0:00

This episode is brought to you by Protect's

0:02

REST Supplement, a new take on getting deeper

0:04

and more restorative sleep. I was introduced to

0:06

this by former Navy CEO, Nick Ferriss, who

0:08

has been on this podcast. And as

0:10

a mutual friend of ours put it to me,

0:12

it's very annoying that Nick has no

0:15

physical weaknesses. This is somebody

0:17

who can climb like

0:20

a spider. He has incredible

0:22

abilities as rock climber. He is incredibly

0:25

wrong and can also run ultra marathons.

0:27

It's quite something to behold. And this is one of

0:29

the tools that he uses. So protects rest supplement

0:31

and help provide consistent restful sleep

0:34

without any habit forming ingredients or

0:36

groggy side effects. Simply add it to your

0:38

last glass of water before bed and it goes

0:40

to work. So I tried

0:42

this after I mentioned to Nick that

0:44

I was avoiding melatonin due to next day

0:46

sluggishness and its possible effects on testicular

0:49

functions. So I was trying to weed myself off of

0:51

things like that. This rest

0:53

cocktail has worked wonders and

0:55

I've made it part of my sleep

0:57

toolkit. Is literally on counter

0:59

about twenty feet from where I'm recording

1:02

this right now. Pro tip number one, if you

1:04

have trouble opening the packets, just use

1:06

scissors. Don't make it hard. Pro tip number

1:08

two, mix the packets with water.

1:10

Do not skip that step and shug it alone.

1:13

I've learned from experience. Rest has

1:15

no added sugars, artificial sweeteners,

1:17

or artificial ingredients. Protect is

1:19

veteran owned and they make all of their products right

1:21

here in the USA. And right now,

1:23

Protect is offering you thirty percent

1:25

off both flavors of their rest formula.

1:28

Visit protect dot com slash Tim

1:30

that's spelled PR0TEKT

1:33

dot com slash Tim to claim this special

1:35

offer today. One more time, protect

1:37

dot com slash Tim, PR0TEKT

1:41

for thirty percent off. Use code, tim

1:43

at checkout, and just

1:46

to remind you, protect dot

1:48

com. That's protect with a k. Dot

1:50

com slash Tim and code Tim.

1:56

This episode is brought to you by Shopify

1:58

Shop is one of my favorite companies out

2:00

there, one of my favorite platforms ever.

2:03

And let's get into it. Shopify is a platform,

2:05

as I mentioned, designed for anyone

2:08

to sell anything anywhere, giving

2:10

entrepreneurs the resources once reserved

2:12

for big business. So what does that mean? That

2:14

means in no time flat, you can have a

2:16

great looking online store that

2:18

brings your ideas, products, and

2:20

so on to 651. And you can have the tools

2:22

to manage your day to day business and drive

2:25

sales. This is all possible without

2:27

any coding or design experience whatsoever.

2:30

Shopify instantly lets you accept all

2:32

major payment methods, shop FIUS thousands

2:34

of integrations and third party apps

2:36

from on demand printing to accounting to

2:38

advanced chat bots, anything you can

2:40

imagine. They probably have a

2:42

way to plug and play and make it happen.

2:44

Shopify is what I wish I had had

2:46

when I was venturing into e commerce way

2:48

back in the early two thousands. What they've

2:50

done is pretty remarkable. I first met the founder

2:53

Tobey in two thousand eight when I became an

2:55

adviser, and it's been spectacular.

2:57

I've loved watching Shopify go from

3:00

ten to fifteen employees at the time to

3:02

seven thousand plus today, serving customers

3:04

in one hundred and seventy five countries with

3:07

total sales on the platform exceeding four

3:09

hundred billion dollars. They power

3:11

millions of entrepreneurs from their

3:13

first sale all the way to full scale, and

3:15

you would nice a lot of large companies and also

3:17

use them who started small. So

3:19

get started by building and customizing your

3:21

online store again with no coding

3:24

or design experience required. Access

3:26

powerful tools to help you find customers,

3:29

drive sales, and manage your day

3:31

to day. Gain knowledge and confidence

3:33

with extensive resources to help you succeed.

3:35

And I've actually been involved with some of that way

3:37

back in the day, which was awesome,

3:40

the build to business competition and other things.

3:42

Plus, with twenty four seven support,

3:44

you're never alone. And let's face

3:47

it, being an entrepreneur can be lonely, but

3:49

you have support. You have you

3:51

don't need to feel alone in this case.

3:54

More than a store Shopify grows with

3:56

you and they never stop innovating providing

3:58

more and more tools to make your business

4:00

better and your life easier. Go to shopify

4:03

dot com slash Tim to sign up

4:05

for a one dollar per month trial period.

4:07

It is a great deal for a great service,

4:10

so I encourage you to check it out. Take your business

4:12

to the next level today and learn more

4:14

by visiting shopify dot com slash

4:17

Tim. One more time, shopify dot

4:19

com slash Tim, all lowercase.

4:21

Optimal, minimal. At this

4:23

altitude, I can run flat out for a

4:25

half mile before my head starts shape.

4:28

Can I answer your question?

4:46

Hello boys and girls, ladies and germs.

4:48

This is Tim Ferriss. Welcome to another episode

4:50

of the Tim Ferriss show where it is my job to

4:52

de construct world class performers across

4:54

many, many disciplines by guest

4:56

today. I'm so happy to have him. Is

4:58

Bill Girley, you can find him on twitter

5:00

at B Gurley, that's GURLEY.

5:03

Bill has spent more than twenty years as a general

5:05

partner at Benchmark. Before entering the

5:07

venture capital business, Bill spent four

5:09

years on Wall Street as a top ranked research

5:11

analyst, including three years at Credit

5:13

Suisse First Boston. Bill also maintains

5:15

a blog on the evolution and economics

5:17

of high technology businesses called above

5:19

the crowd, which you can find at above

5:21

the crowd dot com. Over his venture

5:24

career, he has worked with such companies

5:26

as Grubhub next door open table,

5:28

Stitch Fix, Uber, and Zillow,

5:30

among many others. Bill has BS in

5:32

computer science from the University of Florida and

5:34

an MBA from the University of Texas.

5:36

He's also chartered financial analyst.

5:38

Bill is a board trustee at the Santa

5:40

Fe Institute, a research and education

5:43

center focused on the study and understanding of

5:45

complex adaptive

5:46

systems. Bill, nice to see

5:48

you. Thanks for making the time. Thanks

5:50

for having me. Alright. So I

5:52

wanted to start with

5:54

a number of things that I've seen pop up repeatedly

5:57

in various forms

6:00

in doing prep for this conversation.

6:02

And the first I wanted to get

6:04

your take on and expansion on

6:06

is Michael Porter's book

6:08

competitive strategies, subtitle techniques

6:10

for analyzing industries and competitors.

6:13

Which you have described I believe

6:15

as the most efficient short form

6:17

NBA that one can find.

6:19

Could you please explain what

6:22

this book is. And if you

6:24

still hold that opinion, why you hold that opinion?

6:26

If you do go to get an MBA, like,

6:29

in the first semester or whatever.

6:31

They usually make you take a corporate strategy

6:33

course, and it's the first book assigned.

6:35

It's like extremely well known

6:37

business book. There's a

6:39

belief in Silicon

6:41

Valley that an NBA is

6:43

worthless and that people shit

6:45

on all the time. Yeah. Right.

6:48

Yet. I would say eighty

6:50

or ninety percent of the entrepreneurs

6:52

I meet would benefit

6:55

greatly by reading the first three

6:57

chapters of this book. And -- Mhmm.

6:59

-- it's really just about trying

7:01

to understand the dynamics of

7:03

industry. One of the most

7:06

common mistakes entrepreneurs make

7:08

is they come up with some kind of technological breakthrough

7:11

in their own mind. But they don't

7:13

spend any time analyzing the industry

7:16

structure or whether the go to market

7:18

is gonna be possible or not. And

7:20

The book just has a wonderful framework for

7:23

thinking about the competitive dynamics of

7:25

an industry and whether or not you'd be

7:27

able to break in or be successful

7:29

or maintain success. This

7:32

actually makes me think

7:34

more broadly of frameworks and

7:37

recipes as it were for

7:40

making decisions. So back in the day, you

7:42

are a sell side analyst, and I would love

7:44

for you to define what that is and also

7:46

just define buy side versus sell

7:48

side. And at

7:50

that time, at one point, at least,

7:53

the top three analysts covering

7:55

the PC industry retired and

7:57

reading from a trans script

7:59

on medium. Dan Benton,

8:02

David Kors, if I'm correctly pronouncing

8:04

that, and Charlie 651.

8:06

And Then Charlie became the advisor

8:08

for young analysts, so you got to work with

8:10

him. Here's the part that's of interest to me.

8:12

So, Dana, David became friends, so they both gave

8:14

me all of their financial models.

8:17

And I'm curious what they gave you. What

8:19

were those models? And

8:21

do you still use any of those models?

8:24

No. No. No. So they're not specific.

8:26

But let me let me start with the sell

8:28

versus buy. So and the industry has changed

8:30

over the years. But a sell side analysis

8:32

is someone who does research

8:34

on behalf of an investment bank

8:37

that is presumably gonna make

8:39

money from trading on their

8:41

trading desk. That someone send

8:43

your way because they valued the

8:45

research. So you're providing research

8:47

to help sell stocks. Like, that's how

8:49

you're gonna get paid. The buy side

8:52

is anyone at a mutual fund who's

8:54

trading for their own account. If you're a

8:56

buy side analyst, you don't publish. Like, it

8:58

doesn't get public because you're

9:00

using that as proprietary information.

9:02

The sell side stuff

9:04

gets broadcast broadly. And

9:06

usually when someone changes

9:08

estimates or you hear about a rating changed. That's

9:10

usually a sell side analyst that has

9:12

made made a decision to

9:14

change something, and that's what ends up on

9:16

CNBC. And so a sell

9:18

side analyst, this is gonna sound very

9:20

naive, but a sell side analyst could

9:22

make a recommendation to buy. Right? They could

9:24

make a recommendation like, a strong buy recommendation

9:26

or a strong sell or somewhere in between

9:28

the two. That's, I think, where my Long

9:30

Island brand was having trouble -- Yeah. --

9:32

trying to reach the basics. About that. Because the

9:34

bank makes money in multiple ways and they

9:37

question whether people are biased or not. And

9:39

then you had the whole Henry budget

9:41

thing from the two thousand

9:43

period So it's complicated.

9:45

And could you describe I know

9:47

that you don't, at this point, use

9:48

them any longer, but the company specific

9:51

models. What were the models I

9:53

was super fortunate in my entire

9:55

career by being in

9:57

places where windows opened up that I was

9:59

able to take advantage of, but the

10:01

situation you just scribed I landed on

10:03

Wall Street, very fortunately,

10:06

started covering an And

10:08

there's a magazine called institutional investor

10:10

that pulls the buy side and ranks the

10:13

analysts and the top three analysts

10:15

in the category and it all retired

10:17

within one year. And they all

10:19

became friends, partially from my

10:21

own networking like I was reaching out to them,

10:23

but the models they shared were the models they

10:25

had built on the companies we

10:27

all covered. So their own

10:29

particular version of Dell's financials

10:31

are ComPact or Microsoft.

10:33

And so it just helped me have a

10:35

better understanding. I could see what they

10:37

did, that kind of thing.

10:39

Got it. What I'm trying to unpack

10:41

also through the course of this conversation is

10:43

just how you think about models

10:45

and making decisions. I find that interesting.

10:48

So this is hopping around a little

10:50

bit, but you did not pursue Google

10:52

as an investment in two thousand

10:54

two. And you

10:56

had some investing rules of thumb in place.

10:58

And I suppose what I'm wondering is

11:00

how you think about rules

11:02

as an investor. And since it

11:05

seems like perhaps to be

11:07

consistently great, you need rules. But if you

11:09

have rules, you're inevitably gonna

11:11

some great opportunities, but perhaps that's okay.

11:13

How do you think about rules

11:15

like was missing Google, the result

11:17

of a flaw in the set

11:19

of rules that you had at the

11:21

time? Or was it just a

11:23

collateral miss

11:26

let me work my way back to it because it's

11:28

clearly the biggest mistake of my career. So I

11:30

it's something I thought about a lot. Yeah.

11:33

So I think any investor starts

11:36

with just building bedrock

11:38

and that comes from reading. And

11:40

there's a ton of books

11:42

in history you can read. You can

11:44

go through, you know, all the

11:46

Buffet letters as an example. You

11:48

know, you can read Peter Lynch's one

11:50

up on Wall Street. There's my good

11:52

friend, Mike Mobison, has put out some

11:54

amazing books. There are a lot more

11:56

nuanced about stuff rock prices. That kind of

11:58

thing, a random walk down Wall Street

12:00

by Burton McHale. And then you

12:02

build your bedrock, and then, you know,

12:04

you're out there. And you're

12:06

looking for an

12:08

opportunity. So I think public stocks are a

12:10

little different than venture, but there's some

12:12

overlap. But you're looking for a

12:15

reality that you think is gonna

12:17

emerge. It's not priced into the stock.

