' basement. I

35:41

don't have the money. Like, well, what

35:43

advice would you give to that person? Well, first of all, I

35:45

didn't have the money. I

35:48

never ever, ever spend money on a credit card you

35:50

do not have. But I took out a Chase Sapphire

35:52

Preferred credit card that gave me 50,000 points when I

35:55

used it for the first $3,000. And I had to buy a

35:57

new computer anyway. So I just bought the new computer. Boom,

36:00

I have points. I can fly wherever I want. Right?

36:03

So that's it. And

36:07

with time, and by the way, if you have

36:09

a business and you're running ads, run them on

36:11

credit cards, you'll get unlimited fly points. So

36:15

it has nothing to do with money. And even like I

36:17

didn't need to go in person. I could learn 80% on

36:19

Zoom. Great.

36:23

So that's what you do. Just do Zoom calls with people.

36:26

Yeah. It strikes me that like I get a bunch of

36:28

these sorts of emails. And I'll see the notification on my

36:30

Apple Watch and be like, yep,

36:32

and within like five milliseconds I can tell,

36:34

is this a copy-pasted email or is this

36:36

someone who's genuinely taken the time to write

36:39

a personalized email? Absolutely. And it

36:41

seems like a bunch of my other creative friends get this.

36:44

I'm sure you do as well, where these

36:47

marketing bros, they're probably doing some kind of course that's like,

36:50

hey, Ollie, I checked out your content on olliebzell.com and I

36:52

realized that your articles could be better. And

36:54

it's just so obviously copy and pasted. But

36:56

if I ever do get an email where someone has actually taken

36:58

the effort to even write just a few lines and let me

37:00

think, oh, this person actually knows my

37:02

content. Absolutely. They've actually been following me for a

37:04

while. Now I'm far more likely to respond to that, especially

37:07

if they make it super easy to say yes. Correct. Then

37:10

it's just almost a no-brainer at that point. Every

37:12

email I've ever gotten where someone mentioned five

37:14

different pieces of content that I made and

37:17

showed me what they learned from those peepses and then asked me

37:19

a question, I will always answer that question. I'm sure the same

37:21

is true of you. Yeah. Right? I'm

37:24

putting the time to show that you've actually been following

37:26

and I'm very glad to interact. And by

37:28

the way, sometimes I'll miss it. But

37:30

if you politely follow up, as I do,

37:32

great. Yeah. So

37:34

programmatically figuring out, hiring

37:37

people in the Philippines to scrape the internet for

37:39

contact details from five different platforms for a hundred

37:41

different people, and then you're reaching out to all

37:43

of them. And you're doing all of this while you're trying to build

37:45

your company. I mean, this is how I build

37:47

my company, right? You got to learn

37:49

how to do the thing. So yes, in

37:51

the same time, we've got the team who are

37:54

designing, who are programming, and I'm

37:56

responsible to bring the users. So I got to learn how to bring

37:58

the users, right? Learning is your job. So

38:00

I'm learning. Even

38:03

now, I mentioned

38:05

this thing about CapCut. So I had a call right before

38:08

this. I'm still doing this. I'm

38:10

obsessed with learning how to run ads

38:12

on TikTok profitably. TikTok

38:14

is not yet a platform that works really well for us.

38:18

Every company that is succeeding in

38:21

TikTok, the CEO, the president, the head of growth, they've

38:23

all received emails from me. And if they responded, I've

38:25

had a call. If they haven't, they're going to keep

38:27

receiving emails until I get a call. And

38:30

if those don't work, I will find a friend who knows them

38:32

and get them to introduce me, which I've done many times. I've

38:34

had calls with the CEO,

38:37

a show who's the CEO of TikTok

38:40

outside of China. I've been introduced to him. Everybody

38:43

down the chain. So I found this guy who

38:46

is an amazing company, smaller than Speechify,

38:48

but he's crushing it on TikTok. I

38:51

have a weekly call with him now where I help him

38:53

with Facebook and he helps me with TikTok.

38:55

I've seen five creators, not creators, entrepreneurs

38:58

who have companies that are smaller than Speechify where

39:01

they figured out TikTok and I'm regularly on calls

39:03

with them learning how to crack TikTok. And you'll

39:05

see that in 2023, TikTok will become

39:08

one of our number one acquisition channels. But

39:10

I just need to learn how to do it. Now,

39:13

this episode of Deep Dive is being very

39:15

kindly sponsored by Trading 212. Now,

39:17

investment in trading can seem pretty complicated, especially if

39:19

you're a beginner. I've made a bunch of videos

39:21

about this. And the thing I always recommend is

39:24

to just invest in low cost, broad stock market

39:26

index funds, which you can actually do completely for

39:28

free with Trading 212. It's a

39:30

great app that lets you invest in stocks and shares

39:32

and funds completely for free. And what I love about

39:34

Trading 212 is that it's also got a bunch of

39:36

really helpful features that especially if you're a beginner to

39:38

the world of investing, it can help you kind of

39:40

learn the ropes without actually putting money at risk. You

39:42

can actually trade initially with fake money and you can

39:44

see how you would have performed without actually putting real money

39:47

on the line. And then once you're ready to put real

39:49

money in, you can always switch to the real trading interface.

39:52

And the other feature that I really like is

39:54

Pies and auto investing. So the Pies are different like

39:56

allocations of stocks and shares and funds that various people

39:58

on the internet have come up with. with and

40:00

some of them are even like professional finance people

40:02

and then if you want you can just copy and

40:04

paste someone else's allocation of funds and stocks and shares

40:07

into your own account so you can benefit from other

40:09

people's experience without actually having to do any research for

40:11

yourself. Obviously the thing I always recommend is to

40:13

just invest in index funds but if like I

40:15

do you like to play around with like five to

40:17

ten percent of your portfolio with individual stock picking the

40:20

PIs and auto invest feature is great for that. There

40:22

is also support for multi-currency accounts which is absolutely amazing

40:24

so if like I do you invest loads of

40:26

money in the S&P 500 that is a US

40:28

fund and normally when you

40:30

invest in US funds at least from here in the

40:32

UK you're hit with all sorts of foreign exchange fees

40:35

and all that stuff and so the fact that Trading

40:37

212 has multi-currency accounts means that you can trade without

40:39

getting hit with all these various fees. If you have

40:41

an invest or an ISA account then Trading 212 gives

40:43

you a bonus as well which is interest on uninvested

40:45

money. The interest is paid out every day and you

40:47

have full flexibility. There are no minimums or ceilings and

40:50

you can withdraw the money literally whenever you want. They've

40:52

also just launched a new offer of 5% yield on

40:54

GDP, 4.2% on euros and 5.1% on US dollars and

40:56

many more currencies through qualifying

41:01

money market funds. With over 24,000

41:04

reviews on Trustpilot, Trading 212 is an amazing

41:06

platform to get started with investing so if you click the

41:08

link below in the video description or in the show notes

41:11

that will take you a page which will give you instructions

41:13

on how to sign up and when you sign up you'll

41:15

also get a free share which is worth up to 100

41:17

pounds so it's free money and there's no reason not to

41:19

do it. Thank you so much Trading 212 for sponsoring this

41:21

episode and let's get back to it. What was

41:24

the journey of speechify in the very early days when it was

41:26

just you? Yeah because we kind of blitzed over and sort of

41:28

like you and then there's 28 people like what was the

41:30

you and then how did how did that progress?

41:32

So I started working on it. I

41:35

graduated university next. I

41:38

started flying around the country to

41:41

go to conferences for people learning differences. The first one I

41:43

went to was happened to be in Florida. I didn't

41:47

have the money for the conference. I didn't have

41:49

the money for the hotel so I didn't buy

41:51

a ticket. I emailed every single person on the

41:53

board to ask for a free ticket. No one

41:55

gave me one so I rented

41:57

this very small Airbnb like 15 minutes away. I

42:00

took an Uber there and I sat in the lobby until somebody

42:02

gave me a ticket. They

42:04

were just like, yeah, I'll say I forgot my ticket,

42:06

whatever. I went in. Keynote

42:08

speaker finished speaking. I look around. There's

42:11

like a couple of thousand people in the audience. I jump from the

42:13

stage. I plug in my computer and I demo speechify and no one

42:15

takes me off. And when I got off, 12 school

42:17

heads offered to fly me to their schools and you show the kids how

42:19

to use speechify. Wait, hang on. So the

42:22

keynote speaker finished speaking and you just jumped on stage? Yeah,

42:24

because the event was – that talk was over. People

42:27

were going to go to lunch. So

42:29

I jumped on the stage like, hi, my name is Cliff Weitzman.

42:32

I am 22 years old and I'm a student at Brown

42:34

University and I'm severely dyslexic. And my number one issue is

42:36

the fact that when I get handed out to school, I

42:38

can't read them. So I built a tool

42:40

called Speechify that does this for me. Let me show you how

42:42

it works. And then I plugged in my computer and I demo speechify.

42:45

And then people were like, wow, that's cool. And

42:49

a bunch of them gave me their cards.

42:52

And then I emailed all of them. And then they were like,

42:54

yeah, oh, you're in Providence, right? I'm cool. Come to this school. Come

42:56

to this school. Come to this school. Sometimes they even cover my flights.

42:59

And I would sit – and

43:01

then they invite me to talk. I'd talk

43:03

in front of like 100 kids,

43:05

whatever. But then I'd be like, I'm already at your school.

43:07

Can I sit in on some of the classes? Great. And after

43:10

my talk, I would go to every single class. And

43:12

I would give a mini presentation on speechify and make sure every person downloaded

43:14

it. And I would see what bugs they had on their computer. And if

43:16

they had a bug, I'd sit in the back of the class and I

43:18

would code to fix it. And that's how

43:20

I got the problem, the program to be good.

43:23

And I did this for maybe like six, eight months. And the one that's

43:25

just you? Yeah. And you're starting

43:27

off with a Chrome extension or – This was a Mac

43:29

app. A Mac app. The first thing

43:31

was a Mac app. The problem with a Mac app

43:33

is people don't have the habit of downloading Mac apps.

43:36

People do download iOS apps. They do download Chrome extensions.

43:38

But not a lot of people download Mac apps. So

43:40

then I built an iOS app as an ad for

43:42

the Mac app. That

43:44

started to do pretty well. And then

43:46

I was in the dining hall. I had barely

43:48

slept a night before. And

43:51

there was this guy talking to next to me. He was a

43:53

freshman. His name is Sam Rochelle. And when he finished talking to

43:55

his friend, I was like, hey, can I talk to you for

43:57

a second? You said that you're – a

44:00

video editor. Do you have a YouTube channel?"

44:02

He's like, yeah, so he gave me his channel and I looked

44:04

at it and it was amazing. He was such a good creator.

44:06

And I was like, you're free to meet Saturday at 3 p.m.

44:09

I want to make a kickstarter for speechifying. He's like, yeah, sure. We

44:12

met up. We had this great day of shooting.

44:14

We made a video. It was

44:16

a really good video. I put it online

44:18

and I think it got 5,000 views. And

44:21

I put up a website, a very basic website

44:23

with a Stripe button that lets you prepay

44:25

$100 for the software. And

44:28

I edited a video showing

44:30

how it would work if the program

44:32

was actually really good. And then

44:35

he made one that was a lot better than included me talking

44:37

about it. People started to pre-purchase

44:39

the product. And I was like, great.

44:41

I validated it was a real product. And then

44:43

I started hiring... How many pre-purchase did you get? Do

44:45

you remember? Like 10 or 20. Very

44:47

few. But enough to validate that

44:49

this is something people would participate with. And

44:53

then I hired, I think, three or four

44:56

freelancers on Upwork to help

44:58

me code the Mac app. It was

45:00

a... I remember specifically... And

45:03

none of them worked out. I think each worked

45:05

for maybe like six hours. They kept being

45:07

flaky. I was like, I'm just going to do it myself. So

45:09

I just set a goal of how many

45:11

hours it was going to code. Worked

45:13

on it, worked on it, worked on it. And then I

45:16

used to teach computer science at a place called

45:18

Make School. And I was severely underqualified. But three

45:20

of the other instructors there were really good. And

45:22

one of the students was really, really good. So

45:24

then some of them came as like

45:26

basically interns to speechify to help me. And I

45:28

had this amazing friend, Chaitu, who

45:31

was like one of the best computer science students

45:33

at Brown. He had started a YC company called

45:35

DocDoc, but he needed to finish a year of

45:37

Brown. And he was literally in Palo

45:39

Alto going through YC. And I

45:41

went to a computer science class at

45:44

Brown that was like an entrepreneurship computer

45:46

science class. And I convinced the professor to let

45:48

me into the class. Along

45:50

with Sam, this video editor, Yunnan,

45:52

who was a designer from RISD, and Chaitu,

45:55

who was not at Brown, and

45:57

get him to give us a class credit for

45:59

doing this class. even though I wasn't even

46:01

a student anymore. So Sam, Yunnan, and

46:03

Chaitu all got credit for doing this

46:05

class. And

46:08

he verbally told the class that if you're not

46:11

attending the class in person, you can't participate. And

46:13

then I stood up in the middle of the class

46:16

and I was like, hey, to the class, not

46:18

to the professor, Chaitu on our team really

46:20

would love to participate, but he has to be in Palo Alto

46:22

for 60% of the time. Is it okay

46:24

with you, the class, if Chaitu participates, and we make a special

46:26

exception? Put your hand up

46:28

if you oppose. So nobody put their hand up. And

46:31

I was like, no one opposes, And

46:33

he said yes, so Chaitu got to be in the class. So

46:35

Chaitu, who was a lot better than me at computer science, helped me debug

46:37

all my code in the Mac app that was bad. And

46:41

he went on and started an amazing company. Three

46:43

years later, rejoined Speechify, and now he's the chief

46:45

product officer of Speechify. And

46:48

oh my God, Simon, cool. So then

46:51

I went and I was still working on

46:53

this, I was trying to get the Mac app to work,

46:55

and I came up with this media hack of how I was going

46:57

to contact a bunch of reporters to write about the app, because when

47:00

my brother Tyler was building iPhone apps in high school, I reached out

47:02

to a lot of reporters to have them cover the apps. So

47:05

I posted on my Facebook, on Hackathon

47:07

Hackers, Hackathon Hackers Europe, a couple different places. I was

47:09

like, hey, I'm building this app, I need help in

47:11

this, this, and this. This guy

47:13

from Bulgaria messages me. He's

47:16

like, hey, I checked out your website, it looks

47:18

pretty cool, would love to help. I've messaged four reporters.

47:20

I was like, that's so nice of you, thank you.

47:22

He's like, anything else I can do to help? And I was like, well, I'm

47:24

trying to rebuild my website, and I'm busy on the Mac app. I

47:27

know you don't know iOS, so maybe you can help me with the

47:29

website. Here's the file for what I want it to look like.

47:31

If I go to sleep, I'll wake up in the morning, it's done.

47:34

Hmm, I'm like, wow. And then

47:36

he's like, how else can I help? And I'm like, well,

47:38

maybe you can help me hire more iOS engineers. So him

47:40

and I sat, I showed him how I was looking on

47:42

Medium and LinkedIn and trying to fill out engineers

47:45

who I thought were good. I

47:47

wake up in the morning, it's filled to 100. The next

47:49

day, I give him my email, a

47:52

new email, like [email protected], and he's messaged every

47:54

single one of them. Finally,

47:56

I got him a visa, and

47:59

that took me a while to learn. had to do. So

48:01

now everything to do with

48:03

law and companies I feel very comfortable with. I

48:05

got a visa. I got

48:07

in an apartment in San Francisco and

48:10

then an optimal flight moved into SF.

48:13

He was sitting in Birmingham University and

48:15

he had a year where you could do an internship. So

48:18

the internship was now going to be a speech of mine

48:20

and he was amazing. This guy is my rock. And at

48:22

that point we always started having some success. I hired one

48:25

senior engineer from Snapchat,

48:27

one from Apple. They lasted like

48:29

three months. They overbuilt things. Simon

48:32

was in the office till like 11 p.m. every

48:35

single day. And then we

48:38

worked together for a while and then I hired a bunch

48:40

of the students from Make School and people who I did

48:42

hackathons with to come. I got a big house in Palo

48:44

Alto and eight of us lived there for a summer

48:46

and then grew

48:50

from there. And then Simon and I went together to Europe and we

48:52

stayed in a bunch of different places. Drew

48:55

the team more and more started hiring people remotely.

48:57

So we've been remote first for five and a

48:59

half almost six years now. So now there's folks

49:01

working in like 22 different countries who worked at

49:03

speechify. How did you

49:05

first get the revenue to hire people or the

49:07

investment to hire people? What was the? Speechify

49:10

is 140 bucks after a three-day trial. And

49:13

so we had people who were, you know, who

49:15

needed it, right? Kids with dyslexia. So

49:17

you're getting paid users through this kind of school

49:19

hustle that you're doing? Yeah, and it's not the

49:21

school's paying. It's like, you know, kids

49:24

and their mom's paying. Yeah. It was like a real product

49:26

that people needed and I needed. And

49:29

then, you know, 5% of the

49:31

public school system is diagnosed with dyslexia.

49:34

17% have it but are not diagnosed. And

49:36

then something like 32% of people self-diagnose

49:38

is having ADHD. And

49:41

so these people were finding it from Facebook

49:44

groups, Reddit, Twitter, whatever else, and

49:47

made progress little by little. And they're doing a three-day

49:49

trial. And then what percentage are converting to paid users

49:51

at this point? Oh, my gosh, I

49:53

can't remember. So it would have been

49:55

like a not. It was a pretty good percent because

49:58

they were not people who were coming the. There

50:00

were people who had social proof, a friend told

50:02

them about it, whatever it might be. And

50:05

yeah, at that point, I was already in Palo

50:07

Alto, had a couple of my next door

50:09

neighbor gave me some money, literally

50:12

that level of hustling. And

50:15

we grew from there. And then I met a

50:18

bunch of different people who had dyslexia who

50:20

wanted to help. And

50:23

so I met Mike Krieger, who is the

50:25

co-founder of Instagram. And

50:28

then I met Ev, who is the

50:30

founder of Twitter and Medium. And

50:33

the founder of Audible was super helpful, Don Katz,

50:35

and a bunch of different people who just helped

50:37

teach me as I continued to grow. And

50:40

so are you getting funding from these people

50:42

as well? Sometimes yes, sometimes no,

50:44

like small amounts. Yeah. Because

50:46

I guess if someone is thinking of starting a company, it's like... Because

50:52

it sounds like you were semi-bootstrapped in that you're

50:54

just kind of hacking away this app yourself. Yeah.

50:58

And then you're just through the gate who are paying you the $140 a year

51:00

for it. And then you're using that money to hire more people. Yeah. But

51:02

that feels like it would be a hot thing because my understanding is in that in

51:04

tech salaries are pretty high. And so how

51:07

would you possibly afford it? Great. Great

51:09

question. Simon, I'm not going to tell you how much

51:11

he was paid, but he was paid almost another. And

51:16

I was hiring people in Ukraine to

51:18

work for us. This is part of why

51:20

the Apple Snapchat people didn't work out, like

51:23

their salaries were just unrealistic. A

51:26

lot of people were like interns, right, who were university

51:29

students and people who didn't make school with me. And

51:31

like often people who were in high school. And

51:33

so to this day, I have this philosophy that I don't care

51:35

what your pedigree is. I don't care if you went to Brown

51:37

or Harvard or Stanford. I don't care if you worked at Facebook

51:39

or Google. I care that you learn fast.

51:42

You have fire in the belly for the product, high

51:44

loyalty to the team, and you're able to ship features

51:46

fast and move metrics. And that's all I care about.

51:50

And the core philosophy at Speechify is extreme

51:52

product quality and user obsession, leading with love,

51:55

we talked about Caledon, frugality

51:57

and speed. That's all we care about.

52:00

I'm super happy to take a chance on

52:02

someone if they

52:04

look legit. And if they don't work

52:06

out, they don't work out. But if they work out, amazing. Um,

52:09

and so, you know, most of the leadership team in,

52:11

in speechify are, we're all on, like most

52:13

of us are in our twenties. Um,

52:15

and we have multiple high schoolers who are working

52:17

at future five because they're great. Like Yann was

52:20

in Uruguay. He was 16 and

52:22

he found the app and he emailed me incessantly.

52:25

He sent so many messages. Um,

52:28

and he sent me like a 10 page, uh,

52:30

report on how to optimize our keywords for TikTok.