12:19

So that requires you both to think

12:21

you know where the world's going. And

12:23

also to know what the expectations are

12:25

currently embedded in the stock.

12:27

Because if you just have the same opinion that's

12:29

already embedded, you're not gonna make any money.

12:31

There's a a great piece by Howard Marks

12:33

where he talks about. You have to be right

12:36

and contrarian. You know?

12:38

You can't just be right. You have to be

12:40

right and contrarian, and that's

12:42

more difficult. There's a famous

12:44

saying that I'm sure has been uttered on your

12:46

podcast before, but strong opinions,

12:48

Lucy Hell. And -- Yes. --

12:50

I think I think all investors have

12:52

to work within that framework because

12:54

things change. There's many, many

12:56

variables none of them are constant. They're all

12:58

dynamic. And the minute

13:00

you said a very hard

13:02

rule, then you might be

13:04

sending yourself up. For a mistake.

13:07

And venture I have found

13:09

is a world where

13:11

that happens frequently. And

13:13

so the Google example

13:15

And it's really important to say

13:17

that we didn't lay chase. Like, they

13:19

presented there were twenty five employees there and

13:21

Sergei presented. They presented

13:23

to, I'm sure, a whole bunch of firms and

13:25

lay chase. We should have laid chase, but we

13:27

didn't. At the time, Yahoo! Was

13:29

at ten dollars down from like

13:31

eighty excite was going

13:33

bankrupt. So search didn't look that

13:35

exciting from an external

13:38

viewpoint. You had two

13:40

PHD founders who had never been a

13:42

CEO before and were insistent they were gonna be

13:44

good as a CEO. Normally,

13:46

that's a red flag. You know,

13:48

two PhD founders. There's just a

13:50

number of things where you would make a list

13:52

and say, oops, you know, this probably isn't

13:54

what you wanna do. Now, I'd

13:56

like to highlight and I always do the

13:58

two best venture capitalists in the world at the

14:00

time John Door, Mike Mritz, kind

14:02

of locked hands and said yes.

14:05

Right? And so it'd be

14:07

erroneous for me to say, like, oh, if you

14:09

were a well studied venture capitalist,

14:11

you obviously get to know at that point

14:13

in time because the two best

14:15

didn't. And I think it's just the subtlety of

14:17

the game. There's a great

14:19

reality adventure that a lot of people

14:21

talk about now, but you can only lose one

14:23

time your money. And

14:25

in a case like Google, you make,

14:27

what, ten thousand times your money.

14:30

And that asymmetric result

14:32

means you have to bias

14:34

towards positive in a

14:36

situation like that. The odds are just

14:38

ridiculously different. And so

14:40

what should have happened is I remember one of my

14:42

partners looking at Larry and saying,

14:44

what does it take to get this deal done

14:46

right now? Because he was a

14:48

closer he was a closer type. And

14:50

Larry said 651 hundred and twenty pre.

14:53

And we should have said, how about

14:55

one fifty? Like, if I could go

14:57

back if I

14:59

could go back. Right? But you can't go

15:01

back. Deal got done at, like, eighty, so

15:03

below what Larry had told us,

15:05

but still, obviously, historic.

15:08

So how did you then

15:11

revise your rules of

15:13

thumb or rules

15:16

moving forward after that, if at all,

15:18

like, is a case like that, granted it

15:20

turned into what Google

15:22

is now. So it's easy to maybe

15:24

punish yourself for not

15:26

laying chase. But how did

15:28

things change after that? You know,

15:30

after that, the firm did

15:32

Twitter snapchat. Uber,

15:35

you know, we got some right, you

15:37

know, after that. Yeah. And

15:40

around that time, I remember

15:43

we used to give a book out to our

15:45

LPs, our limited partners, our investors

15:47

at every annual meeting. And I

15:49

think Bruce had just read directional

15:51

optimist, which is a Matt Ridley book. And

15:54

he started using a phrase at our

15:56

partner meeting. What could go right? Uh-huh.

15:58

And because of this asymmetric

16:01

outcome thing where you could

16:03

make ten thousand times your money and only

16:05

lose once on the downside. It was the

16:07

right frame of mind. So

16:09

it's very easy to get into a

16:11

trap adventure where getting

16:13

no right feels like a win.

16:15

And it's just not yet. I mean, obviously, you can't

16:17

do every deal. You know, you can't do every investment.

16:20

You go broke, but getting

16:22

overly jazz

16:24

about correctly identifying

16:26

a negative or a

16:27

no. It's just not that big a deal. It's not

16:29

the job. The job is

16:31

to find outliers.

16:33

So my my experience

16:35

granted it's limited in some respects.

16:37

My experience and perception of

16:39

benchmarks. You guys are very, very selective.

16:42

More selective than a lot

16:44

of venture capital firms and a

16:46

lot of angels. Who maybe like

16:48

the poker player who doesn't mind his or her

16:50

bankroll. Yes, you can only use one x your

16:52

money, but if you do lose all of your

16:54

money, the jig is up. Yep.

16:56

That's the problem with the complete

16:58

opposite there. You can't just do every

17:00

investment, you mean? We have another

17:02

challenge, Tim, which is we've

17:04

for a variety of reasons have chosen a strategy

17:07

where we don't let our money walk around

17:09

without our work product and our

17:11

involvement. And so Right?

17:13

We go on a board if we make an

17:15

investment and we usually become the

17:17

largest shareholder on the board. And

17:19

as a result, our limitation is

17:22

our board seats more than the

17:24

capital, actually. You can't do -- Yeah.

17:26

-- twenty of them. Not well.

17:28

So you mentioned Howard Marks, Oaktree

17:30

Capitol, who's been on the podcast

17:33

twice, very, very impressive

17:35

fellows. So I've seen you

17:37

mention investors, I'm sure you've

17:39

mentioned more, but two have come up repeatedly

17:42

in various interviews and so on

17:44

Howard Marks and Stan

17:45

Drucker. Could you please speak

17:48

to what makes them impressive

17:50

or interesting to you? Sometime in the

17:52

last four or five years when interest wouldn't get

17:54

away from zero for so long.

17:56

I told myself, I have to learn

17:58

more about macro because I

18:00

just know nothing and it's causing all kind

18:02

of problems in my industry. I had been

18:04

reading Howard to work for years. One of the things

18:07

I just love about

18:09

people like Howard and Buffet does it,

18:11

but, like, people who

18:14

archive their thought

18:16

process, just as part of their own

18:18

process. But it's quite

18:20

kind if you're a learner in

18:22

the field to have someone of

18:24

that capability that

18:26

insist upon writing these letters

18:28

and making them public, which

18:30

they've done. I started reading Howard's work

18:32

when I was on the sell side twenty

18:34

five years ago, and I've gotten to know him

18:36

now. And he's he's in addition

18:38

to being very, very he's also just a

18:40

wonderful human and fun to

18:42

interact with. There aren't many

18:44

people. If you study financial

18:46

history, most people in

18:48

Buffet included will tell you macro's

18:51

impossible. Like, you shouldn't even try. And

18:53

the two individuals you mentioned are

18:55

are two of the only ones that

18:57

are known for being successful

18:59

in macro investing, powered mostly

19:01

by being one of the

19:03

most successful and and longest

19:05

tenured investor in the bond market and

19:07

stand for taking more kind of single

19:10

individual bets that are macro

19:12

and naturals going back to his

19:14

success with

19:15

Soros. He's become

19:18

kind of refamous in

19:20

the past twelve months for predicting

19:22

the inflation situation we're in and

19:24

being very loud about it. Could you, if you

19:26

don't mind, just give a definition, it

19:28

could be simple of macro investing for people

19:30

who may not know that

19:31

term? If you go to to

19:34

business school, there's two economic

19:36

clashing, take microeconomics and

19:38

macroeconomics. Microeconomics is

19:40

a lot about what's discussed

19:42

in competitive strategy, the Michael Porter

19:44

book. So it's about the interaction between

19:47

firms within an industry pricing

19:50

that kind of thing. Competitive dynamics.

19:52

Macro is the study of

19:54

economies, you know, writ large.

19:57

And I'm fascinated with

19:59

complex systems and our

20:01

economy is certainly one of those things

20:03

as is weather and whatnot, which is why

20:05

I've gotten involved with the Santa Fe

20:07

Institute, but they're

20:09

nearly impossible to predict

20:11

and where you get into real,

20:13

real trouble. There might be a variable

20:15

you're not tracking that

20:17

has never flipped from

20:19

zero to one. And when it

20:21

does slip from zero to one, all your

20:24

models, all your planning are out the

20:26

window because this other thing's different

20:28

this time. And you can I mean, it's why no one can

20:30

predict the weather more than five days

20:32

in front of us? Right? Because there

20:34

is just two dynamics too

20:36

complex, and most people feel that way about

20:39

macro. Let me go back in time a little

20:41

bit, revisiting the the

20:43

sell side analyst side of

20:45

things. So I believe at one

20:47

point, and I could be getting the terminology wrong, but you

20:49

were an institutional investor, which you already mentioned,

20:51

institutional investor all American or

20:53

all Ferriss research team.

20:55

Parts because those people left the field. But,

20:58

yes, what allows

21:00

someone to land on

21:02

that list? So it's a

21:04

poll of your customers. It's a

21:06

poll of the buy side. So they

21:08

literally pull the buy side in each

21:10

industry and say who was most

21:12

helpful. I will tell you that and

21:14

I don't know. This probably came from

21:16

some book I read in business school, but when

21:18

I showed up and this is

21:21

framework someone could use. But when I showed

21:23

up on Wall Street in

21:25

one of the very first weekly meetings,

21:27

they introduced us to the Salesforce. The

21:29

Salesforce is the

21:31

individual at your firm responsible

21:33

for that account, for fidelity, for

21:35

Wellington, for teachers,

21:37

you know, of Texas or whatever,

21:39

and they own the relationship. And

21:42

somewhere in my youthful

21:44

wisdom, I decided to ask each

21:46

salesperson, is there

21:48

one client that will spend

21:50

thirty to forty five minutes with me as

21:52

a new analyst, and just tell me

21:54

what they want. Like, I'm just

21:56

gonna ask them questions. I'm not gonna have

21:58

anything for them. I just wanna know how

22:00

I can serve them best. And I

22:02

did, like, twenty of those interviews.

22:05

Before I had started the job.

22:07

That's a roundabout answer to how do you

22:09

how do you get on the list? Because I knew what

22:11

they were looking for at that point

22:13

in time. What made you a good analyst? And let me explain why

22:15

I'm asking. I'm asking because one of the

22:17

advantages that you bring to a board, which you've

22:19

already discussed, to a board of

22:21

a startup in the art of building

22:23

is your in-depth understanding of how the

22:25

public markets work. And if a

22:27

company is aiming for going

22:29

public and so on. You have knowledge that

22:32

scares a lot of other folks who

22:34

might intimidate or

22:37

be fuddles folks who might

22:39

otherwise wanna be on the board. So so

22:41

I'm curious to know what made you a

22:43

good analyst. The thing I learned

22:45

in those interviews both

22:47

in terms of the because I would also ask him

22:49

who does a really good job of this, and then

22:51

if I could, I'd try and be friend that

22:53

person. But what I heard

22:55

frequently back was and this

22:57

is a little different from what the world

22:59

perceives. You know, I don't

23:01

really need you to make this buyer

23:03

sell recommendation like what I

23:05

would really get a huge benefit

23:07

from is if you provide a

23:09

point of view or a piece of

23:11

analysis that causes us to

23:13

think differently about a particular

23:15

company or industry. Can you go off and

23:17

do some work that other

23:19

people haven't thought of that causes us

23:21

the question that makes us want to talk

23:23

to and hear what you're thinking. And

23:26

so that became really

23:28

the essence of of what I was focused

23:30

on it. Partially from listening to

23:32

them and partially from mimicking

23:34

and copying David Kors, who you

23:36

mentioned, I started doing a weekly

23:38

fax at the time. Mhmm.

23:41

And David was doing that before

23:43

he quit. And so I started

23:45

doing it too. And interestingly,

23:48

the Wall Street

23:50

firm tries to keep your

23:53

content closed within

23:55

their customer and

23:57

this is not very loyal to my firm,

23:59

but it it became very obvious to

24:01

me that a sell side analyst that

24:03

was more well known was

24:06

more powerful, more impactful.

24:08

And so I intentionally

24:10

started expanding the distribution of

24:12

this weekly piece as as far and wide

24:14

as I could. How did you

24:17

expand that distribution? I

24:19

would leverage the Salesforce and get them

24:21

to give me fax numbers. That's where

24:23

I started. I then started developing

24:26

industry relationships, which

24:28

is important because you're covering these

24:30

companies, you start going to investor

24:32

days, the buy side is talking

24:34

to these companies as well, and so I started getting

24:36

some of them onboard. And

24:38

then probably the most successful hack

24:41

of my career happened when

24:43

I was invited to attend

24:46

Stuart Alsop's agenda

24:49

conference. Which I think was in Phoenix at the time and used to

24:51

be it was the conference everyone

24:53

went to. So Gates would be in the front

24:55

651. Ellison would be

24:58

there. They would stay for the whole

25:00

thing. You could walk up and talk to

25:01

him. Michael Dell was there. Like, everyone

25:04

was there. How

25:04

many analysts were invited to such an

25:07

event? I don't know. Charlie had been

25:09

invited, and Charlie got me in.

25:11

And I don't even remember. Rick Cherilyn was probably

25:13

there. He's a famous Microsoft

25:15

analyst. And, you know, it's kinda what code

25:17

is today, but their

25:19

difference back then way back

25:21

then was the most famous

25:23

founder, CEO, set through the

25:25

whole thing and were available the whole time.

25:27

Today, if one of them speaking,

25:29

they come into back door, they

25:31

go on stage, they leave. Like, they're

25:33

not around. So it was pretty

25:35

cool. What was the hack at the

25:37

conference? So this was around

25:39

the time where the Palm Pilot launched.

25:41

And so in the lobby of the

25:43

conference, they were selling them. Weren't free.

25:45

They were selling selling a Palm Pilot. And

25:48

I think it was, like, two hundred dollars, three hundred

25:50

dollars, I can't remember. But

25:52

they had put the contact information for every

25:54

attendee at the conference in

25:56

the Palm Pilot. And

25:59

so I ran some quick math people

26:01

weren't really doing, like, cost of customer

26:04

acquisition back then, but I think it was,

26:06

like, seventy cents

26:08

a name. Or something like that

26:10

of the most influential people in the

26:12

entire tech industry. And

26:14

so I bought the palm pilot. I

26:16

took it home and I spam

26:18

the four, five hundred most important

26:20

people in the tech industry with

26:22

my weekly newsletter. Did

26:24

you end up developing

26:28

close or closer relationships

26:30

with any of them because of

26:32

that newsletter? I think so. And I think

26:34

also just a reputation. Right?

26:36

It's actually not that surprising today.

26:38

Right? There's plenty of content influencers

26:40

all over the place now. So stack

26:42

and everything. It was just -- Yeah. -- a

26:44

version of that when there weren't as many people

26:46

doing it. It started with facts and became

26:48

email. And today, I just mostly

26:50

tweeted. I I don't force the distribution out of

26:52

these other things anymore. And I'm not as

26:54

frequent. But but I mean, it's a

26:56

common way to build

26:58

reputation in a network I

27:00

mean, you do this with own life. So Yeah.

27:02

I do. I do. So

27:05

so let's come back to the

27:07

point you made of helping your clients

27:10

think differently, presenting that was

27:12

something that helps them to think differently. And I'd

27:14

like to Connect

27:16

that to story I'd love you to

27:17

tell, which is about

27:20

what you learned from a food

27:22

analyst. And the question that might

27:24

come before that is how did you meet the food

27:26

analysts? Why were you having a

27:28

conversation? Well, first of all, all the analysts

27:30

share a floor. So

27:32

we're all physically proximate. So the person

27:34

that covers food or telcos

27:36

or electric utilities were all in

27:38

the same group and -- Mhmm.

27:40

-- the firm had efforts

27:43

underway which Charlie Wolf ended up

27:45

running to try and

27:47

educate the analyst

27:49

in a common framework or whatever so that we

27:51

would all be better at what we do and

27:53

that kind of thing. And

27:56

for reasons that I I I guess birds of a feather

27:58

flock together. For reasons I don't

28:01

remember me and

28:03

my Moses and the food analyst started spending a lot

28:05

of time together. And we're still close friends

28:07

today. He's involved at Santa Fe, and and

28:09

I see him frequently. And

28:11

he's done amazing things himself 651 several

28:14

books, and and he's quite well known.

28:16

So we started hanging out together.

28:18

He had just read a

28:20

bunch of books on a framework called return

28:22

on invested capital, which --

28:24

Mhmm. -- Stearns Stewart had published on Mackenzie

28:27

had a book called valuation. They

28:29

still do. That uses this type of

28:31

analysis and he was spreading

28:33

it was the word proselytizing through

28:36

the analyst group. And

28:38

you know, I was a sponge at that point in time.

28:40

So it says, sure. So I took the

28:43

framework and ran it on all of my companies.

28:45

It turned out just

28:47

by happenstance that

28:50

Dell stood out like a sore

28:52

thumb with ridiculously

28:55

high ROI c numbers versus the

28:57

rest of the industry. Like, at night and day,

28:59

like, twenty to wasn't even

29:01

close. And in fact, it was so

29:03

ridiculous. Like, the first time I

29:05

showed him, Michael didn't believe it was true. And

29:07

so we got in the numbers and that kind of

29:09

thing. And there's probably, like,

29:11

ten super lucky things that happened in my

29:13

life, but him giving me that

29:15

framework. And then at the exact same

29:17

time, Dell as a company had

29:19

made two stumbles. So they had options

29:22

trade that went bad and their laptops had

29:24

caught on fire. And so the stock

29:26

was in in the ditch. It

29:28

was trading six times earnings.

29:31

And we had discovered, if

29:33

you will, through this

29:35

framework that they had a

29:37

massive competitive advantage

29:39

of this return on invested capital thing.

29:42

And so we went to a strong

29:44

buy on a broken stock and it went

29:46

up 651 x in the public markets

29:49

from there. And I became close

29:51

friends with Michael and and Tom Meredith,

29:53

who is CFO at the time. They

29:55

did all the work. I was just along for

29:57

the ride, but it was very

29:59

fortunate. But that helped

30:01

put Bill Girley on the map in the same way

30:03

that maybe eBay really

30:06

thrust benchmark into the

30:08

into the limelight? Very much.

30:10

So especially with the buy side

30:12

community. So Michael, I just want to double

30:14

check this. There's a chance I've read a book of

30:16

his. Did he write a book called Think Twice harnessing

30:18

the power of Counter Intritioners.

30:21

Oh, yeah. He's he's written four or

30:23

five books. Mhmm. And he's

30:25

and he's got great content. He did a

30:27

Google Talk that's on YouTube that I'd

30:29

highly commend people

30:30

watch. It's just fascinating.

30:32

He's been on a shynessy a couple of

30:34

times. So yeah. He he's he's worth

30:37

checking out. So in that particular

30:39

case of proselytizing, did

30:42

Michael do that? Was he

30:44

proselytizing and sharing this

30:46

ROIC? Idea

30:48

because he had peers who

30:50

were noncompetitive and it

30:52

was sort of deposit in the

30:55

Karmic bank account to hopefully

30:57

have some reciprocation or is he

30:59

just a nice guy? It just seems like

31:01

the environment that that would not be true in

31:03

all environments. Right? That someone would

31:05

take this new insight

31:07

and share it

31:07

widely. I think he's a natural

31:10

learner, and maybe that's

31:12

what you would attribute.

31:14

Howard Marks writing to, you know, you ask

31:16

people who do that, you know, ask Buffet, why

31:18

do you write your letters? Like, I think

31:21

people believe that it helps

31:23

their mental frameworks if they write

31:25

stuff down. And it

31:27

challenges them. Even in when

31:29

I write a blog post, from the minute

31:31

I have an idea or a

31:33

compulsion to say, hey, I'm gonna write about

31:35

this topic. The process through

31:38

which you actually put the

31:40

words to paper and structure the paragraphs, structure

31:42

the argument. You know, you get

31:44

smarter. Sometimes you decide, oops. I was around,

31:46

like, you learn by putting it

31:49

all together. And Michael's always

31:51

been that way. He's pretty much

31:53

built a career out of

31:55

being someone who studies

31:58

companies valuation frameworks,

32:00

how investors win I mean, he's

32:02

gone deep on things like

32:04

Myers Briggs on different investors and,

32:06

like, you know, structural, how do you organize an

32:09

investment team to be most all

32:11

kinds of stuff like that. It's what

32:13

fascinates him and he's a

32:15

synthesizer. So I

32:17

think some of the best non fiction writers

32:19

or sensitizers people

32:21

say, oh, Michael Porter competitive

32:23

strategy, fourteen other people had

32:25

written that stuff before. Well, he wrote it in a

32:27

really compact way that's easy to

32:29

read. That's super helpful. And

32:31

Michael does that. With,

32:33

you know, he he goes out

32:35

and reads the stuff that's super hard

32:37

to

32:37

read, that's overly academic or whatnot,

32:39

and packages in a way that people

32:42

can consume. Mhmm. Never ever Kevin

32:44

Kelly said to me once and I'm

32:46

paraphrasing, but I really

32:48

enjoy Kevin Kelly for people who don't

32:50

know. Look them up. Kk

32:52

dot org. And he was

32:54

saying, I don't write to

32:56

express what I think. I

32:58

write in an order to think or

33:00

discover what I think -- Yeah. -- or clarify what

33:02

I think. So the taking of the

33:04

ROIC from, in this case, let's just say,

33:06

food and applying it to Dell was this sort

33:08

of translational move. So I

33:10

wanna read something from I think this

33:12

is a vox interview,

33:15

and I can't believe everything you read on the Internet, but it

33:17

would make sense. So

33:19

this is related to OpenTable. So having

33:21

come out of OpenTable being successful, this is

33:23

quoting you. I was trying to think of other industries

33:25

where if you put a network on top

33:27

of it, it would absorb waste

33:29

and make it more efficient and more

33:31

usable. And this is within the context

33:33

of of looking for something

33:35

that would basically appear like Uber. Right?

33:37

So not working within the taxi

33:40

framework but with black cars. And I'm just

33:42

curious where else you've applied that

33:44

type of translation where you see

33:46

a case study

33:48

or a successful proof,

33:50

right, of a network being laid on top of something

33:52

and then applying it to something else. Are there

33:54

other examples of translating in that

33:56

way? Absolutely. And and by the

33:58

way, OpenTable was our impetus to

34:01

do OpenTable was based upon

34:03

something like that. So in nineteen

34:05

ninety six, Brian Archer,

34:07

who was at the Santa Fe Institute back

34:10

then, published an article in Harvard Business Review

34:12

called increasing returns in the two

34:14

worlds of business, and it was really the first

34:16

piece that talked about network effects. Of

34:19

course, Microsoft was already starting to really take

34:22

off, but this idea of

34:24

network effects is that some

34:26

industries are gonna restructure where you get

34:28

win or take most. And the more

34:30

successful you are, you get locked

34:32

in. And Brian talked about

34:34

things just like the Microsoft UI. So,

34:36

like, you know how word works. You get

34:38

comfortable with how it works. And then, like,

34:40

switching has cost and and those kind of

34:42

things and sharing documents and

34:44

collaboration and, you know, Zoom

34:46

obviously has massive network

34:48

effects. So we started looking

34:50

for those things because they

34:52

tend to cause outlier

34:54

outcomes. So I can remember, and this

34:56

gets into it back your rule stuff

34:58

when I met with Chuck Timmston, the founder of

35:00

and he had three restaurants. You

35:02

had to believe a lot to get

35:04

from that point to the global phenomenon that

35:07

it became. Yeah. And the bet

35:09

that we talked about making when we

35:11

said, okay, let's go do

35:13

this thing is if we get enough restaurants on

35:15

this thing, then the consumers will come. And if the

35:18

consumers come, then people will have to get on. You know? And

35:20

that's what

35:22

happened. But you had to believe

35:24

it. Because otherwise, at the time, we were selling PCs to restaurant

35:26

owners. And guess what? They didn't

35:29

have connectivity at the

35:32

time. So we had to partner with someone to like,

35:34

broadband installed, which wasn't easy.

35:36

I mean, it was all bad.

35:40

So your normal rule set, we're putting a piece of

35:42

hardware in a small to medium

35:44

business that didn't have a lot of money.

35:46

And we had to provision broadband.

35:50

So you would normally go, don't

35:52

do that. But I remember my favorite

35:54

proof point of that network effect, and then I'll

35:56

get to some of the other models. At

35:59

one point, we the sales force productivity and

36:01

our model when we scaled up

36:03

the business that we had -- our e

36:05

salesperson had to close four

36:07

restaurants a month. And at the

36:09

time, we were up to seven point seven. So we're feeling pretty good. Things

36:11

were starting to break our

36:14

way. And in

36:16

the board meeting, they laid out the list of all

36:18

the salespeople and one of them had

36:20

done thirty five -- Thirty

36:22

five restaurants in a month.