52:32

And he had built a TikTok following a 250,000 people. And

52:36

eventually it was a great boom. It's just

52:38

a credit card. Like go hire micro

52:40

influencers to help us make ads. And he hired

52:42

200 people, uh, who would

52:44

make us ads. And so we fed that into

52:46

three creative producers who would then make a hundred

52:49

ads a week that we would test on the

52:51

different platforms. And so, yeah, I

52:53

just don't care how old someone is or, you

52:55

know, if you're good. Nice. So

52:58

what, what year are we?

53:00

Okay. If you can, so, uh, I

53:02

graduated from Brown 2016. Okay. I

53:06

was a visiting scholar at Brown.

53:08

Yeah. Um, and the beginning of the year by

53:11

March of that year, I had moved to San

53:13

Francisco and, um, a

53:15

couple of months later, Simon moved from, uh, Birmingham

53:18

slash book area to San Francisco. I

53:21

like at this point, like how much revenue is coming

53:23

into the business? If you don't mind sharing? Not

53:26

like enough to, again, we talked about this right before

53:28

I got here. Uh, my brother Tyler was

53:30

a lot more successful than me at this point in time. He

53:33

had gone to Stanford, um, was the president

53:36

of his fraternity. Yeah. Had made enough money in high school

53:38

to pay for Stanford. Um,

53:41

and, uh, started a cybersecurity business that was

53:43

doing amazing. He rented

53:45

this incredible office space similar to this one

53:47

across the street from Twitter in San Francisco.

53:50

And I was living across the street and something that was the size of

53:52

a shoe box, literally imagined the

53:54

size of three of these tables. Didn't

53:57

even have a window properly. Um,

54:00

like shower was outside of my

54:02

little room. And then eventually

54:04

Ty was like, Cliff, you're living in like squalor.

54:07

Just come live in my house. Again, this is my

54:09

brother who's 18 months younger than me. So

54:12

I had an air mattress in Tyler's house. And

54:15

I borrowed a table from my parents that

54:18

I put in the house. And then Simon and I would

54:20

work on this table. And as the team got bigger, we

54:22

all worked on this table. For the first year and a

54:24

half of speechify, I slept on air mattresses in people's couches.

54:26

And then the next year I slept in this guest

54:29

house. There was no office during any of this time. Like

54:31

I was like trying to make it work. And

54:35

the only times I did pay

54:37

rent is I would negotiate a

54:39

three month lease with someone for a house because

54:41

then I could have interns living in the house

54:43

with me. And essentially that was their compensation. Maybe

54:46

a little bit towards saving. So like, not

54:49

that much. But

54:52

when we lived in Palo Alto, we would go

54:54

to Stanford every single day and we'd give the app to

54:57

students and we'd have them like

54:59

tear at the shreds. And

55:02

it was so emotionally difficult because you're spending all

55:04

your time, for years trying to make this thing

55:06

work well. And you're like, I

55:09

can't believe you clicked on that button. I

55:11

can't believe you can't, don't you see

55:13

there's a play button right there? Like why are you

55:15

not clicking the button you're supposed to click? And

55:18

so a lot of it was just user experience

55:20

research. And even today, the designers

55:22

of speechify have so many calls with users

55:24

just to figure out how to make it

55:26

better. Natalie who runs product operations for us,

55:30

every user that doesn't convert, she

55:32

will call it and try to figure out like, what was the problem? Why

55:34

did they not have a good experience? And then try to fix it to

55:36

make sure that it never happens again. Oh,

55:38

this is the best thing that I ever did. 15%

55:41

of the screen real estate at speechify for the first three years was

55:44

a button that says, message us slash help.

55:46

And if you clicked it, you

55:48

entered into an iMessage conversation with me, my

55:50

personal phone number. And then when

55:52

Simon started to just like, eventually he went from

55:55

an iOS intern to iOS engineer to

55:57

head of engineering, the head of recruiting, the

55:59

head of operations. and now he's COO, it became

56:01

a chat with me and Simon. And

56:03

it doesn't matter if it's three in the morning,

56:05

four in the morning, we were responding. And people

56:08

were like, what's happening? Like what type of customer

56:10

support is this? You're always available and

56:12

you're iMessage-ing me. So the beautiful part is I didn't

56:14

need people to enable notifications, it was not like a

56:16

clunky email. It was just inside of iMessage. I

56:19

got to talk to so many users to the

56:21

point that my iMessage broke because iMessage

56:23

is not designed to have this many chats. And

56:25

I had to contact a friend at Apple and

56:27

it was not fixable and the only way to

56:30

fix it was someone at Apple had to delete

56:32

my entire iMessage history and then refresh it and

56:34

then it worked. Nice. Okay, so

56:36

you're grinding away at this, it seems, for like

56:38

a solid two years at this point. Two and

56:40

a half, three years, yeah. And talking

56:43

to users all the time, really being lean

56:45

and frugal with how you're operating. So then

56:47

here's the shift point that you're looking for.

56:51

2019, end of

56:54

the year, I'm in Europe. I'm

56:56

already towards the end of the year, started

56:58

messaging all these people. Around this time, I

57:00

had this amazing experience with this company, Reflectly,

57:02

that were amazing at buying Instagram ads and

57:04

they were based in Denmark. I

57:06

started living in London. Simon

57:09

to this day, we went to this hostel

57:11

we stayed in that

57:13

literally, there were cockroaches on the floor,

57:15

we went outside, someone tried to attack

57:17

us. It was bad. But

57:21

they had wifi. And we

57:23

both worked until we were exhausted. And this was around

57:25

the time that the product was starting to take off,

57:27

people were listening to more than 100 million words per

57:29

month. This is when we built

57:31

the Google Ads strategy for speechify. And so to

57:33

this day, the Google Ads account for speechify is

57:36

in pounds and not in dollars because Google doesn't

57:38

let us shift it to USD because it started

57:40

in the UK. And

57:44

then I kept hiring people internationally.

57:47

And then I moved to LA, to a one

57:49

bedroom apartment with my friend Taylor

57:51

Offer, who has a company called FeedSocks,

57:53

feed.com. And Taylor

57:56

is one of the best people I've ever met at

57:58

performance marketing. and

58:01

we shared a house there. Simon came to live with us. By

58:05

the way, Simon, for one of these years, when I

58:07

was in that guest house, he went back to Birmingham,

58:09

finished his degree, wrote a textbook on neural networks. Previously,

58:11

he was ranked number one in math and Bulgaria in

58:13

high school, and he just grinds harder than anyone I

58:16

know. So it was a long

58:18

flog. So Simon's now back, amazing. And

58:20

at this point, Simon got a green card, which is a really, really

58:22

big deal. Tried to inform me that

58:24

he was finishing up his other company, I flew to

58:27

Michai to convince him to join Speechify,

58:29

he joined us. I convinced my

58:31

friend Valentin to break his lease and come join us as well.

58:33

So now there's five of us living in

58:35

a one bedroom apartment in LA, and

58:38

then COVID hits. How does that work? I'll tell

58:40

you how it works. And I'll send you a video so you can share

58:42

here, it's hilarious. My office

58:45

was a folding Costco table in

58:47

the closet in the bathroom. You open the door and

58:50

you saw the toilet, and then you exited the bathroom,

58:52

and you went into my room, where I shared a

58:54

bed with Valentin. Because it was COVID, we've

58:56

put a bench press in that room. You and I had

58:58

our first podcast on that bench press. And

59:01

then if you go into the living room, there's a dividing window

59:03

in another bed, and two people are sleeping in that bed, and

59:05

there's an air mattress, and someone else is sleeping on that air

59:07

mattress, and there's like two tables in the kitchen that everybody's working

59:09

on. That's the setup. Nice. That

59:11

was quite fun. It was the best time

59:13

ever. And we do super intense workouts every

59:15

single day. All of us were working

59:18

towards running a six minute mile, so we'd run every day.

59:20

It was like summer camp. We

59:23

lived there for like maybe four months. And

59:26

by the end of those four months, speech class started to do really good.

59:28

So we got a bigger house. And we

59:30

had more people. And we went

59:32

to that house, and we stayed there I think for eight months. And

59:35

then we moved to Miami. All

59:37

of us once again had the same house together. Then

59:40

we came to the house in London,

59:42

and then we went back to Miami,

59:44

New York, Seattle, back

59:47

here, and then we'll be back in London towards

59:50

the end of this year, or mid-March. We'll go

59:52

back to London. Sorry, we'll go back to Manhattan.

59:56

And all this time, Simon

59:58

and I are constantly hiring engineers. nonstop.

1:00:01

We're looking for the best people in the world to work with. We

1:00:04

had more founders join us. Tyler, my brother, joined

1:00:07

us earlier this year. Rohan Povluri, who

1:00:09

founded Upsolve, joined us earlier this year.

1:00:11

Rajiv joined us from Amazon where he

1:00:13

led a 100 person team. Rahil, very

1:00:15

similar story, joined us to lead the

1:00:17

web team. We

1:00:20

just kept growing and growing and growing and growing and growing. And

1:00:22

then we launched audio books. Hmm. Well,

1:00:25

yeah, what's the story there? So

1:00:27

number one, the number one,

1:00:29

if you look at how the Chrome extension is used,

1:00:32

the Chrome extension has now been the number

1:00:34

one app in its category for about two and a half years. The

1:00:36

iOS app has been the number one app in its category for about three,

1:00:38

three and a half years. We launched

1:00:41

the mobile Safari extension where you get a play button

1:00:43

on every single page on Safari on mobile and you

1:00:45

can play in a reason. It's amazing. We launched Paltrow's

1:00:47

voice. We launched a partnership with Snoop

1:00:49

Dogg. So we have Snoop Dogg's voice. We're

1:00:52

about to add MrBeast's voice. And

1:00:56

most popular website on Chrome is Gmail,

1:00:59

then Google Docs, then PDFs, then the

1:01:01

rest of the internet, and then like

1:01:03

fan fiction websites and like indie author

1:01:05

websites. We're like, huh, very interesting. And

1:01:08

I love

1:01:11

audio books, but

1:01:13

there's really only one app where you can buy audio

1:01:15

books today. And even in iTunes,

1:01:17

they get all their audio books from this

1:01:19

other app, which is owned by Amazon. And

1:01:24

Amazon is not very nice to authors. It

1:01:26

used to be that for audio books, they would keep

1:01:28

20% of the profits and publishers

1:01:30

and authors would get 80%. With

1:01:33

time, because it's been a monopoly, so Apple was

1:01:35

sued in the early 2000s by the federal government

1:01:38

in the US for colluding as the American consumers to

1:01:41

set the price of audio books and

1:01:43

books. So they went out of it.

1:01:47

Now Amazon keeps 80% and

1:01:50

gives 20%. And if you give

1:01:53

them an exclusive deal, and you're a big

1:01:55

author, they'll let you keep 40 and look

1:01:57

at 60. If you look at every

1:01:59

other platform like Steam games, whatever. It's

1:02:01

usually 7030. Yeah. So

1:02:05

we didn't think that was particularly fair. And

1:02:08

so we went and worked to get

1:02:10

a license system we did. So now almost

1:02:13

every audiobook you can imagine you can buy

1:02:15

on speechify. We built this incredible back end

1:02:17

that allows us to process all these audiobooks.

1:02:20

We launched now we are the other than

1:02:22

Amazon the only company in the world that

1:02:24

has a credit based audiobook subscription. And so

1:02:26

now you can use speechify to read any

1:02:28

PDF any physical book your emails, your text

1:02:31

messages and any book you desire. The

1:02:33

next step is we're launching

1:02:35

the function to listen to ebooks.

1:02:38

So you'll there's about 450,000 audiobooks,

1:02:41

but there's millions of ebooks and normal books. And

1:02:43

so the next step is one, we're using the

1:02:45

model that Tyler is team built. So Tyler joined

1:02:47

the company. And in seven months, it

1:02:49

went from one AI engineer to another eight AI

1:02:52

engineers working at speechify working on

1:02:54

national language processing, optimal character recognition,

1:02:56

transcription, translation, and

1:02:58

speech synthesis. And they've built models that are like

1:03:00

incredible. So if you played around with stuff like

1:03:02

open AI and chat GPT, very similar

1:03:04

to what they've created, but for speech. And

1:03:07

so a lot of that innovation will launch towards the

1:03:09

end of the year. I

1:03:11

have lots more questions about the speechify stuff. So we've talked about

1:03:13

kind of growing the team from basically

1:03:15

zero to now you're about 100 people, I guess,

1:03:19

for most people listening to this, and actually, for me as

1:03:21

well, can you just give like, if you were to zoom

1:03:23

out, and just riff on what are the

1:03:25

steps to build a startup? Oh, yeah, I said someone's listening to

1:03:27

this and they're like, I love the idea of building my building

1:03:29

my own startup. I don't want to work for the man. I

1:03:31

want to do my own thing. Yeah, I love the idea of

1:03:33

like building a product and like making a difference in people's lives

1:03:36

and making decent money while doing it and working with a team

1:03:38

in real life to sound sick. Like, what are

1:03:40

the steps? Cool. First thing is, let's

1:03:42

go even more meta. How

1:03:47

Do you find the right idea? So The first

1:03:49

thing is while you're walking around the world, always

1:03:51

think to yourself, this sucks, how can I fix

1:03:53

it? Right? Excel can't do this, this sucks, how

1:03:55

can I fix it? Right? I Have this like

1:03:57

T and it's whatever, how can I fix this?

1:03:59

That's one way. Cameo player the other frameworks

1:04:01

coming up with ideas is thing about the

1:04:03

following What technology exists today or will exist

1:04:05

in three years that has not exist so

1:04:08

far and will take build on it. And.

1:04:11

That means that I will not have competition try because

1:04:13

I'm building for where the football is going. That where

1:04:15

the football is. Now that number one. Number

1:04:18

two is was a simple. To.

1:04:21

Build like Apple website because of use that it'll

1:04:23

be better. Am that has a

1:04:25

business model that makes sense and then you have to decide?

1:04:27

Okay, we'll Am I going be to be so. Sorry.

1:04:30

Selling to businesses are immigrant be to see selling

1:04:32

to give serious if you Geisel to consumers are

1:04:34

eager to get people's your product. Are they going

1:04:36

to her for each other because it has a

1:04:39

viral loop com? Are you going to run ads

1:04:41

successfully so that the cost the customer acquisition is

1:04:43

less than the first payment? your it. Or.

1:04:47

Are you going to have a sale

1:04:49

thing with a more and you know

1:04:51

something like that and then I'm before

1:04:53

building a product. I recommend. Closing.

1:04:56

Says else so. Either.

1:04:58

Has. Some companies to you are somehow some individuals

1:05:00

to you and if that's difficult, start by getting

1:05:03

into, give you an email to for example don't

1:05:05

linked and and write about what the idea is

1:05:07

and say i'm going to build this comment your

1:05:09

email if you're into it. And the nice thing

1:05:11

about linked in his engagement is like sheriff's every

1:05:13

time someone coming through email it gives her to

1:05:16

the networks. If you have a good enough idea

1:05:18

people freak out. When. I

1:05:20

was a college. Or this is my philosophy

1:05:22

and when I had like a good idea I'd I'd put

1:05:24

it out there and you know what? The one that the

1:05:26

best was. I think a

1:05:28

picture of a cheese spray can and

1:05:30

said i'm building spray able mattel on.

1:05:34

This thing got like hundred organic shares on

1:05:36

facebook back into the in So much engagements

1:05:38

as I I will sell it for fifteen

1:05:40

dollars com and if you want the first

1:05:42

batch. Everybody wanted Rebels is hella.

1:05:45

So just the that like right amount of

1:05:47

twitter right the modeling dance and then get

1:05:49

he will sign of for an analyst What

1:05:52

I should have done was it basically did

1:05:54

but it could have done it more is

1:05:56

made an email list for people with dyslexia

1:05:58

and like parents at Giza Dyslexia. And

1:06:00

sent out an email every day or every week with

1:06:02

all the resources that I had made a video every

1:06:04

day or every week. and by the time I started

1:06:07

selling my product I would have had an email us

1:06:09

at one hundred thousand deal. That would be amazing. I

1:06:11

basically did this by writing a book published online. But

1:06:14

I could have been even more intentional.

1:06:16

But so am. Whether it's selling to

1:06:18

businesses or selling to consumers, start by

1:06:20

creating content around it. Email.

1:06:22

List is the easiest one. You can also

1:06:24

you to channel you can post about it

1:06:26

only done regularly whatever it may be but

1:06:29

capture people the next as number one. Once

1:06:31

you've done that are concurrently with doing that

1:06:33

but definitely not. This is a second a

1:06:35

separate of the first step. Start building a

1:06:37

project. Ah, You.

1:06:40

Can build ninety percent of what

1:06:42

products on google sheets. Only

1:06:44

website woody me I'm a given example.

1:06:46

So this thing about find Me scholarships.com.

1:06:50

And we that a vi the Autumn.

1:06:52

Quick mathematically identify the best freelancers and

1:06:54

up work com and we would auto

1:06:56

generate. A. Google sheets for you with

1:06:58

a hundred rows of scholarships that you were

1:07:00

most likely to win depending on your Gp,

1:07:02

A, your background, what you studied, where you're

1:07:04

from, etc. And so we literally gave the

1:07:07

people who signed up for the service a

1:07:09

Google seat and then the virtual assistants would

1:07:11

sell out this Google sheets and I had

1:07:13

a google sheet of all the virtual assistants

1:07:15

what their hourly rates were, how many hours

1:07:17

they were doing, a higher on the entire

1:07:19

company to school seats because it just didn't

1:07:21

wanna waste the time building you know, databases

1:07:23

and Google she is a database as the

1:07:25

everything the you could show. Inside of any

1:07:28

of these things as there's a google seats so

1:07:30

start with google sheets and if you need to

1:07:32

charge people if you're american just venmo request people

1:07:34

are paypal request people by always it was elected,

1:07:36

paid a paper I didn't even need to and

1:07:39

com and stripe arm and it again later I

1:07:41

did when the beginning I didn't for i'm. I'm

1:07:44

sources like. you

1:07:46

said other i was in i was in pakistan for

1:07:48

the last two weeks and i was on a road

1:07:50

trip with all my cousins his trying to go to

1:07:52

don't stop us and we were talking about his struggles

1:07:54

he spent two years building the product before reaching at

1:07:56

any one and doing any kind of conversations or any

1:07:59

kind of self thing anything like that. And

1:08:01

so the advice I was giving him to him as

1:08:03

well was a everything is downstream lead generation, like you've

1:08:05

got to generate the leads before you even build a

1:08:07

thing. And secondly, I asked him the exact question, what

1:08:10

does the Google Sheets version of your product look like?

1:08:12

And he was like, his mind was blown. He was like, I've

1:08:14

just never thought in that way before. He thought he was building

1:08:17

iOS app, he's going Android app, putting it to web app as

1:08:19

well. Also that he could then show the

1:08:21

people he's trying to pitch the idea to two years later that

1:08:23

I've already built this thing and it's really good for you. And

1:08:26

I think, yeah, just building Google

1:08:28

Sheets. The other thing that is

1:08:30

sad and will blow people's minds negatively

1:08:32

is look, I've rebuilt the app at

1:08:34

least 50 times. I've rebuilt the onboarding

1:08:36

at least 1000 like

1:08:38

or someone in the team did. It's

1:08:42

just a matter of like levels of polish, you got

1:08:44

to polish, polish, polish, polish, polish. And to show people

1:08:47

the idea, he didn't need to build an iOS app,

1:08:49

he could sketch it on a piece of paper with

1:08:51

a pencil. That's what I did with Board Break, I

1:08:53

would pitch people literally the sketch. And then later, after

1:08:55

they gave me a lot of feedback, I built the

1:08:57

actual product. So like you said, everything is

1:08:59

downstream of lead generation. And it is not the case that

1:09:01

if you build it, they will come. If you build a

1:09:03

good website, if you build a good app, people will not

1:09:05

come to your app or website, figure out how

1:09:07

you're going to make them come. Half

1:09:10

of your time can go to building the product, but

1:09:12

at least half of your time should be figuring out

1:09:14

how people will come to use your product. By the

1:09:16

way, one of the amazing things is that YouTube provides

1:09:19

is a huge free billboard that if

1:09:21

you have videos that convert, i.e. they have

1:09:24

good retention and they have good click through rates,

1:09:26

you will get more users. And

1:09:28

so YouTube is an amazing customer

1:09:30

acquisition funnel. Same thing for search engine optimization,

1:09:32

same thing for app store optimization. Choose

1:09:35

which one you are picking, and then

1:09:37

go down that rabbit hole. And

1:09:40

then the next level is strategy. Okay, if you

1:09:42

have a business with a moat, where you have

1:09:44

monopolistic tendencies, where the big you get the more

1:09:46

challenging it is to compete with you, that's a

1:09:48

really good feature to have in a

1:09:50

business. And then after that, at

1:09:53

the end of the day, a company, the definition of the

1:09:55

word is a collection of people. The

1:09:58

best people in the world should be working with you. the

1:10:00

most important thing, the most important thing.