36:24

And so I asked the question, who's

36:27

that salesperson? And OpenTable started in San

36:29

Francisco and we played local

36:32

game. We didn't go everywhere at once. We

36:34

built liquidity city by

36:36

city. Anyway, that salesperson

36:38

was the one salesperson left in

36:40

San Francisco. Where we ninety

36:42

percent penetration. So the thirty

36:44

five were coming out of that last ten

36:46

and that individual was

36:49

basically taking orders. And that,

36:51

you know, to me, was like,

36:53

yeah, the network effect's really

36:55

working here. Yeah. And so, anyway,

36:57

we look for that. Some

36:59

other frameworks, SaaS. Could I actually pause

37:02

for one second before you get to SaaS? So you

37:04

talked about you have to believe you mentioned you have to

37:06

believe a lot of things can go

37:08

right in order to invest in

37:10

a business like OpenTable

37:12

at the time. I would love

37:14

to know what contributed to

37:16

your ability to have conviction.

37:18

I mean, how much of it was potential

37:20

network effects versus total addressable market

37:23

versus founder versus other

37:25

stuff? I would say

37:26

in some of these verticals, both

37:28

OpenTable and Uber, there were years

37:30

and years and years where everyone thought

37:32

the TAM was too small. In fact,

37:34

oh, here's a great story. This was in ninety nine. So this

37:36

is a twenty three year old story, but

37:39

I had successfully recruited a

37:42

CFO from a public company, which you could do back in those

37:44

glory days to come

37:46

into open table. And one

37:49

day, I showed up early for a board meeting and

37:52

and the CFO comes to me and he

37:53

says, Bill, I'm gonna quit.

37:56

And

37:57

I said, okay. I go, why are you gonna quit? And he goes, this

38:00

business will never work.

38:02

And I said, okay. Why will

38:04

never work? And he says, well, my

38:06

model says I'll never work. So I said show

38:08

me your model. So when you look at the

38:10

model, I dive in

38:12

and he has frozen

38:14

penetration in each city

38:16

at seventeen percent. And he

38:18

had come from a retail business, and I

38:20

I go, why'd you freeze it at seven

38:22

ten percent. He said, oh, no one gets more than seventeen percent

38:24

market share. All the businesses I've worked

38:26

with because I believe the network effects,

38:29

so I was like, we're gonna

38:31

get ninety nine. We're not stopping at seventeen. We're gonna get ninety nine

38:33

percent. You don't understand how this is

38:35

gonna work. And That's

38:38

because I believe the network effects and he didn't. And when

38:40

we filed the S1I really wanted

38:42

the

38:42

FedEx. Did you stay there?

38:44

No. He left. He quit.

38:46

And I decided not to to FedEx in the S1I

38:49

figured he'd seen them. 651 by then,

38:51

we were kicking off, you know,

38:53

massive cash flow.

38:56

Each quarter. Yeah. So you're talking about models and I

38:58

interrupted you. You were gonna jump Yeah. I was

39:00

just gonna mention a bunch of them. I mean Yeah.

39:03

Please. She has so

39:06

I did my first SaaS deal in ninety nine, and that

39:08

category of companies is still

39:10

bearing for people who've made tons

39:13

of money, you know, and now it's common, but

39:16

but it's time it wasn't. And you

39:18

transformed an entire industry

39:20

open source, you know,

39:22

benchmarks probably had, I don't

39:24

know, eight or nine successful open

39:26

source companies where you've used that

39:28

model to attack a

39:30

industry that's preexisting, but where

39:32

someone has proprietary technology, social networks. I mean,

39:34

we are fortunate to be in

39:36

Twitter, Snapchat, Instagram,

39:40

and Matt Koller, who is a partner at Benchmark, developed kind of

39:42

a sixth sense on what a

39:44

breakout social network looks like. And we've

39:46

had a few

39:48

that missed too. I mean, but the outsized nature of the winds are

39:50

so high. So there's, like, four

39:52

or five different ones. Right? 651 we

39:54

look for network effects? And

39:57

you get good at, you start to understand what

39:59

works, what doesn't, what leads to

40:02

success, what doesn't. I mean, there's

40:04

nuances. Right? In

40:06

open with that business model, which is you're

40:08

basically packaging support and

40:10

reliability, because someone could

40:12

just download it for free. So how do

40:14

you charge? And this goes

40:16

back to Michael Porter. But if you have a very

40:18

consolidated industry of big

40:20

companies, they don't pay. They

40:22

just hire the people that know how to use

40:24

the product. So We were in

40:26

MySQL as an example.

40:28

Google and Yahoo were two of the biggest customers and

40:30

never paid us a penny. You need

40:32

that product to go into the

40:34

corporate world where people

40:36

want that kind of handholding, and that's where

40:38

you get paid. So you can't just

40:40

be, oh, I love OpenSource because

40:42

then you get into

40:44

something where it's serving a much more finite set of customers

40:46

and then you can't make the model work. So

40:48

you have to learn the esoteric

40:50

nuances. But anyway, yeah, there

40:52

are many

40:54

veins in and venture they get mined over and over and over

40:56

again if they're

40:58

big enough. Just

41:01

a quick thanks to one of our sponsors and we'll be right back to

41:03

the show. This episode is brought to you by

41:06

athletic greens. I

41:08

get asked all the time. What I would take if I could only take one

41:10

supplement? The answer is

41:12

invariably AG1 by athletic

41:14

greens. If

41:16

you're traveling, If you're just busy, if you're not sure if your meals

41:18

or where they should be, it covers your

41:20

bases. With approximately seventy

41:22

five vitamins, minerals

41:24

and whole foods sourced ingredients, you'll be hard pressed to

41:26

find a more nutrient dense formula on

41:28

the market. It has a multivitamin, multimiral

41:32

greens complex probiotics and

41:34

prebiotics for gut health and

41:36

immunity formula, digestive enzymes, and

41:38

adaptogens, you get the idea.

41:40

Right now, athletic Greens is giving my audience a special offer

41:42

on top of their all in one formula,

41:44

which is a free vitamin

41:46

d supplement and

41:48

five free travel packs with your

41:50

first subscription purchase. Many of us

41:52

are deficient in vitamin d. I

41:54

found that true for myself, which is usually produced in our bodies

41:56

from sun exposure. So adding a vitamin

41:58

d supplement to your daily routine is a great

42:00

option for additional

42:02

immune support. Aphthnic Greens is

42:04

running a special offer just for this

42:06

month of January. Usually, when you try

42:08

athletic greens, they give you a five

42:10

day supply of travel packs plus a supply

42:12

of vitamin d three plus k two.

42:14

But for January, you get ten

42:16

travel bags plus that d

42:18

three plus k two. So If

42:20

you've been thinking about giving AG1 a try, there no time like

42:23

today. You can try AG1 by

42:25

athletic greens at athletic greens

42:28

dot com slash Tim. One more time, that's athletic

42:30

greens dot com slash Tim for a

42:32

free one year supply of vitamin

42:34

651, and

42:36

ten free travel packs? It's

42:39

maybe a a novice

42:42

question, but I'm curious. What

42:44

rules did you guys have around

42:46

bet sizing, right, to use those sort

42:48

of -- Yeah. -- maybe a poker

42:50

analogy. How do you

42:52

think about the

42:53

parameters around the size of the check that you cut

42:56

just in terms of

42:57

long term strategy and Benchmark's

43:00

a very particular firm we've

43:02

wandered off course a couple of times, but we've always come back to our

43:05

kind of home base. And we've remained

43:07

committed and focused to

43:10

very early stage investing. And we're pretty

43:12

much market takers on what

43:14

that first check size is.

43:18

What do you mean by that? There's

43:20

a competitive dynamic in the industry that kinda defines

43:22

what a series a check looks like

43:24

or, you know, sometimes series b.

43:28

we thought something was gonna be an outsized winner, we wouldn't miss

43:30

it because the market was

43:32

saying that had to be fifteen or

43:34

five, it's more like

43:36

a ticket. Fifteen or five you're talking about, you know, fifteen or five

43:38

million. Right? So would the

43:40

guiding principle in a case like that be the

43:42

percentage of ownership that you would

43:44

end up

43:46

with? Something we care way more about than the size of the town.

43:48

Yeah. Yeah. And and we're not the

43:50

only one. A lot of mentioned firms operate

43:52

that way. So board seats

43:55

is the constraint, and then ownership is

43:57

the optimization variable, whatever

44:00

the math is that follows out

44:02

of that is is what you take. To

44:04

be right and contrarian, I'm

44:07

curious how you cultivate

44:09

that in partner meetings. What is the format

44:11

of a successful partner meeting when someone's getting up and saying?

44:13

Well, one thing I I would

44:15

offer as a preamble

44:18

to at least our process

44:20

is the founding partners of Benchmark did

44:22

something that

44:24

was not had been done before in Venture, which is they created

44:26

an equal partnership. And most

44:28

of business partnerships, and

44:30

this includes private

44:32

equity and and law firms and real estate firms,

44:34

there's a hierarchy. And the people that

44:36

have been there the longest sit at the

44:39

very top and they take

44:41

an outsized amount. And some

44:44

of our founding partners had worked at those

44:46

firms and felt like the young people did more of

44:48

the work And therefore, you know, it wasn't conducive

44:50

to the right type of internal behavior.

44:52

And so they created this notion of

44:55

a equal partnership and and

44:58

even today when we go hire a new

45:00

partner, they come in and they have an

45:02

equal seat at the table. We just divide

45:04

the pie. Which is very different. The reason I I think

45:06

that's important to any discussion about

45:08

our own processes, I think

45:10

that has a

45:12

number of dynamics

45:14

that kind of emerge

45:16

from being equal. I think everyone's

45:18

voice is heard, whereas if you had a

45:20

hierarchical firm and boss

45:22

walked in and said, blah blah

45:24

blah. Like, that would carry so much more

45:26

weight. Right? And so

45:28

you feel like your voice is heard

45:30

when you come in. I think we for

45:32

one another. If I'm at a higher

45:34

Google firm that's up or out, I'm kinda

45:36

competing with my

45:38

peers here. Because we're gonna divide it equally. I want our new to

45:40

be as successful as I possibly can. And

45:42

I felt that on the way in,

45:45

and I feel the desire

45:47

to do it on the way out. Those are just natural

45:50

emergent properties. So then to

45:52

come around to your question, like, how do you be

45:54

contrarian? How do you have? Well, I

45:56

mentioned, like, we have this

45:58

phrase, what could go right. Like, you're just

46:00

constantly thinking devil's advocate

46:02

in your mind. And devil's advocate

46:04

on the positive side, like, what would

46:06

it take for this thing to break out? Like, are we making a mistake? Could

46:08

this could you imagine this being

46:10

really, really big? And what would have

46:12

to happen for that to happen?

46:15

someone

46:15

new comes along, you'd be surprised I

46:18

have specific memories of

46:20

new partners coming in. We'll just tell

46:22

them

46:24

bring all the companies you meet in. Like, just bring them

46:26

in. And those companies

46:28

will present and then we talk about them

46:30

afterwards. And that's where your

46:32

partner meeting. Yeah. And that's where

46:34

the learning process gets

46:37

passed down. I'm sure

46:39

just like storytelling in

46:41

a tribe. Three hundred years ago or four hundred years ago.

46:43

Right? Like, that that is

46:45

that dynamic. And a lot of people

46:47

say venture is at the end

46:49

of the day. A pattern

46:52

recognition job. But we also

46:54

talked about those rules can get in

46:56

trouble. So you're trying to create this

46:58

kind of loose

47:00

pattern recognition so that you can be helpful and identify

47:02

things. And then you wanna pass those

47:04

along. And then you gotta

47:06

constantly check

47:08

it because of what we said about how you can miss something. But

47:10

when that new partner comes in and those are

47:12

happening, I can remember, and I

47:15

leave the names out. But one of the new parks, like,

47:17

he brought in company one. No. Company two.

47:19

No. He's very frustrated. And then

47:21

company ten, he brought in. We were like,

47:23

try and close that immediately. Like, it's just the

47:26

complete opposite. And he

47:28

remembers that too. You're starting

47:30

to learn. You're starting to

47:32

pick up the collective

47:34

wisdom of the group. It is a bit of

47:36

an art form because in

47:38

that window between when you meet a

47:40

company and when you might try and

47:42

close the investment, I bet you

47:44

the average is

47:46

under sixty days. Oh, yeah.