1:10:03

And so you just need to be very persuasive

1:10:05

to get people to bind to your vision. And

1:10:09

by and large, people will not join you because they believe in

1:10:11

your vision. They'll join you because

1:10:13

they believe in you. And

1:10:15

so my number one piece of advice, if you want

1:10:17

to be a strong founder, make

1:10:19

yourself the best version of yourself, right?

1:10:22

Read more books than everybody else. Work

1:10:25

harder than everybody else. You know, work

1:10:27

out, philosophize, solve out your issues, talk

1:10:29

to a therapist, figure out, people want

1:10:31

to work with you because you inspire

1:10:33

them. And that's part of

1:10:36

what it takes to be a good leader, right? People

1:10:38

will need to be inspired to want to collaborate with

1:10:40

you. And so a lot of the people who joined

1:10:42

Speechify, especially in the early days, didn't

1:10:45

necessarily believe in the idea, but

1:10:47

they did believe in me, which I'm very lucky to

1:10:49

have people around me who believed in me. But

1:10:53

it was also because I built 36 different products. And like I'd

1:10:55

say, I'm going to do this and then like a

1:10:58

week later it was done. And

1:11:00

I have a lot of people who joined Speechify who told me, look,

1:11:02

I didn't join initially, but you told me

1:11:04

that you were going to do this. And then a week later it

1:11:06

was done. And then you told me you were going to do this.

1:11:08

And then a month later it was done. So many times you said

1:11:10

you were going to do something that I thought was impossible, and then

1:11:13

it happened, right? You and I met a while ago. You knew I

1:11:15

was a fan of Brandon Sanderson. I'm sure I'd mentioned this to you.

1:11:17

And you probably were like, okay, whatever.

1:11:19

And then you're like, oh, all right, that actually happened. And

1:11:22

so this has happened so many times that people were, when

1:11:25

I say things that sounds crazy, they're like, he just has

1:11:27

a track record of doing the things that he said he

1:11:29

was going to do. So if you develop that reputation, you

1:11:31

will end up with a really good team, and

1:11:33

then do your best to get to the most

1:11:35

interesting people who you can learn from and grow

1:11:38

from, not just as

1:11:40

mentors, but as teammates. And then when

1:11:42

you become a really good person, the

1:11:45

best form of leadership is seeing the greatness in others before

1:11:48

they see it in themselves. And so if you can cultivate

1:11:50

the ability to do that, then

1:11:52

you're really setting yourself up for success. And so I

1:11:54

think that, to summarize, look

1:11:57

at a technology that he did not... or

1:12:00

doesn't exist yet but will in the next three years. Find

1:12:04

a business that ideally is monopolistic in the long term. Make

1:12:07

yourself the best person that you can be. Validate

1:12:09

that people will pay for it. Figure

1:12:11

out how to build lead generation before you build a product.

1:12:14

And see the greatness of other people before they see it in

1:12:16

themselves. Okay, so everyone

1:12:18

I've ever interviewed on this podcast about growing the

1:12:20

business or their YouTube channel, whatever

1:12:23

the thing might be, has had a period of time, generally

1:12:27

in the early days, where if you asked them, they

1:12:29

would have said, what work-life balance? And

1:12:31

a lot of these people, as they get older and

1:12:33

become more successful and become multimillionaires, at that point they're

1:12:35

like, well, you know, work-life balance is important. And

1:12:38

it becomes a kind of luxury belief to be like, take

1:12:40

care of yourself, etc., etc. And I'd like to

1:12:42

ask the question of like, you know, do

1:12:44

you think the period of hustling and grinding is necessary

1:12:46

to get to the point of, quote, success? Or

1:12:49

do you think you can have this dream of

1:12:51

work-life balance even while you're building something in the

1:12:53

early days? Well, if you

1:12:55

look at the sentence structure you used, what

1:12:58

does success mean there, right? And

1:13:00

so implicitly success means you are

1:13:02

hypothetically more successful than other

1:13:04

people, either you have more attention, you make more money, whatever

1:13:06

it might be. So extraordinary

1:13:09

results require extraordinary inputs. And

1:13:11

the thing that people typically glaze over is the fact that

1:13:13

if you work 100 hours a week instead of 50, there's

1:13:19

not diminishing but exponential returns to the

1:13:21

additional hour. Oh, okay, tell me more.

1:13:25

Because my intuition on this would be diminishing returns.

1:13:27

In some places that is the case. Yeah, you

1:13:29

would think that, oh, your 99th hour is less

1:13:32

productive than your 31st hour. Cool. Let's talk about

1:13:34

ads for a second. I

1:13:36

went down the rabbit hole in ads. And

1:13:39

I found the arbitrage. I figured out

1:13:41

where to run ads, how to run ads in a way

1:13:43

that other people did not. So while everybody else is fighting

1:13:45

over here, because they're spending 40 hours, I

1:13:48

did the research to figure out a bunch of things. And so

1:13:50

I'm not here. And so I'm

1:13:52

doing something completely different than everybody else

1:13:54

because I'm just a couple of

1:13:56

steps ahead. And so the problem

1:13:58

with customer acquisition is that – it is

1:14:01

a zero sum game. There is a

1:14:03

set amount of impressions per month that exist across

1:14:05

all the different platforms. And

1:14:07

so you want to win those impressions, right? And so either

1:14:09

you figure out how to have a higher retention on your

1:14:12

product, so you can afford to have a higher crack, or

1:14:15

you have better ads, so you have better click-through

1:14:17

rate, whatever it might be, but you end up

1:14:19

with diminishing, sorry, exponential marginal returns if you put

1:14:21

in the extra hour. Same thing for if someone

1:14:23

else has a duplicate app than you, but

1:14:25

you have slightly better retention, all the users would end

1:14:27

up with your app, not with their app, because you're

1:14:30

just objectively better by that 5%. So

1:14:33

in markets and games where it's winner-take-all,

1:14:36

it's exponential marginal returns. And

1:14:39

typically, it's places where learning is involved, or

1:14:41

speed is involved. OK, that's sparking a lot

1:14:43

of thoughts in my mind as well. It's like one

1:14:45

of the things we teach on our YouTuber Academy is that people

1:14:48

were asking, what's the minimum viable dose I need

1:14:50

to succeed on YouTube? Minimum effective

1:14:52

dose, yeah. Yeah, minimum effective dose, yeah. So

1:14:54

I often quote the number of 10 hours a

1:14:57

week. I think if people are putting in 10 hours a week,

1:14:59

that's enough time to make one video every week. If

1:15:01

you've helped towards editing, maybe it's enough time to make two videos a week.

1:15:04

But then people always ask, what if I put 20 hours a week?

1:15:07

Will I be more likely to succeed? And the answer is, well, yes.

1:15:09

Because if you can make one video a week, and if

1:15:13

you're making three videos a week, and someone else is making

1:15:15

one video a week, the extra several hours that you're putting

1:15:17

in is actually not

1:15:20

diminishing returns over time. It's actually exponentially compounding over

1:15:22

time, because now you've got three times as many

1:15:24

data points every single video you make. It's

1:15:27

a lottery ticket for the algorithm to blow up or someone else to find your

1:15:30

channel. Exactly. And you get the

1:15:32

exponential returns, exactly. And so, it's interesting.

1:15:34

I've never thought in this way before, because that doesn't

1:15:37

make a lot of sense. Because my intuition before you

1:15:39

said that was like, hang on, that sounds wrong, because

1:15:42

everything I've ever read about this, just in like articles

1:15:44

and stuff is, hey, don't work too hard because dot,

1:15:46

dot, dot. Like you need to look after yourself. You

1:15:48

don't want to burn out. But you're

1:15:50

right. It's like, take the analogy of like the

1:15:52

Tour de France. You have to get up in

1:15:54

front of the pack. Like the Tour de France.

1:15:58

Oh, yeah, sure. You have to. to crush

1:16:00

it in the beginning to get in front of the rest

1:16:02

of the pack. Once

1:16:04

you're in front of the rest of the pack, it's okay

1:16:06

to cruise a little bit. That's why

1:16:08

companies who get to become the big

1:16:10

fang, right? Microsoft, they can be

1:16:12

slow. They could do whatever they

1:16:14

want and the meandering pace because they've

1:16:17

already, you know, we dominate with the

1:16:19

monopolistic business models that make it difficult for

1:16:21

someone else to succeed. But

1:16:24

if you want to go around the moat

1:16:26

of the monopoly, you have to

1:16:28

work harder than everybody else. Now,

1:16:30

ideally the way that you do that is

1:16:32

you pick three, four, five friends or

1:16:35

at least one other person that could be working on something

1:16:37

different, you and Tamar, right? You and me. We're

1:16:40

in a pod, right? You, me, Valentin, Tamar. We

1:16:42

are constantly working together and I'll have a call

1:16:44

with you or with him and we'll teach you

1:16:46

some of the things that we are learning together

1:16:48

because it's not me against everybody. It's me against

1:16:50

like the masses, but I can have, you know,

1:16:53

Jimmy or Eric or whoever and or

1:16:57

Richard Branson and we can learn from each other. And

1:17:01

so having a pod of people like that who you can learn from is

1:17:03

amazing whether they do the same company

1:17:05

you or not. But if you have seven

1:17:07

of those people working on the same thing as

1:17:10

speechify, we have like dozens of these. Then

1:17:13

you're very unbeatable. Then it's very very hard because like

1:17:16

everything I told you about me and performance marketing, Tyler is

1:17:18

doing the same thing for AI. Tyler every

1:17:21

day is spending like a hundred hours a

1:17:23

week doing math and programming

1:17:25

after having skip

1:17:28

four and a half years of math in high school and that a

1:17:30

math undergrad and in the Masters in AI where you got like a

1:17:32

98 percentile and all the assignments

1:17:35

and now he hired eight other people like him and

1:17:38

Chai is just doing the same thing on product and Simon

1:17:40

is doing the same thing operations and podcast is doing the

1:17:42

same thing on finance and Rajiv and Rajeela doing the same

1:17:44

thing on engineering, right? And

1:17:47

Roman is doing the same thing on it.

1:17:49

Kai and on the back end platform. So

1:17:51

we and you know, Rohan is doing that

1:17:53

on SEO and growth. And so we've had

1:17:55

like there's 20 people in the company leading

1:17:58

organizations that are each putting in. hours

1:18:00

like this and researching and learning and if,

1:18:04

look, Tyler, my brother, can do a backflip. He

1:18:07

wasn't that interested in backflips growing up, but I learned how to do a

1:18:09

backflip. So then, you know, he

1:18:11

had to learn it too. I learned how to make iPhone

1:18:13

apps when I was 19. I didn't

1:18:15

have an interest or attitude for iPhone apps when I was younger, but

1:18:18

he made enough money in high school to pay for Stanford, so I

1:18:20

had to go and learn iPhone apps too. If you have someone next

1:18:22

to you who is so

1:18:24

motivated and outperforming at every single corner,

1:18:27

you're gonna have to pull that out of the bag and do your

1:18:29

best as well. And so surrounding yourself with people

1:18:31

like that is key. What

1:18:34

does success mean to you? When

1:18:36

I was 14, I babysat

1:18:38

some family friends and my mom was like, Cliff,

1:18:40

it's a full moon. You get 30 moon wishes.

1:18:42

So I pulled out a notebook and I wrote

1:18:44

down 30 wishes and then I

1:18:46

collapsed them into the three things that I would

1:18:48

most want. And over time,

1:18:50

I consistently reflect on them and I read books and

1:18:53

I update my beliefs and so the three things that

1:18:55

I want, number one, is

1:18:58

to be the best person that I can be and to

1:19:00

have kids who are greater than me. I

1:19:02

ideally want five to seven kids. Number

1:19:04

two is I want to maximize the love in

1:19:06

my life, whether it be my family, my friends,

1:19:09

significant other. Number three is to create as much

1:19:11

value in the world as possible and elevate the

1:19:13

collective quality of life, ideally

1:19:17

from people who are similar to me. People

1:19:20

with dyslexia, ADHD, low vision, etc. And

1:19:24

the biggest way I can do that in the short term is with

1:19:26

technology. Those

1:19:31

are like the top ones. And then obviously, I

1:19:33

want money to not govern my decision

1:19:35

making. So feel freedom. And

1:19:38

freedom is not just that. I

1:19:40

practice parkour. It's like a mix of gymnastics

1:19:42

and martial arts. Anything that I want to

1:19:44

do, I want to be able to like jump on that and do a front

1:19:46

flip. Like I can do that. That makes me happy. I

1:19:48

want to do a round off backhand swing double backflip. I can

1:19:50

do that. That makes me happy. But I can't do a butterfly

1:19:54

twist swing through court and so

1:19:57

I go and I drill that every week until I learn it. And

1:20:02

then there's other hobbies. I wanna have five

1:20:04

songs that I have written that

1:20:06

my friends play at parties because they like the

1:20:08

songs. Because I like music

1:20:10

and that's a thing on my list. And

1:20:14

I'm very metrics driven. I wanna be 185

1:20:17

pounds, 13% body fat. And

1:20:22

I wanna have a bunch of goals around how much I

1:20:24

wanna bench press or whatever. But the

1:20:26

first three things are what success means to

1:20:28

me. Yeah,

1:20:30

and that has evolved over time. What's

1:20:33

your process for goal setting? I

1:20:36

had- You sent me a screenshot of one of

1:20:38

your Google Sheets and I was just blown away

1:20:40

by how intentional and how systematic

1:20:42

you seem to be about setting goals. And I've

1:20:44

never been particularly systematic about setting goals. So I

1:20:47

feel like, yeah, I'd love to learn from you.

1:20:49

What's your process here? So the

1:20:51

first thing I do is I have a group called Elephants with

1:20:53

Valencine and Barish. And we meet

1:20:55

every quarter. We'll go to like a beach or a hotel or

1:20:57

something. And for three days, we'll write

1:20:59

out our goals. Three days? Yeah,

1:21:01

sometimes it's shorter, sometimes it's longer, depending on how much

1:21:04

time we have. We also like, we'll do adventures in

1:21:06

the mid-time. We'll go like kitesurfing or whatever. But

1:21:09

I write my goals across seven

1:21:11

categories. Love,

1:21:14

adventure, music, spirituality,

1:21:18

intellectual curiosity, business, fitness,

1:21:22

goals for loved ones. And

1:21:24

I'll write goals in one

1:21:26

month, three months, six months, one year,

1:21:29

three year, five

1:21:31

year and 10 year increments for

1:21:33

each one of those categories. Oh, wow. That

1:21:36

sounds intense. Yeah, it's great. And

1:21:38

the way I do it now is I always will

1:21:40

start like a bullet point with an emoji that represents

1:21:43

that. Cool. I'll write that and

1:21:45

then I'll come back. Hypothetically,

1:21:47

I should be doing it every month. I just do it whenever I'm on

1:21:49

a plane. I

1:21:52

will write. Oh, that's clever. Where I do it. Yeah, yeah, yeah,

1:21:54

oh my God. I like almost never pay for wifi on planes.

1:21:57

I just have two notebooks in my backpack and I write out my

1:21:59

goals. and my thoughts and

1:22:01

I process my goals at Google, that

1:22:04

I save offline. So I update

1:22:06

them and so like one month from now,

1:22:08

and the way that I typically do it is instead of

1:22:10

one month, three months, if it's like I'm close to being

1:22:14

whatever age I'm next, I'll fix it to that age. That's

1:22:17

one. Two, I had a goal

1:22:19

system I used in the past. I

1:22:21

used the same similar categories, but I broke it down by

1:22:23

age, like 27, 26, whatever, all

1:22:26

the way up until 200. I think I'm gonna live to be 200. And

1:22:29

I even included like when I'm gonna get married, when I'm

1:22:31

gonna have my first, second, third, fourth, fifth,

1:22:33

seventh kid, when I'll have my grandkids. And

1:22:36

I wrote that with all my goals, and then I

1:22:38

wrote how can I achieve those goals? So

1:22:40

for example, I wrote, okay, I wanna be able to do like

1:22:42

a Webster, which is like this front football kick. Cool

1:22:44

to do that, I need to buy like X amount of protein powder, I

1:22:47

need to have access to gym, I need to have access

1:22:49

to a car, I need to pay for like gymnastics, and

1:22:52

ideally I want like these two coaches to

1:22:54

help me, and I DM them on Instagram and

1:22:56

they said it's $45 an hour, and

1:22:58

I need like a $15 tripod to like, FaceTime them

1:23:00

while I'm flipping, so they can like give me instructions,

1:23:02

so like I'll pay for the instructors, and I'll figure

1:23:04

out exactly how much money I need for all of

1:23:07

my goals, and so my goal is to make sure

1:23:09

that I'm making enough money to pay for everything that

1:23:11

will let me get to my goals, and

1:23:13

then I'll like look at what my goals are for like the

1:23:15

future years, and I base my target income based on

1:23:17

what I need to do to achieve my goals. Bloody hell.

1:23:21

So that's one way of doing it. And

1:23:23

then I have a thing called Life in Cubes

1:23:25

that I picked up from my friend Felix Kraus, who

1:23:28

made Fastlane, and it's just like from the, and

1:23:30

so it's one cube for every week I've

1:23:33

been alive, so 52 cubes in a row, and

1:23:35

then 28, so 28 columns. And

1:23:40

everyone in preschool, high school, this is where I learned

1:23:42

how to do a backflip, like

1:23:44

it goes all the way back. And

1:23:47

then I like project, well, what I want this

1:23:49

to look like in the future. Wait,

1:23:51

so is this like a Google Sheet, or is it like? Yeah, it's

1:23:53

a Google Sheet that shows all of this. Okay, so

1:23:56

like 52 rows and like X number or column depending on

1:23:58

what age you are. Correct. I could

1:24:00

look backwards at the emojis and the titles and

1:24:02

see, oh, interesting, I never thought about this. The

1:24:05

first four rows are essentially, there's

1:24:08

no events. Hypothetically,

1:24:10

I could write, I started walking, I started speaking, whatever.

1:24:12

And then the second two rows, also, almost no events.

1:24:14

But then I learned how to do a back handspring,

1:24:17

and then I learned how to do a backflip. And

1:24:19

then I read the four hour work week, and then

1:24:21

I got a six pack for the first

1:24:23

time. And then I kissed a girl, whatever it might be.

1:24:27

And then I moved to London, I moved to Germany,

1:24:29

and I moved to Paris, and I did this, and

1:24:31

I did that, and Simon joined the company, and Tyler

1:24:33

joined the company, and we closed the deal with Sanderson,

1:24:35

whatever might happen. And I published,

1:24:37

read to you, great song on Spotify about

1:24:40

speech fine. Yeah,

1:24:44

and then I think about what are the things I want for the future. Cool,

1:24:47

I wanna live in New York, I wanna pick up stand up

1:24:49

comedy, I wanna do Brazilian Jiu Jitsu, I wanna learn guitar, I

1:24:51

wanna learn piano. And I put this

1:24:53

on my goals, and then I try to do them. Okay,

1:24:56

help coach me on this. So one of the things I

1:24:58

have on my goals is that I

1:25:03

would like to become

1:25:05

sufficiently flexible in mobile that I can do

1:25:07

acro yoga. Amazing. Cause right

1:25:09

now I can't even do a straight leg raise in terms

1:25:11

of hamstring flexibility and stuff. And

1:25:13

so it's like cool, I've got that as a thing. How

1:25:16

would you kind of break down, like if you have

1:25:18

the goal, like the process of getting there? Fantastic. Well,

1:25:21

that's not too hard. First of all,

1:25:23

you can already do acro yoga, you don't need to

1:25:25

add flexibility. So what is the move in acro yoga

1:25:27

that you wanna be able to do? Straight leg raise.