47:48

Yeah. It might even be shorter. It might

47:50

even be closer to three weeks. And so it

47:53

all happened super fast. Was there

47:55

any feedback that you recall

47:57

or could just hypothetically imagine

48:00

you gave that new partner that

48:02

helped him or her to hone

48:05

in on the tenth that

48:07

was then a yes closes

48:08

immediately? A lot of

48:09

it goes back to what you might pick

48:11

up in competitive strategy. The the

48:13

two books that I

48:16

would put for the venture community for the startup community that I

48:18

put as next level

48:20

right on top of competitive strategy

48:24

or Innovators dilemma, which does an

48:26

amazing job of describing why

48:28

startups can compete with big

48:30

companies. Amazing. And crossing the

48:32

chasm, which does a

48:34

really good job of

48:36

explaining how

48:38

a startup should kinda

48:40

sequence their customer base as

48:42

they grow. And they're

48:44

both fundamental, and you just

48:46

pick up a number of these

48:48

different things. And I'll give you an example because you start developing

48:50

tuition. So a company where we had

48:52

a quick yes on that I'm on the board of is

48:54

called Hacker

48:56

So at the time of hacker

48:58

one's founding, there are four companies that use white hat

49:00

hackers to make their sites more secure.

49:04

Microsoft, Google, Facebook, and Mozilla. They were the only

49:06

four. No one else did it. And

49:10

yet, people would say, I remember

49:12

seeing an interviewer, Sheryl Sandberg, where

49:14

she said someone asked her about privacy and

49:16

security, and she says, well, the best

49:18

thing we have is this bug

49:21

bounty program. And she talks about it for a

49:23

while in the crowd applauds. One of the

49:26

individuals that worked on the

49:28

Facebook program came up with a

49:30

business idea of why don't we build this for the other companies in the

49:32

universe? Like, every other company

49:34

in the

49:36

universe. And when they

49:38

presented, we didn't even discuss the

49:40

company. The minute they left, we were like, okay,

49:42

how are we gonna close this? Everyone

49:44

had just jumped

49:46

to yes. Because it seems so thoughtological

49:48

that that that there's no

49:50

way this thing's great for these four

49:52

companies and no

49:54

one else. And it turned out that those four companies had a group

49:56

of, like, twenty or thirty people running

49:58

the program, which other companies

50:00

couldn't do. So it's

50:02

just natural. That if you ran that

50:04

as a service, people would would

50:06

jump in. That's one that we got too

50:08

fast. So you've mentioned in a

50:10

number of your partners, Matt Coller, who

50:12

was was also very helpful for

50:14

Uber. He seems to be very good

50:16

at using himself as the guinea pig, very

50:18

smart guy. What are some of

50:20

the lessons you've learned

50:22

from your partners? And that could

50:24

be anyone, you know, but Fenton, Laskie,

50:27

Koller, would don't even necessarily have to mention names. I'm

50:29

just curious what their superpowers

50:31

are or what you

50:33

have

50:33

adapted, learned

50:36

from

50:36

them. I'd probably go back to some of the founding partners because

50:38

those were the ones that I that I learned

50:40

from. You know, one of them

50:43

I'd say from Bruce Dunlevey is just have a

50:45

really big tent. We're in a networking

50:48

business where you're trying to look under every

50:50

rock for the next possible

50:52

deal. And cutting off

50:54

avenues of information or

50:56

flow is just a really stupid

50:58

idea. And so it it ties in with

51:00

what could go right but just be super

51:02

open minded. And so Bruce

51:04

had been involved with a company that

51:06

I think was in a software tool

51:09

space the name escaping where we could look it up.

51:11

And had developed a lot of

51:14

friendships with down the

51:16

organizational chart at the company.

51:18

And one day, of the engineers

51:20

from that company comes to him and

51:22

says, I'm working on this

51:24

marketplace thing called eBay. And he

51:26

was from a different industry Street because he

51:28

had developed that relationship

51:30

that came in. Now it turned out that

51:32

Bob Cagle was much more

51:34

excited about eBay than Bruce was. And

51:38

so Bruce made that intro because we're

51:40

equal partnership. That's all

51:42

cool. And Bob ends up doing the deal and,

51:44

you know, becomes the number

51:46

one ranked DC in the land, but it all came from that

51:48

relationship. So the older

51:50

I get, like, I just take the

51:52

time, be

51:54

kind, be available. It pays off. At at least in our

51:56

industry, it pays off big time. So that's

51:58

one. Another one is just and this

52:00

is more specific to the but

52:03

unless you're super lucky. There are two

52:06

VCs I know that were very

52:08

fortunate. Coler and

52:11

roll off. It's Koi who wrote off with

52:13

YouTube and Matt with Instagram where they

52:15

had a hit two years in. But

52:17

for the vast majority of venture capitalists,

52:19

they don't have equity event

52:21

till year eight or nine.

52:24

And so it's easy to

52:26

doubt yourself. And three

52:28

years in, people like to use the child age analogy.

52:30

You've got twelve year olds

52:32

becoming thirteen year olds and

52:34

your whole portfolios

52:36

like that with acne all over their

52:38

face, and you can really

52:40

lose confidence. And so one

52:42

thing that a lot of my partners did

52:44

was just it's gonna be

52:46

okay. Get back out there. You're

52:48

doing fine, you know, that kind

52:50

of thing, which was way more

52:52

helpful than you could

52:54

possibly Because the anxiety -- Yeah. -- was, like,

52:56

spiking. I wanna come back to open source

52:58

in a minute, but first, I'm wondering you've

53:00

mentioned a number

53:02

of books that are broadly applicable to

53:04

entrepreneurship, and I'm sure

53:06

translate to, as you mentioned, investing in a number

53:08

of capacities. If

53:10

you were advising someone

53:12

or just mentoring someone

53:14

who wanted to learn how

53:16

to be a good angel

53:19

investor or venture capitalist intech.

53:22

What might you suggest to them in

53:24

terms of approach or resources books, anything?

53:26

I'll give you a few more books, and

53:29

then obviously see the world's different today. The content

53:31

consumption is so different, but there's

53:33

a book I love

53:35

called StartUp by Kathlin. Yeah. I bought that when I

53:37

first moved to Silicon Valley

53:40

in ninety nine. That was one of the

53:42

very first books. Yeah. It's

53:44

just fantastic. Thing. And he

53:46

-- Yeah. -- he was in

53:48

the the first portable

53:50

computers. Like, every one of them failed Yeah.

53:52

-- every every one of them. But they all raised

53:55

massive amounts of money.

53:58

He, on his ride home every day,

54:00

had a cassette tape and

54:02

left, like, an an archive

54:04

or a diary, which makes the

54:06

book so good because the details

54:08

fantastic. Mhmm. But they

54:10

had the best venture capitalists, they had the

54:12

best advisors they had

54:14

the best executives. And

54:16

the executives that were in this

54:18

was GoCorp, I think, that

54:20

were All went on to do amazing things, but this

54:22

company failed hard.

54:24

And I I think It's

54:28

nice to combine a book of failure

54:30

with all the books of

54:32

success. MustRead. I just think it's

54:34

super educational. Interestingly,

54:36

if you read that, you should probably also

54:38

simultaneously watch the general magic

54:40

documentary. It's so good. Because they

54:42

were competitor with Go. So it'd be

54:45

and and the same story. And and

54:47

the people there all went on

54:49

to wild success. Yeah. It's

54:51

like the Yodaworski's dune, where Geiger who went on

54:54

to create

54:56

the iconic designed

54:58

for alien and aliens. I mean, they're just these these

55:01

failures that contain an

55:03

all star

55:05

team and fail for any number of factors, but general magic. I just

55:08

wanna second that recommendation. Yes. Fantastic.

55:10

If people want an additional

55:12

insight look, listen to my interview

55:14

with Tony Fadel, which goes into great

55:16

depth on

55:16

that. Yeah. No, sir. On on

55:18

Tony, I I read his new book, Bill. It's

55:21

solid. He's very strong. And got

55:23

a lot of it's got a lot of frameworks,

55:25

you know. And you can agree with them

55:27

or not agree with them, but I do

55:29

think entrepreneurs benefit from

55:32

frameworks because just running by the seat of your pants. Like, you need

55:34

a a framework for having weekly

55:36

meetings. You need a

55:38

framework for executive recruiting.

55:40

You need a framework for

55:42

development of your team. And there

55:44

are a lot of good ones. There's there's probably

55:46

twenty good ones of each of those.

55:48

But the worst thing you can possibly do is have no framework

55:50

and just kinda see to your pants. And by the way,

55:53

that's the most frequent. Solution.

55:56

They absolutely have no

55:58

framework. I would mention Shoe Dog.

56:00

I don't know if the

56:02

reality had as many near

56:05

death experiences as the book makes it sound

56:07

like. But it's it's good to

56:10

see the tenacity you

56:12

need. This is the

56:14

Phil Knight. NIKE book. Yeah. But, like, it the

56:16

tenacity you need to make

56:18

it is

56:18

high. Yeah. And so if

56:21

you're starting a company because

56:24

you think it's gonna be a good lifestyle. shit. You're in

56:27

for a rude

56:27

awakening. Yeah. So

56:29

the same writer

56:32

who penned Shoe Dog in

56:34

reality is the same person who did

56:36

open the autobiography of

56:38

Agassi, which is another spectacular

56:40

And apparently, the new Harry book. Yeah.

56:42

No. Don't kidding. Yeah. Alright. That's a

56:45

selling point because his Maringer is

56:47

J. D. Maringer is

56:50

spectacular. Okay. So Shoe Dog. So so far, these are company

56:52

building -- Yeah. -- books, which is totally

56:54

For for kinda high-tech macro,

56:58

it's there is such a thing. I think Ridley's two books, the rational

57:00

optimists, and how innovation works

57:02

are just fantastic and spectacular.

57:04

They're they're much higher level.

57:07

But they're really really good. What makes

57:09

them good? The first book

57:11

Ridley has this point of view that I think

57:13

is very hard to dispute the the vast majority

57:15

of wealth creation and increase the

57:18

standard of living for humans on

57:20

the planet. Come

57:22

from two things, commerce and what he calls ideas

57:25

having sex. So if you

57:27

if you think about if

57:29

someone comes up with a new farming technique, the

57:32

marginal cost of that is

57:34

zero. And if you pass it along

57:36

to someone, their productivity improves. And so, you

57:38

know, it it's super

57:40

powerful in my mind, especially when I think

57:42

about people

57:44

a lot of people on the planet wanting to

57:46

improve standard living with a lot of people. And I look at

57:48

what Jing Xiaoping did in

57:51

China and say, I don't know that any

57:53

other human in the history of the world

57:56

has unlocked as much standard of

57:58

living increase

57:59

is this one human by bringing

58:02

capitalism to China. And,

58:04

anyway, it's something that -- Yeah.

58:06

-- really speaks to me. Yeah. Much

58:09

smaller scale, but people should study Lee

58:11

Kuan Yu -- Okay. -- in this whole

58:13

Singapore story. Well, just the whole Singapore story. I mean,

58:15

basically taking it from a swamp to

58:17

what it is today is just

58:19

an

58:19

incredible, so but but I agree on the day

58:22

shopping piece. Yeah. Yeah. So then on

58:24

on other con intent. Like, the world's way different now. Right? With

58:26

podcasts and blogs

58:28

and Twitter and all these kind of things.

58:32

So no matter what you're doing, but especially if you wanna do Venture

58:34

attack, like, there's lots of people

58:36

that talk their book, that talk their

58:38

game, like, follow,

58:40

listen, read, consume,

58:41

consume, read, read, read. I don't think

58:44

you can get too

58:46

much information. Let's segue to open source. What else do you think open

58:48

source might be able to

58:50

solve? And I've had some

58:52

reasonably direct

58:54

experience with watching,

58:56

I suppose, one example of what you

58:57

described, which is automatic

59:00

and very good friend, Matt Longway. I'm an

59:02

adviser to automatic. And for those who don't

59:04

know, Matt who's been on the show multiple

59:06

times was a lead developer of

59:08

WordPress, very familiar with

59:10

OpenSource, and then has built

59:12

a services and enterprise grade secure,

59:16

stable solution for many

59:18

different businesses and people in the form

59:20

of automatic. So

59:22

I'm deeply interested in this, and I know you are as well, you're a big fan of Open

59:24

Source. What else do you think Open Source

59:26

could be applied

59:27

to? Or might be

59:30

applied to? First of all, I think people are

59:32

relatively familiar with the

59:34

big individual software projects,

59:36

Linux being the most well

59:39

known. It's over twenty years old now. It's

59:42

clearly the most used

59:44

operating system in the world. And

59:47

what people might not know about it

59:49

is a lot of scientists

59:51

believe it's the most secure. And that's

59:53

kinda counterintuitive. Oh, all the codes public, how

59:55

could it be the most Pure. But

59:58

it gets beat up the most, you know,

1:00:00

because it's used the most.