1:25:29

Okay, great. Straight leg raises, your back is against the

1:25:31

ground, and like, and your legs are fully straight, or

1:25:33

can you bend a little bit? Fully straight, I'll do

1:25:35

it here. Okay. What is

1:25:37

the exercise you need to do in order to get to

1:25:39

that level of flexibility? Hamstring stretches. Okay, great. How

1:25:42

many days of hamstring stretches do you need to get there? Apparently 10

1:25:44

minutes a day for 30 days. Great. Set

1:25:47

a calendar event. I have this on my phone right now,

1:25:50

not for stretching, but for making music. Every

1:25:52

night, I'm supposed to spend an hour

1:25:55

writing songs. I'm currently obsessed

1:25:57

with NF, the rap. NF? Oh,

1:25:59

okay. And so every day, 9 p.m. to 10

1:26:02

p.m., music writing, and

1:26:04

the song I'm currently working on is a song called The Fire in My

1:26:06

Chest. And I write,

1:26:09

so have an alarm, set

1:26:11

for that time, and the calendar event, and

1:26:13

you have to stretch, and if you don't stretch,

1:26:15

text me, and I'll pay by request you $50.

1:26:19

Or get Taymor or anyone of your roommates

1:26:21

to do it with you. If

1:26:24

they have the same goal, great. If they have a different goal, they can

1:26:26

spend 10 minutes on the goal that they care about. And

1:26:28

that's it, and just do it for 30 days. And

1:26:31

you have to do it for 30 days, and if you see up one day,

1:26:34

you have to restart the entire 30 days. And

1:26:38

add some consequence to yourself. Run

1:26:40

10 miles, whatever it might be, if you miss the goal.

1:26:44

But more importantly, you

1:26:49

can already do acro-yoga without doing straight leg raises, I

1:26:51

promise you. Yeah, I've been for a few lessons, it's

1:26:53

quite fun. Exactly, so how far away is it from

1:26:55

your apartment? Like 10 minutes. Walking

1:26:58

or Uber? Amazing. Do you have

1:27:00

a friend in acro-yoga? Amazing. So

1:27:02

one, is get the phone numbers of three people in acro-yoga,

1:27:05

and every day at 5 p.m., be

1:27:07

like, hey, I'm going today, are you guys going? And then they'll tell you

1:27:09

that they're going, and then they'll be much more fun. And

1:27:12

then when they go and you're not there, they'll message you,

1:27:14

hey, Ali, where were you? And you'll feel like it's family,

1:27:16

like you wanna go. And

1:27:18

have it in your calendar, and never miss it.

1:27:20

So I just never miss gymnastics. And

1:27:23

I found I missed it a couple times, so what

1:27:25

I started doing is I ordered, I rescheduled the Uber,

1:27:28

like five hours beforehand to

1:27:30

just arrive at my house 30 minutes

1:27:32

before gymnastics happens. And so I'm rushing,

1:27:34

grabbing things, because the Uber is outside.

1:27:37

And then I never missed gymnastics. One

1:27:39

of the things we alluded to earlier was you said that something

1:27:43

around, if you want it enough, you'll find a way to

1:27:45

make it happen. I find that when

1:27:47

I set goals, like this acro-yoga thing, or like

1:27:49

for the last seven years, I've fixed my posture,

1:27:51

or kind of become more flexible, and these kind

1:27:53

of things, or even last year, before,

1:27:55

try and get six-pack abs. But when it

1:27:57

comes to the hard bit of those particular

1:27:59

goals, goals, I realized, I'm

1:28:01

not really sure I want this. Oh, and

1:28:04

so like, you didn't want

1:28:06

enough. Yeah. So people,

1:28:09

for some reason, it's a meme recently to say,

1:28:11

don't tell people your goals, bullshit, tell people your

1:28:13

goals, tell your parents, tell your girlfriend, tell the

1:28:15

guy down the street, so that when they meet

1:28:17

you five days from now, and they go, how's

1:28:19

Acre yoga going? You're way too embarrassed to say

1:28:21

I gave up on my goal. No

1:28:24

one likes the guy who gives up on his goal. Yeah. commit

1:28:27

your goals, platform all over

1:28:29

your Instagram. Oh my god, there's

1:28:31

this influencer. I love her. I can't remember her

1:28:34

name. But her bio on Instagram is I will

1:28:36

be on SNL one day. God

1:28:38

damn it. I love people who have a mission like that. I've

1:28:41

actually thought to myself, what is the one liner that

1:28:43

I can put that is such an inspiring goal. And

1:28:47

so, you know, now I'm the guy

1:28:49

who's taking down audible, or

1:28:52

the guy who wants to be responsible for

1:28:54

people reading a trillion words. Recently,

1:28:56

I've gotten really into the idea of gamifying speech

1:28:59

or five where there's a store

1:29:01

with rewards if you're a kid between the

1:29:03

ages of five and 15. And

1:29:06

if you read a certain amount, and it will give you

1:29:08

like an NLP based quiz at the end, that you will

1:29:10

verify that you actually understood, you'll gain points like an arcade

1:29:12

and you can get like a ps4 or an Xbox or

1:29:14

a basketball or whatever. Like I

1:29:16

want to be responsible for kids learning how to read. And

1:29:18

then we have a language learning algorithm that teaches you other

1:29:21

languages that's infused inside of the normal reading that I want

1:29:23

to release. So like, I'm

1:29:25

trying to figure out how to actually wordsmith that

1:29:27

so that it makes sense. And

1:29:30

so yeah, enough people learn about the goal, and then

1:29:32

I'm committed to it. But also, it's on my sheet,

1:29:34

it never goes off my sheet. And

1:29:36

the only way it gets off my sheet is either I accomplish it or

1:29:39

I decided I don't want to do it anymore. And it's almost never the

1:29:41

case that I've decided I did not want to do something that I put

1:29:43

on my head. So you break the

1:29:45

goal down into is it that you start off with like the

1:29:47

10 year goal and you break it down into one month and

1:29:49

so on? Or is it like, like, well, how do you how

1:29:51

are you doing this? Okay, so such

1:29:54

a good question. So a lot of it is like, where's the

1:29:56

goal even coming from to begin with? And

1:29:58

so the most important skill in all of the The

1:30:01

most important skill for motivation, for goal setting, is

1:30:04

you need to hone the ability to dream. And

1:30:08

for me, it came from reading fantasy books. So

1:30:11

I love the way of kings. I love Caledon. The way

1:30:13

Caledon leads his group, I want to be that. I want

1:30:16

to lead with love. I want to be the

1:30:18

person that when someone else is not succeeding and the

1:30:20

world is shitting on them, that

1:30:22

I see the value in them, that I stand up for that person, that

1:30:24

I'm the one who picks them up. That's who

1:30:26

I want to be. And if I read

1:30:29

all the biographies that I read, and I've read hundreds

1:30:31

and hundreds and hundreds, my

1:30:33

two favorite people are not Malala,

1:30:36

but Malala's dad, if

1:30:39

you read her biography, and

1:30:41

Teddy Roosevelt's dad, who's

1:30:43

just such an incredible character, but also Teddy

1:30:45

Roosevelt himself. And when I think

1:30:47

about Teddy Roosevelt, he was super astigmatic when

1:30:49

he was young, had asthma, got really bullied,

1:30:52

got himself into weight and bodybuilding even before it

1:30:54

was like a thing. So

1:30:56

his famous president, like this really big chest. And

1:30:59

then literally at the

1:31:01

height of his political success, he left

1:31:03

his position to go lead a platoon.

1:31:06

He recruited himself in the war

1:31:08

in Cuba in the Spanish War. And

1:31:11

so the vision I've had of this man, and

1:31:13

he's amazing and dreaming, Teddy Roosevelt is like, Teddy

1:31:16

or me, like on top of some hill, big

1:31:19

chest, carrying a flag, and like a bunch of people

1:31:21

following to war with like some big goal. Like that's

1:31:23

kind of like the vision that I have for myself.

1:31:25

And then like being a good dad, by the way,

1:31:27

you asked, like, what's my definition of

1:31:29

success? Probably it's being a good dad, more than anything else.

1:31:31

I don't have kids. I'm not married. I won't be for

1:31:33

a while. But that's the

1:31:35

vision that I have for myself. Amar

1:31:41

called me the other day from yesterday, and he's like, Cliff, you know

1:31:43

who you remind me of? You're like

1:31:46

Andrew Tate, but standing for all

1:31:48

the opposite things. So like, I think like the definition of a

1:31:50

good man is just someone who's a good dad. And

1:31:55

that's the dream that I have of who I want to be.

1:31:57

And that's a very clear dream. flow

1:32:00

a lot of other things. So like well

1:32:02

that guy is well read. I couldn't read when I was

1:32:04

a kid. I had to practice really hard. That

1:32:07

guy you know is built. Well I was

1:32:09

very small when I was a kid. I had to work on

1:32:11

that. That guy has resources. I

1:32:13

didn't have resources growing up. Okay

1:32:15

well what are the other attributes that this person has?

1:32:18

And even if you took away all material positions, you

1:32:20

took away the ability to code, you took away really

1:32:22

most of the things. The tenacity in the drive and

1:32:24

when I talked about before the AQ is still there.

1:32:26

And by the way, Caledon,

1:32:28

Kelsier, all these characters, the

1:32:31

tenacity is the thing I admire the most. That's why

1:32:33

they're so likable to me. And

1:32:36

so the most important thing is to learn how to dream. And

1:32:42

have a clear vision in your head of who you want to

1:32:44

be. And so is

1:32:46

there a vignette in your head of you

1:32:48

doing a kroyoga on your back with someone

1:32:50

you know held up by your feet? Probably

1:32:52

yes. Great. Draw that.

1:32:56

Like make a poster of it. Like

1:32:58

I'm sure that you've trained Dali already

1:33:00

on your face. Tell it. Give

1:33:04

me a photo of Ali doing

1:33:06

a kroyoga. With an AI themed

1:33:08

vision board. Exactly. Why not? Yeah.

1:33:11

Nice. Just

1:33:14

visualize it. But also just FYI, you could do

1:33:16

a kroyoga right now. You can have your knees

1:33:18

bent a little bit. Yeah I mean I've doubled.