1:00:02

There's a great piece of writing,

1:00:04

an incredible piece

1:00:06

of writing called the cathedral and the bazaar that

1:00:08

was the first kind of

1:00:10

magnum opus on

1:00:12

why this open source

1:00:14

thing might work. And

1:00:16

another thing people don't get

1:00:18

is the cathedral and the

1:00:20

Lazar compared. Are you trying to

1:00:22

build up like the

1:00:24

Gowdy Church or a bazaar which

1:00:26

is super flat and

1:00:28

wide and obviously

1:00:30

built by people. And the thing is open source is way

1:00:32

better complex problems than

1:00:34

simple problems. And it's

1:00:37

a very complex problems, it'd be hard for a single

1:00:39

company to do. And so if you're building an operating

1:00:42

system, it might have all

1:00:44

kind of

1:00:46

edges on how it integrates with other systems, you know, different

1:00:48

drivers you might need. And

1:00:50

the world is able to

1:00:54

build Whereas an individual company would be very

1:00:56

difficult. And so we

1:00:58

started to see that in MySQL

1:01:00

and there have been others like

1:01:02

MongoDB and we're in one called

1:01:04

Elastic and, like, there's been a lot of

1:01:06

successful companies around single

1:01:08

software frameworks. Something

1:01:11

started happening about ten to fifteen years ago, which is

1:01:13

people started using open source in more

1:01:16

complex and then sometimes using it

1:01:18

defensively rather than

1:01:20

offensively. So

1:01:22

the most well known is probably Android. Apple would

1:01:24

come out with this smartphone. You could

1:01:26

only get it on AT and T.

1:01:29

It's scared to shit out of everybody, not

1:01:31

just Google, but it's scared to shit out

1:01:33

of all the other telcos. It scared

1:01:36

to shit out of all the other

1:01:38

handset manufacturers. And so

1:01:40

Google did this clever thing. They said,

1:01:42

oh, we're gonna create a competitor, but it's

1:01:44

gonna be open you can trust

1:01:46

us. Now, there's

1:01:48

different versions of how open something is,

1:01:50

and and Android is not very

1:01:52

OpenSource. But at the time,

1:01:54

Compared to the threat of Apple, it seemed like a

1:01:56

much better

1:01:57

trade. And so they got the

1:01:59

world behind them on this thing -- Yes. --

1:02:01

and it took off.

1:02:03

then that model's been

1:02:06

repeated in some esoteric ways.

1:02:08

Facebook has something called the open compute

1:02:10

project. And if you want

1:02:12

your hardware, to go into

1:02:14

the Facebook data server

1:02:16

room, it has to be compatible with

1:02:18

their open source framework. And what

1:02:20

that means is no

1:02:22

one can have an

1:02:24

IP claim against Facebook

1:02:26

because you basically sworn off your

1:02:28

IP. So

1:02:30

they're commoditizing the stuff they're

1:02:32

gonna purchase. You know, since

1:02:34

then, AT and T and

1:02:36

China Mobile have worked with the Linux Foundation

1:02:38

to do the same thing for the wireless

1:02:40

and wireline equipment stack. So they're defining open

1:02:42

standards for their next gen

1:02:44

products. And if you're

1:02:46

a supplier, that

1:02:48

wants to sell to them, you have to agree to

1:02:50

the those standards. And once

1:02:51

again, you're you're out of the IP

1:02:54

game. So they don't get held up. It's

1:02:56

very

1:02:58

clever stuff. Would

1:02:58

you mind expect just expanding on that a little bit? Well, let's

1:03:00

go to the Facebook one because it's more understandable.

1:03:02

It's called the Open Compute Foundation.

1:03:05

It's managed by the the

1:03:07

Linux Foundation acts as a

1:03:10

steward much in the way that the crypto

1:03:12

world thinks a Dow

1:03:14

manages this loose federation. The Linux Foundation does that.

1:03:16

Linux has been a Dow for twenty five

1:03:18

years. It's it's super interesting.

1:03:20

And so

1:03:22

that group, they're like a nonprofit overseer of

1:03:24

the project. Brake ties -- Yep. --

1:03:27

the new patent defense, which I think is

1:03:29

super interesting. So they pull patents

1:03:32

and no one sues within the open source project, but if

1:03:34

someone were to come attack it, you

1:03:36

know, you'd go out after them. So

1:03:39

anyway, Facebook has all these equipment in their

1:03:41

data center, networking equipment storage,

1:03:44

equipment, computers, software. So

1:03:47

they just created this thing called the Open Compute

1:03:49

Foundation that defines Open Standards

1:03:51

for how all these products

1:03:53

work. And so if you wanna be on

1:03:55

their purchasing list, You

1:03:58

say, yes, we are compatible with this

1:04:00

open standard. And once again,

1:04:02

it basically commoditizes

1:04:04

that equipment. So it just

1:04:06

gives Facebook more leverage -- Yeah. --

1:04:08

with a broader spectrum of suppliers.

1:04:10

Every one of these things has a website and

1:04:12

you can go see it. So now you know,

1:04:14

Amazon and and Google, they're all

1:04:16

part of open compute. I got it.

1:04:18

So you can't get if I'm understanding you

1:04:20

correctly, tell me if I'm missing something. They're

1:04:22

trying to create

1:04:24

conditions such that they wouldn't end up

1:04:26

back in the day when I moved to Silicon Valley.

1:04:28

I worked in storage area networking.

1:04:31

And EMC at the time was

1:04:33

pretty famous for kinda being a black box.

1:04:35

Like, if anything broke, if anything went wrong,

1:04:37

like, you EMC to fix it or upgrade

1:04:39

it, so to avoid being held hostage

1:04:42

or losing leverage with

1:04:44

suppliers. Facebook made this move. Is

1:04:46

that a fair description? Absolutely. And

1:04:48

then another High profile

1:04:50

one. Google was very worried about

1:04:52

Amazon. That's genius.

1:04:54

I mean, it's very smart. Oh, it's super

1:04:56

clever. Google was very afraid of Amazon running

1:04:58

away with the cloud services

1:05:00

business in AWS. And so

1:05:02

they had a piece of technology called

1:05:04

Kubernetes. And this was right around

1:05:07

when docker and containerization took

1:05:08

off. And Kubernetes was an

1:05:11

orchestration layer for containers. What a

1:05:13

bizarre name? What was it? Kubernetes sounds

1:05:15

like an Italian

1:05:16

pasta. Well,

1:05:17

here's the thing though. They decided that it

1:05:19

was in their best interest to

1:05:21

take this technology and gift it

1:05:23

to an open source consortium.

1:05:26

They had the Linux Foundation to manage it,

1:05:28

and they went out and recruited

1:05:31

IBM and HP and all

1:05:33

these other vendors to say, oh,

1:05:35

yeah. We'll support Kubernetes because

1:05:38

everyone wanted to make

1:05:40

sure that people weren't locked

1:05:42

in to Amazon. And eventually,

1:05:44

it got so successful that

1:05:46

Amazon had to announce support for

1:05:49

Kubernetes. And now if you

1:05:51

need to move a workload, from Google

1:05:53

to Amazon, there's a common framework for which to do

1:05:55

that. So you're less locked in. So

1:05:57

this seems like

1:06:00

it's become at least based

1:06:02

on the examples you gave, like, a a

1:06:04

very refined,

1:06:06

reliable counterpunch that Google

1:06:09

uses. I call defensive corporate strategy. Yeah.

1:06:11

Yeah. I'll give you two more very recently.

1:06:13

Well, there's always been open street maps,

1:06:15

and that's kinda interesting because it's data

1:06:17

rather than, you

1:06:20

know, but the Open Street Maps team was

1:06:22

very kind of noncommercial. So

1:06:24

they didn't like when Apple and the

1:06:27

and Google's obviously running away

1:06:30

with the maps business. So if you just look this up like within the

1:06:32

past four weeks, there's a new

1:06:34

open source map group that's gonna live

1:06:37

on top of Open Street maps

1:06:40

that as Facebook and Microsoft and

1:06:42

the other parties that don't want Google

1:06:44

to run away with it. And quite

1:06:47

frankly, I think that's pretty cool. If you talk to

1:06:49

actually, someone you should have on it. I think that

1:06:51

would be really cool as Jim's Evelyn.

1:06:54

Who's run the Linux Foundation for this

1:06:56

whole time? Because

1:06:58

he's a huge believer, and

1:07:00

it kinda ties into Matt Ridley's ideas

1:07:03

having sex that not having

1:07:05

IP is actually great for the world because

1:07:07

it just creates constraints

1:07:10

for people being able to take advantage

1:07:12

of things. And so

1:07:14

mapping super interesting to me because it's data

1:07:16

oriented. There's something called Ferriss, which

1:07:18

is a open source processor, believe it or

1:07:20

not, that has a lot of momentum now. You'll

1:07:24

see people in on

1:07:26

the earnings calls for ARM, they'll start. Is

1:07:28

this a competitor? Is this gonna be a problem?

1:07:31

Because it's completely free license.

1:07:33

And China is a big backer

1:07:35

of open source could understand

1:07:37

why because the West has

1:07:39

accused them of of IP theft. So this is where we -- Yeah. --

1:07:41

you can't be accused. Safe

1:07:44

space. Safe space. They're investing

1:07:46

like crazy. But I wonder

1:07:48

what's possible. I'll give you some kind of

1:07:50

grander ideas. First of all, I think

1:07:52

autonomous vehicles should definitely be

1:07:54

open source. I think everyone

1:07:56

benefits. Safety is higher.

1:07:58

Communication layers are better.

1:08:00

The idea that you would use artificial

1:08:02

intelligence to figure out whether a light is

1:08:04

red, yellow, green is really fucking stupid because it's a state machine.

1:08:06

It is in one of those three

1:08:08

states and that could be communicated into

1:08:11

the software. Like, You don't

1:08:13

need to infer that. That's a known thing, but you need

1:08:15

a common language. And your test

1:08:17

suites could all be used by

1:08:20

everyone academia could

1:08:22

be working on the same thing that the corporations

1:08:24

are, safety is higher, easier for

1:08:26

government to get involved if it's

1:08:28

single standard. And I wonder about

1:08:31

other things like the NIH gives out

1:08:33

forty billion dollars a year. And many

1:08:35

of these projects end up as research that

1:08:37

leads to a drug that

1:08:39

leads to having a

1:08:42

seventeen year patent life and get -- Yep. -- sold at three hundred grand a year or whatever like --

1:08:44

Yep. -- why wouldn't we

1:08:46

say if you take NIH dollars

1:08:51

your research is open source. Why do we use

1:08:53

government dollars to fund stuff that

1:08:55

becomes proprietary?

1:08:57

That would make any sense to

1:08:59

me. Yeah. This is something

1:08:59

that you and I can

1:09:02

have a conversation about separately.

1:09:04

Yeah. This is I look at

1:09:06

the state of nuclear energy where the

1:09:08

cost are high

1:09:10

because of regulation, not because of the actual product. And I wonder

1:09:13

what 651 the

1:09:16

globe had a standard for,

1:09:18

you know, vision based nuclear reactor, like, wouldn't we get to lower price points? Wouldn't we get

1:09:23

to safer product? X. Like, wouldn't they get

1:09:25

more evolved? It seems possible to me. Are there any companies

1:09:27

that are on a short list

1:09:29

of companies you would love to

1:09:32

see exist or types

1:09:34

of companies, like anything that's just kind of a be in your bonnet or has been for any period of time?

1:09:36

Yeah. I can give

1:09:38

you a few. So one

1:09:41

651 this one's kinda

1:09:43

well known in the venture industry, but

1:09:45

everybody talks about something called

1:09:47

the interest graph. And they

1:09:49

wonder why there isn't a Internet

1:09:51

website that kind of links everyone that is tied to a specific

1:09:54

interest. And there are companies

1:09:56

that people

1:09:59

talk about as being close like Pinterest

1:10:01

or Quora or Twitter.

1:10:04

But I don't think any

1:10:06

of them have really pulled it

1:10:08

off. Would read it far. Sure. I think

1:10:10

Especially with some of the subreddits, they're getting close. Yep. You know? Yeah. And people

1:10:12

talk about kind of like

1:10:14

a holy grail, but, like, there's

1:10:17

not anyone that's, I think, just really nailed it. Really, really nailed

1:10:20

it. And if you did,

1:10:22

you would have this combination of

1:10:27

really cool unlock for people because if

1:10:29

you're into quilting, you'd be immediately

1:10:31

connected with everyone else, you know,

1:10:33

that's on your level and you

1:10:35

can imagine that kind of thing. But

1:10:37

then the advertising performance would just be off the charts because you've kind

1:10:39

of bucketed everyone

1:10:43

into these places. I actually think Twitter still

1:10:45

has a huge opportunity on this front, but it'd have to build a top

1:10:47

down version of Twitter rather than this feed

1:10:49

thing, which is super hard for a lot

1:10:52

of people

1:10:54

What do

1:10:54

you mean by top down? I think

1:10:57

you could take all of the

1:10:59

information that's flowing in Twitter and

1:11:01

all of the influencers that are

1:11:03

in there and build an algorithm that would score who's smart

1:11:05

about certain things. Let let's

1:11:07

make it super simple to

1:11:09

convey the point if you

1:11:11

look at stocks. So they already

1:11:14

have a UID, a unique individual identifier for each stock with the dollar symbol

1:11:16

thing. But

1:11:20

right now, The only way you can kinda

1:11:22

you could do a search, but there's a ton of noise. What if you had a page

1:11:25

where for

1:11:28

each stock, you had a list of the

1:11:30

top stories of the day, and that -- Mhmm. -- could be pulled out

1:11:32

of the Twitter feed

1:11:35

by knowing which people are

1:11:37

the access on that individual stock, which you could infer simply with

1:11:39

the data that's already in there. You could imagine

1:11:41

that for sports teams. So I

1:11:44

could have Twitter

1:11:46

sports that is a top down

1:11:48

version that's using the information in

1:11:51

the feed, but then presents it

1:11:53

more like a standard newspaper would.