1:33:21

When did you develop this kind

1:33:23

of drive this vision led approach

1:33:26

to things? When I was

1:33:28

six or yeah. Like

1:33:31

I you know, no one

1:33:34

reads when they're not with no one but like reading is not

1:33:36

a thing when you're four but we

1:33:38

had I would draw we had like

1:33:40

acting and I really wanted to

1:33:42

get the solo. There was like a band

1:33:44

I was in called the Pune National and it was

1:33:46

like a singing dancing troupe and it was a solo

1:33:49

and I really wanted it and I didn't get it and so

1:33:52

what I did is I practiced for

1:33:54

like straight up the

1:33:56

week before but especially the 24 hours beforehand and

1:33:58

my thesis as a kid I didn't even know

1:34:00

where thesis meant, was if I do 24 hours

1:34:03

work, no one else will do this. I

1:34:05

will be the only person who worked this hard on the song.

1:34:08

And the audition happened and I

1:34:10

didn't get it. But it was so

1:34:12

clear to the people who ran the choir that I had

1:34:14

to outwork everybody else in the room, they just gave me

1:34:17

another solo. And

1:34:19

they realized that my personality was

1:34:21

so tenacious that actually there's

1:34:23

a certain MC acting role that they

1:34:25

just automatically started to give to me. And then I started

1:34:27

winning some of the solos and I was practicing so hard.

1:34:30

And by the way, interestingly enough, that's exactly what we talked

1:34:32

about before. When you just are willing to work harder than

1:34:35

everybody else, people

1:34:37

notice. And

1:34:40

you collect all the space that there is

1:34:42

where no one else is willing to go.

1:34:46

Right, that arbitrage. And so

1:34:48

I literally picked it up when I was a kid doing choir.

1:34:51

How do you think about the balance between self-improvement

1:34:54

versus self-acceptance? Wow.

1:34:58

Okay, I actually have a very strong thesis around this

1:35:00

too. I set

1:35:02

very tenacious, very ambitious goals. And

1:35:06

the second I do not accomplish them when the day comes,

1:35:09

I forgive myself. And

1:35:11

then I reset the goal for a week, a month, a

1:35:13

year later. And then I go for it again. I

1:35:16

don't punish myself internally ever for setting a

1:35:18

goal and not achieving it. Even

1:35:21

if I got distracted, I didn't put enough work. There's

1:35:24

a reason why, you know, it went somewhere else,

1:35:27

like there was another priority, et cetera. Because

1:35:31

the desire is there. The desire is always there.

1:35:34

And things will always get in the way. You'll

1:35:36

have dyslexia, you'll be short, you'll get injured. You

1:35:40

know, I had a bunch of goals around weightlifting and

1:35:42

I injured my shoulder and for a year, I couldn't

1:35:44

really lift. And I couldn't really train tricking. But that's

1:35:46

okay, I'm not mad at myself. But

1:35:48

did I do my best to heal myself? Did

1:35:51

I go to every doctor who could help

1:35:53

me? Did I ice my shoulder? Did I

1:35:55

do the stretches? Absolutely. This

1:36:00

is like, it's not fair

1:36:02

for me to talk about the things that I'm good at. Let's talk about the things I'm bad

1:36:04

at. I'm so bad at going to sleep. I

1:36:07

have no desire to go to sleep. I'm so excited to wake

1:36:09

up and I don't want to go to sleep because there's so

1:36:11

many things I'm interested in doing. I felt guilty

1:36:13

for not doing the things I was supposed to do,

1:36:16

specifically going to sleep. So what I did is

1:36:19

I got a safe and I looked at

1:36:21

my phone and my computer in the safe every night before I

1:36:23

went to bed and I put an eight hour timer on it.

1:36:25

And then I configured my computer and

1:36:27

my phone. Everything locks at

1:36:30

midnight except for Google Maps and

1:36:33

Uber. And I just can't

1:36:35

use technology after midnight. I have no choice. I can

1:36:37

either play guitar or journal or go to sleep. That's

1:36:40

it. And so in the places where I'm

1:36:42

not good, especially because I have ADHD at

1:36:44

self control, I just set up

1:36:48

habits and mechanisms that make it so that I can't

1:36:51

fail. For the

1:36:53

same reason I don't drink alcohol, I

1:36:55

don't smoke weed, I just don't indulge

1:36:57

in things that are dangerous

1:36:59

to me and my personality. What's

1:37:04

the deal with ADHD? So

1:37:07

attention deficit disorder. It

1:37:11

is defined by a challenge in executive

1:37:13

functioning in the brain. So

1:37:16

your brain tells you do this and yes

1:37:18

you don't. The

1:37:21

regulation of dopamine is different in brains of people

1:37:23

with ADHD compared to people without ADHD. One

1:37:26

easy test to figure out if you have ADHD is when

1:37:28

you drink coffee is it easier for you to concentrate and

1:37:30

now you're more calm as a result of coffee. There's

1:37:35

a bunch of medication you could take,

1:37:37

Adderall, Ritalin, Concerto, My Vance. I

1:37:39

don't take any. I personally don't recommend it. Obviously

1:37:41

there's exceptions. One of the songs I'm working on

1:37:44

right now is about this topic. Retie

1:37:47

was about dyslexia. This other

1:37:49

song, Fire in My Chest, is about ADHD. At

1:37:53

the same time, kids with ADHD, they're

1:37:56

not stupid. They're very, very smart. And

1:38:00

they can hyper focus on the things that

1:38:02

they care about and that they're into. And so you

1:38:04

see me doing this all the time. I hyper focus on things, but like

1:38:07

if I don't want to do something, it's so difficult to

1:38:09

get myself to focus on it. And

1:38:12

you just got to exactly accept yourself. Say,

1:38:14

hey, it's fine. I have dyslexia. I

1:38:17

have ADHD. Cool.

1:38:19

What now? Well, I'm going to listen to

1:38:21

audiobooks. I'm going to convince kids in

1:38:23

my class to come over to my house and

1:38:26

study together, but really, they're going to read the book.

1:38:28

And I'm going to listen and explain what I understood.

1:38:30

And gosh darn it, if

1:38:32

there's no audiobooks, I'll just build my own software that

1:38:35

makes it easier for me to focus because it highlights the words

1:38:37

on the screen as it reads. And like, even if I get

1:38:39

distracted because there's a donut, it's still reading

1:38:41

in my ears and I can come back and I don't need to restart my place.

1:38:45

I'm going to commit to friends that I'm going to do it. And if

1:38:47

I'm not, I'm going to run. I'm going

1:38:49

to work out really hard every single day because by doing

1:38:51

that, I don't need to take medication.

1:38:56

Yeah, it's on a spectrum. So,

1:38:59

you know, different people who

1:39:01

wear glasses have different prescriptions because they

1:39:03

have different intensity of how

1:39:07

their eye is formed. Some people

1:39:09

are very ADHD. Some people are like mildly

1:39:11

ADHD. If you think you have it

1:39:13

and you're still in school, I highly recommend to go get diagnosed.

1:39:18

It is helpful to have the diagnosis because it helps

1:39:20

you understand yourself a little bit more. That's definitely the

1:39:22

key for me. There's

1:39:24

a thing I've heard on the grapevine recently, which

1:39:26

is that I'm not sure

1:39:29

to what extent this is true at all, but that loads

1:39:32

of people are now or the straw man would

1:39:34

go that like loads of people

1:39:36

now are self-diagnosing themselves with ADHD because they think I

1:39:38

struggle to focus when I'm bored of what I'm doing.

1:39:42

But everyone's struggling to focus when they're bored of what they're

1:39:44

doing. So, like, why are we trying to medicalize something which

1:39:46

is very natural? I

1:39:48

don't know anything about this topic, so I'm not curious. No, no, it's

1:39:50

a good question. So,

1:39:52

interestingly, dyslexia and ADHD are similar

1:39:55

in this regard where it's an invisible

1:39:57

learning difference. see

1:40:00

it in a person. Everybody

1:40:04

has challenges focusing, yes, and by

1:40:06

the way, so the other

1:40:08

problem is like we can't measure these things

1:40:11

well, right? Like I said ADHD

1:40:14

is caused by a regulation

1:40:17

of dopamine in the brain, but like we have

1:40:19

no idea what's going on in the neural cleft

1:40:21

at all. So it's not something you can measure

1:40:23

on a spectrum. Dyslexia,

1:40:25

the best literature indicates that it's

1:40:28

defined by, there's a thing called

1:40:30

mini columns in the brain where

1:40:33

if they're really short, you have autism, if they're

1:40:36

really long, you have dyslexia and further apart. So

1:40:38

hypothetically, eventually we'll have a way of

1:40:41

measuring people's dyslexia. I

1:40:44

don't think regulation for ADHD. So yes,

1:40:46

there might be more ADHD, less ADHD,

1:40:49

you might cross the spectrum, whatever. I

1:40:52

have friends who find it very easy to focus

1:40:55

on things and they tell themselves they

1:40:57

want to do this thing. Like their behavior is very aligned

1:41:01

with their brain's instructions to them.

1:41:03

By the way, people with ADHD

1:41:06

are much more susceptible to addiction.

1:41:09

They have much more probability of

1:41:11

having really big mood swings, being super happy

1:41:13

or super mad in a very short period

1:41:15

of time. Interestingly,

1:41:17

they don't hold grudges very much. Very

1:41:21

happy-go-lucky often, considered

1:41:23

irresponsible often. And

1:41:29

there's a bunch of reasons why someone might be like this. It

1:41:31

might be that you didn't sleep enough, it might be that you

1:41:34

have a headache, it might be that you're a period, it doesn't

1:41:36

matter what. But worthwhile to

1:41:38

go potentially get diagnosed, we just don't know

1:41:40

enough about the brain. Bye-bye. What

1:41:43

are the diagnosis involved? You

1:41:45

go and you talk to a psychiatrist. And

1:41:49

if you're a kid, you'll be there for like

1:41:51

six hours, maybe over two sessions. And

1:41:54

they'll give you a bunch of like little tests and

1:41:56

they'll see how difficult or not difficult it is for

1:41:58

you to complete them. And

1:42:00

it could be something like playing with cubes, or

1:42:03

often an inclusive sort of IQ test. For

1:42:05

dyslexia, I can just tell you mine. I

1:42:10

scored in every category in the IQ tests