1:11:55

If you understand what I'm getting That one's one. I'm highly

1:11:58

interested in people. Feel free

1:12:00

to

1:12:02

reach out in

1:12:03

any variation or new take on LinkedIn. I just think

1:12:05

it kinda stopped, you know, and

1:12:07

it stopped ten years

1:12:10

ago. One idea that There

1:12:14

aren't you semi retired, Bill? Are

1:12:16

you kinda semi retired? People will price

1:12:18

you. Your version of retirement is super intense.

1:12:22

Yeah. I people may steal this, and and that's

1:12:24

okay, but I hope they reach out to me.

1:12:26

It feels to me like you

1:12:29

could have paid drink

1:12:30

for people. You know? Okay. And so I don't think the

1:12:33

skill thing on LinkedIn works because

1:12:35

it's public. So you get

1:12:37

all this performance art where

1:12:39

people are just you know, being nice.

1:12:41

Right. But what if you had people maybe privately

1:12:44

opining on who they

1:12:46

think is smartest on particular

1:12:48

topics? You you

1:12:50

could develop a really cool product that would be unique to each individual

1:12:52

because what you see, Tim, would

1:12:54

be different than what I see.

1:12:58

Because it starts by who you trust. And then

1:13:00

what you see is anyway, I

1:13:02

think that's a really interesting concept. Yeah.

1:13:06

I wonder, you know, if

1:13:08

Twitter could actually use the literal page

1:13:10

rank by correlating verified accounts, although that's

1:13:14

become kind of unusable for me because

1:13:16

it was kind of pay for play for

1:13:18

a while, so it's difficult to filter now.

1:13:20

But if people have websites

1:13:22

that are their personal websites, just looking at the page rank, it seems like

1:13:24

from a computing side,

1:13:27

pretty easy to do. 651

1:13:29

thing Twitter could do that would

1:13:31

just be super industry is create a leader board for topic under

1:13:36

the sun. And then the

1:13:38

things you could build once you had that would be super compelling. Yeah.

1:13:41

Let's talk about

1:13:44

Twitter specifically a tweet

1:13:46

thread and feel free to amend this, but I'd love to have you

1:13:48

walk me through this particular

1:13:50

thread, this from spring of twenty

1:13:55

twenty two. It's not that long ago. And

1:13:58

here it goes.

1:14:00

So there are four

1:14:03

points under an intro, and I'll just read it if you don't mind.

1:14:05

So an entire generation of entrepreneurs and

1:14:07

tech investors built their entire

1:14:10

perspectives on valuation during second half of a thirteen year amazing bull

1:14:12

market run. The quote unquote unlearning

1:14:14

process could be painful, surprising, and

1:14:17

unsettling to many,

1:14:19

I anticipate denial. Number one. And maybe we could just

1:14:21

do this point by point. Well, we don't have to do two at a time. Number one,

1:14:24

previous all time highs

1:14:26

are completely irrelevant. It's not

1:14:28

quote unquote cheap because it is down seventy

1:14:30

percent. Forget those prices happened. Yeah. I got I screwed myself on

1:14:34

that first one. Second one, and valuation multiples are always a hack

1:14:37

proxy. Dangerous to use, if you insist

1:14:39

ten x should be considered amazing

1:14:41

in an upper limit over

1:14:43

that silly. So let's sit

1:14:46

on those two for a second.

1:14:48

Would you mind just expanding on either of both of those?

1:14:50

There's an unfortunate reality in the venture world that

1:14:55

really became very crystal clear

1:14:57

to me through a conversation with

1:14:59

Howard Marks, actually, but

1:15:02

it's structurally set up

1:15:04

you know, people talk about boom

1:15:06

busting cycles, but this is set up more like a sawtooth. So risk on

1:15:09

happens very slowly, almost like

1:15:11

the roller coaster and

1:15:14

neck, neck, neck, neck going out. Yeah. But

1:15:17

when it crashes, if

1:15:19

it's interesting to explain or

1:15:21

do my best job of explaining

1:15:23

why it's structure this way. When it crashes, it happens all

1:15:25

at once. So it's more like

1:15:27

a sawtooth, like, then

1:15:29

a sine wave, and

1:15:32

it just crashes, and it's

1:15:34

painful. And that just happened. And it happens it looks

1:15:36

like it happens every seven

1:15:38

to fifteen years. You know? And

1:15:43

I thought this was gonna happen six years

1:15:45

ago. I even wrote some things. I was

1:15:47

way early. It took six

1:15:49

more years, and it

1:15:51

got ridiculously crazy. But it took so long

1:15:53

from o nine, which was the last kinda and o nine wasn't as

1:15:56

hard to

1:15:58

reset as o one. But from o nine to thousand thirteen,

1:16:00

because so many entrepreneurs are

1:16:03

young, you know, you had

1:16:05

people grow up that had

1:16:07

never seen a reset. And

1:16:09

-- Yeah. -- risk on is a lot like the boiled frog. Like,

1:16:11

you don't know what's happening. Would you mind

1:16:13

just quickly defining risk

1:16:15

on for people listening? What

1:16:19

do you mean by that? So the community as

1:16:21

a whole takes on more

1:16:23

risk gradually without

1:16:25

realizing they're doing it. And their middle models

1:16:27

and their frameworks adjust daily

1:16:29

to what's happening. And

1:16:32

so they're thought

1:16:34

about how the world works is

1:16:36

really a window of five years

1:16:38

or maybe three years. Not -- Yep.

1:16:40

-- thirty years. For many of

1:16:42

them, they don't have the thirty

1:16:45

year perspective. And when the going gets good, greed takes over. And

1:16:47

you weigh the data points that feel

1:16:51

good to you and are gonna make you the most

1:16:53

confirmation biases. Oh, like crazy.

1:16:56

And so then you've

1:16:58

shortened your window to the

1:17:00

last twelve months. You know, this is

1:17:02

how the world works. And because things got so sloppy

1:17:04

with interest rates

1:17:07

being near zero speculation so I money

1:17:10

everywhere. We taught a lot of people not only

1:17:12

valuation things that will

1:17:15

never be true again. But,

1:17:17

like, growth at all costs. Like, spend as much money as you can. You can raise money every

1:17:19

nine months if you want to because you

1:17:22

could. You know, the failure rate of

1:17:24

companies in

1:17:27

the five years prior to this reset. It's probably, like, the

1:17:29

super low. Probably the lowest it's

1:17:31

ever been, rated startups just

1:17:33

because it was so easy to

1:17:36

raise money. And so you

1:17:38

develop mental models, and then the world shifts dramatically,

1:17:40

hundred and eighty degrees,

1:17:42

whatever you wanna say, like,

1:17:45

it couldn't be more dramatic, how fast it shifted. And, you

1:17:47

know, even today, entrepreneurs will say, well, I just need

1:17:50

to hold on till things get back to

1:17:52

normal. And

1:17:55

then I'm not the one or the VCs getting normal. This

1:17:58

is normal, dude. Like, that was a

1:18:00

fantasy you were in, and you

1:18:02

need to forget it fast, but you

1:18:04

can't. And there's

1:18:06

another painful thing that I I don't even jest about because it's a real it it creates real problems it's

1:18:09

actually quite

1:18:12

unfortunate. But founders,

1:18:14

whatever that peak evaluation is, they ran the math where they took their ownership, they multiplied

1:18:16

it by that number, and they

1:18:18

thought about their net worth that way.

1:18:23

And that can just be super destructive. Like,

1:18:25

once it's no longer true.

1:18:28

You mean psychologically disturbed? Psychologically.

1:18:30

Yeah. I think it's super quote

1:18:33

comes terms with 651 you've been through that. What do you mean by

1:18:35

valuation multiples are always a hack proxy?

1:18:40

So this gets into some of

1:18:42

the earlier stuff we talked about, just about how deep you go on investing history understanding

1:18:48

investors. But If there is a scale of financial

1:18:50

sophistication between one and ten, and you would say a really smart

1:18:52

person in New York is

1:18:54

a eight and a

1:18:56

half, The average Silicon Valley

1:18:58

person on financial literacy is it too. And

1:19:01

it's

1:19:01

funny because they make

1:19:04

fun of Wall

1:19:06

Street, but it's just out of ignorance. They

1:19:08

don't know anything. And so

1:19:10

most of them think about

1:19:13

valuation by a price

1:19:15

to revenue multiple. Which couldn't be a

1:19:17

recruiter tool. And at at one point, I wrote a blog post

1:19:19

called the keys to the

1:19:21

ten x revenue club. And

1:19:24

I took all

1:19:26

the public tech stocks and laid them end to end based on price to revenue. And one of them was at twenty, one

1:19:28

of them was at point one, and

1:19:30

it was just a big curve. So

1:19:35

there's no line there. There's no reason to believe that

1:19:37

price to revenue is how you should

1:19:39

value anything, but

1:19:42

it's how just because it's easy and these

1:19:44

companies are young and immature. It's hard to

1:19:46

do a DCF. It's hard to do something

1:19:49

more sophisticated. It's the common language. Of

1:19:51

the group, but things changed overnight. And

1:19:54

so someone was pointing

1:19:56

out 651 think

1:19:58

on Twitter, that and they used price to gross margin instead of

1:20:01

price to revenue still. Twilio went

1:20:03

from seventy times gross

1:20:05

margin to three.

1:20:06

In a very short window. I

1:20:08

mean, talk about valuation reset. That

1:20:10

is just -- Yeah. --

1:20:13

radical. And so it shakes the industry. And no one no one's gonna feel sorry for

1:20:15

Silicon Valley, and I'm not trying to elicit empathy. But

1:20:17

he just in terms

1:20:20

of under standing

1:20:23

what happens. It is so foundational.

1:20:25

The change is so

1:20:28

radical that the

1:20:30

best thing possibly happen is if you

1:20:32

can adjust your mental models fast and get on

1:20:34

with the new world, but it's very hard

1:20:36

for people to do. And that's And by

1:20:38

the way, what I I write that kind

1:20:41

of stuff in part to help

1:20:43

the industry. And I'm, you

1:20:45

know, I'm super grateful, Sequoia, in o nine put

1:20:47

out. There's a famous deck they put out so

1:20:49

long, good times or something like that.

1:20:51

And these things

1:20:54

help people adjust faster. Having structure, having smart people

1:20:56

tell them it's okay. It gets

1:20:58

them there faster. That I think relates

1:21:00

to number three. And you mentioned DCF just

1:21:02

for people who don't have that reference

1:21:05

discounted cash flow. So number three is you may be shocked to learn that people want to value your

1:21:07

company on FCF. That's free cash flow.

1:21:10

Am I getting that right? Yeah.

1:21:12

Yep. And

1:21:15

earnings. Facebook trades at fourteen times GAAP

1:21:17

EPS. Am I pronouncing that

1:21:20

correctly? Yes.

1:21:22

Yes. And is growing twenty three percent what earnings multiple are

1:21:24

you assuming? Question mark. So could

1:21:26

you just walk us through that

1:21:30

that bullet I'm gonna use a

1:21:31

little

1:21:31

bit of detail so you can understand. But, like,

1:21:33

I am that best of two on

1:21:35

the financials. If you think

1:21:37

all companies could trade it, like, ten times

1:21:40

revenue, here is one of the most

1:21:42

successful companies of all time that is

1:21:44

producing massive amounts

1:21:46

of positive cash

1:21:47

flow. And GAAP audited

1:21:50

earnings that's trading at a very low multiple

1:21:51

of those GAAP earnings. And

1:21:56

earnings are a small percentage of your revenue. Right?

1:21:58

So if you're trading at ten times revenue, you're probably trading at fifty times earnings.

1:22:00

Right? And here, they're trading at fourteen

1:22:02

or you may not even have earnings.

1:22:06

Because most there's a great

1:22:08

graph. We should try and find it once that

1:22:10

someone can see a link of the percentage

1:22:13

of companies at IPO that are profitable.

1:22:15

And it's this nice cyclical wave that

1:22:17

goes with these boom bus

1:22:19

cycles. We'll find it and put

1:22:21

in the show. So,

1:22:23

we does

1:22:23

risk on also. Right? And so in in

1:22:25

really dark times, the percentage of

1:22:28

companies IPOing, they're

1:22:30

profitable is like ninety,

1:22:32

but by twenty twenty, twenty twenty

1:22:34

one, that number is five percent. Like, the vast majority of companies are

1:22:36

losing money as

1:22:39

they go public. And Wall Street's encouraging that

1:22:42

behavior. And so I've often said Wall Street is the buyer of what venture

1:22:45

capital produces.

1:22:48

And if Wall Street wants high growth money

1:22:50

losing businesses. We will create as many as they can possibly

1:22:52

consume. Those

1:22:55

are a lot easier to build than the profitable ones. Way

1:22:57

easier, which is part of the cycle.

1:22:59

That's part of why you end

1:23:02

up in this cycle thing. So

1:23:04

anyway, The point of highlighting that is

1:23:06

just to try and get entrepreneurs and founders reset on

1:23:09

where the world

1:23:11

is today. Why did you and I'm sure a lot

1:23:13

of people listening will think this is a stupid question because they already know the answer, but why

1:23:16

did you highlight Facebook in

1:23:18

the way that you highlighted

1:23:20

Facebook?

1:23:20

I think on a historical basis, Facebook

1:23:22

is trading at a very low multiple. It

1:23:25

looks like a very

1:23:27

cheap stock. I mean, Coca

1:23:30

Cola, you know, for years trading at thirty, thirty five times earnings, Facebook's at fourteen

1:23:32

or mad at whatever they

1:23:34

like to be called. Yeah. And

1:23:40

twenty three percent is a pretty impressive

1:23:42

growth rate for a company of this.

1:23:44

Yeah. Right?

1:23:46

For

1:23:46

sure. Like Coke is less than ten percent growth five or three.

1:23:48

Since you mentioned meta, what do

1:23:50

you think of I mean, on

1:23:53

on one one level I have have to admire. They're

1:23:55

just doubling down on the meta

1:23:57

verse. What does your take

1:23:59

on this sort

1:24:02

of direction

1:24:03

So years ago, you know, I read snow crash

1:24:05

when it came out. And oh, it's

1:24:07

so good. I

1:24:08

thought it was the best

1:24:10

thing that I'd ever consumed. And

1:24:12

I was in hook line at Tinker.

1:24:15

And so when -- Yep. -- Philip Rosedale started second life, I was knocking on

1:24:17

his door. I served on that

1:24:19

board for twelve years. I

1:24:23

have immense knowledge on this

1:24:25

kind of immersive stuff. And --

1:24:27

Mhmm. -- Philip and I

1:24:29

actually did a postmortem podcast

1:24:32

recently, which is I would point

1:24:34

people to -- Cool. -- but what we

1:24:36

found is I

1:24:38

think there's a difference between the kind of gaming stuff this

1:24:41

idea that people

1:24:44

want to

1:24:46

live

1:24:47

experiences like they do in the

1:24:50

real world, in this virtual world.

1:24:52

And What we learned in

1:24:54

that second category. You know, I see this in the

1:24:56

Facebook demos and stuff like, oh, we're gonna

1:24:58

do a board meeting in world or

1:25:01

whatever is -- Yeah. -- that that doesn't make a lot

1:25:03

of sense. Like, the -- Mhmm. --

1:25:06

the number of people that love

1:25:08

escapism, first of

1:25:10

all, young people will

1:25:12

do. Yeah. You don't wanna wait

1:25:14

in line at the postal office in your second life. Right. Right. But

1:25:16

but but young people do it, they

1:25:18

role play a lot. Yeah. And so

1:25:22

Yep. That makes sense. And then a handful

1:25:24

of adults do it. They they have

1:25:27

wooden swords in the park.

1:25:30

Burning Man is that experience. a

1:25:32

high percentage of humans. And

1:25:34

one thing we found quite

1:25:36

interestingly is a lot of

1:25:38

the people that love it are

1:25:41

looking for an escape, so they may actually have mental health

1:25:43

problems or they're in a tough spot in their

1:25:48

And it reminds you that

1:25:50

both snow crash and ready player one were dystopian novels. Right?

1:25:52

Yeah. People were escaping

1:25:54

a world that that sucked.

1:25:58

And so we did a bunch

1:26:00

of in world board meetings, and this

1:26:02

is pre Zoom. Like Zoom is an

1:26:05

amazing substitute, which is one of the

1:26:07

frameworks from competitor strategy to the notion

1:26:09

of being in the world. And

1:26:11

so it got even harder.

1:26:13

Right? Because Zoom I think Zoom's way better for

1:26:16

a a board meeting than making

1:26:18

everyone get an avatar and sitting

1:26:20

around them. Mhmm. But I just don't

1:26:22

think that's gonna happen. It's a long answer, but I think the

1:26:24

premise they have is wrong. I don't

1:26:26

think this becomes the next platform,

1:26:29

the next smartphone. I

1:26:31

don't see that. So how do

1:26:34

you reconcile that with the position that they're trading cheaply?

1:26:36

I guess if that's just based on the math.

1:26:38

Oh, it is the -- All street. -- it

1:26:42

on my side on this one. I think if

1:26:45

they shut down the

1:26:47

VR effort, not only

1:26:49

well, the profitability would soar because they're spending

1:26:51

real money, like, five to ten billion

1:26:54

a year. But I think the stock

1:26:56

does. Yeah.

1:26:59

Where would you suggest they redirect

1:27:01

their resources? I won't even take

1:27:03

credit for this because people are saying

1:27:05

it all over Twitter. There's really good

1:27:08

buzz on some of their

1:27:10

AI

1:27:10

tools. Mhmm. And the world's super excited about that. Yes. They are. My

1:27:13

god. Look at

1:27:15

open AI and Microsoft

1:27:18

investment and so on. I mean, it's incredible. Longer conversation, but I think WhatsApp has some interesting

1:27:21

things going on.

1:27:24

You know? What's

1:27:26

happened in India? Should he give ATT shirt

1:27:28

in India? If you read about what's

1:27:30

happened in

1:27:31

India, it has a

1:27:33

similar place in the world that

1:27:35

in like accomplished more. So those things

1:27:38

are more interesting to me.

1:27:42

Yeah. So I'm curious, you know, I've had Mark Zuckerberg

1:27:45

on the podcast and he's a

1:27:47

man of strong conviction.

1:27:49

I'm curious from your perspective like, zero to

1:27:51

a hundred percent, where would you

1:27:53

put the likelihood that he

1:27:56

would ever

1:27:58

can the VR efforts and redirect.

1:28:00

Yeah. A lot of people love to discuss this.

1:28:02

And and you know what's that? Other there's

1:28:04

a bias, like, when you get pipe committed.

1:28:06

Maybe that's confirmation bias too. But, like,

1:28:09

you like, if you've already bought something,

1:28:11

you like it way more than -- Oh,

1:28:13

yeah. Some caught -- lost some caught

1:28:15

penalty. Yeah. It's super And he's had people telling

1:28:17

him what I just said for two

1:28:19

years now. So it's

1:28:23

interesting because if you look at other big bets like that,

1:28:25

I think the two most amazing

1:28:27

are AWS and Android. They were

1:28:29

both kind of out of left

1:28:32

field, not part of your

1:28:34

core business, but super successful. And I just I think

1:28:37

you've already run

1:28:40

the clock you've already spent

1:28:42

way more than either of those did in the development of those just

1:28:44

don't have the

1:28:47

numbers. Someone was there was

1:28:50

a thing going around Twitter that the

1:28:52

year over year sales of headsets is down

1:28:54

globally for the

1:28:55

industry. Mhmm. It's not happening. Alright.

1:28:57

Let's off to number four. So revenue and earnings quality

1:29:00

matter. Could

1:29:04

you

1:29:05

please explain the word quality. That's like a nuance. They like, if you

1:29:07

talk to someone who'd been investing

1:29:10

on Wall Street for

1:29:13

for fifteen years, you could talk about

1:29:15

revenue earnings and would exactly talking about.

1:29:20

But it's not something the average

1:29:22

person would know. And if that blog post, again, that I wrote called the keys

1:29:24

to the ten x revenue

1:29:26

club, I go through, like, twelve

1:29:30

different things that signify quality.

1:29:32

Okay. And and we'll link to that

1:29:34

in the shutters. Yeah. So, I mean,

1:29:37

a simple one is margins. Like,

1:29:39

if you are reselling use cars and your

1:29:41

revenue is the price of the cars you're selling, but you're only making

1:29:43

ten percent on a

1:29:47

car. That's really low revenue quality compared to

1:29:49

a SaaS vendor with ninety

1:29:51

percent gross margins. Their

1:29:54

incremental dollar of revenue

1:29:56

creates ninety cents of

1:29:58

gross margin. Yours creates seven cents of gross margin. You can't value

1:30:00

those companies both on price

1:30:02

to revenue, like they're very different.

1:30:06

651 that would be revenue quality. Earnings

1:30:09

quality typically relates to

1:30:11

cash flow. So you

1:30:13

might have really

1:30:15

good GAAP earnings but because of different factors

1:30:17

in your business, your cash flows may not be nearly as good. There

1:30:19

could be timing differences,

1:30:22

those kind of things. And so

1:30:24

anyway, you go through the the posts that are wrote, but they're just --

1:30:26

Yeah. -- elements of whether or not you have a competitive advantage,

1:30:28

whether or not you have

1:30:31

churn in your business. Is

1:30:33

this customer gonna stay around forever or they might

1:30:35

leave tomorrow? One of the reasons Coca Cola trades at

1:30:38

a high multiple is

1:30:40

everyone imagines

1:30:42

Coke will still be here fifty

1:30:44

years from now. They have no

1:30:46

reason not to believe that. Whereas

1:30:48

some of these tech companies, you

1:30:50

know, why is Facebook at fourteen? They don't know,

1:30:52

could take time, you know, take away their business and

1:30:54

then they're down overnight. Those become bigger risks

1:30:57

for tech companies sometimes

1:30:59

because they look like they might be

1:31:01

disruptive

1:31:01

more. You mentioned and you also write about and speak about competitive

1:31:04

advantage

1:31:06

a lot. What are some lesser known or undervalued

1:31:08

competitive advantages? I mean, we already

1:31:10

talked about network effect like that's

1:31:13

one that just comes

1:31:15

to mind right

1:31:16

away. Lock in. We had hinted at, like,

1:31:18

is there reasons why switching cost, which is also a, you know, back in

1:31:21

it's in the

1:31:24

order book. Like, are there switching costs that

1:31:26

make it hard to leave for a customer to leave? How many

1:31:28

substitutes are there for your product?

1:31:30

How unique is it? Right? That's

1:31:34

a competitive advantage like, hear you in

1:31:36

and of one. And some of the network effect

1:31:38

companies become that way. Like, I could create an

1:31:40

Instagram competitor, but they don't have everyone on

1:31:42

it. So the user experience is a function of everyone being

1:31:45

on it, so it's hard to

1:31:47

compete with that thing. So you

1:31:49

have a strong competitive advantage.

1:31:51

It could be performance like in an

1:31:53

enterprise product, you look at something like Snowflake. People just

1:31:55

say this this database does things, no other

1:31:57

database will do. And so then

1:31:59

that would come a

1:32:02

competitive advantage. But it's how hard

1:32:05

is it for someone to

1:32:07

find an alternative to

1:32:09

you and trade you

1:32:11

out? This is the easy way to say it. This is not

1:32:12

directly related to competitive advantage, but I

1:32:14

wanted to just revisit the total

1:32:18

addressable market your open table

1:32:20

story with the spreadsheet or the model that had

1:32:22

been capped at seventeen percent. We're going to

1:32:25

ninety nine percent. It's

1:32:27

not seventy percent. Just to reflect back

1:32:29

also on Uber since you and I were

1:32:32

both along

1:32:34

for that. Right? Unintended. And in the early

1:32:36

days, you know, there's three black Gurley. Two or

1:32:38

three black cars in the very very

1:32:41

early Gurley, sort

1:32:43

of prototyping. And I

1:32:46

remember sitting with Garrett and looking at some of the

1:32:48

potential market sizes, but the

1:32:50

assumption always was that the

1:32:55

high could grow and should grow if

1:32:57

Uber's functioning effectively and that's what

1:32:59

ended up happening. So

1:33:01

you can't say you know, there are seven hundred

1:33:04

black cars in San Francisco. What

1:33:06

percentage of those will use Uber

1:33:08

and the the upper limit is seven hundred

1:33:10

because then in the matter of a handful

1:33:12

of years, you have a thousand plus

1:33:14

-- Yep. -- black cars because of the rider demand. So I just wanted

1:33:16

to mention that since you'd

1:33:19

also mentioned it earlier. Yeah.

1:33:22

I wrote a very long piece on this because there's a famous, I think, NYU professor,

1:33:27

Azwoth, his last name is

1:33:29

not coming to me right now, but he's famous for valuation work. He's always on And

1:33:33

he wrote

1:33:36

a piece that said Uber wouldn't be worked

1:33:38