The Power of Data: Lessons from Shazam Creator

Thank you.

Thank you. It's gonna be hard to live up to the energy of

that video, but, so my son's about to turn 17

and I said to him just recently, I said,

I have all these ideas for gifts,

this is your last year being a kid.

What gift do you want? And he said, I want money.

So you're in the right business.

So who here uses Shazam?

All right, that's a lot of Shazam users.

Well, have you ever had…

I have a question for you. Have you ever had a crazy idea

for something that's impossible?

Like a time machine.

A time machine that you can use to go back in time

and be the very first person

to buy front row Coldplay tickets?

I forgot. Be careful who you bring.

Or maybe here's one that I want. An app.

An app that you can use when you're doing your laundry

and it finds the missing sock, right?

An idea for something that's impossible.

Well, I had an idea for something that was impossible.

I thought, gosh, well we have these mobile phones,

wouldn't it be so cool if you could just pull out your

mobile phone when you hear a song

and it would just figure out what the song is?

And you could use it anywhere.

You could use it like driving in your car, listening

to the radio, in a bar, a club, a cafe, a movie theater.

Even in a little grocery store.

I was like, wow, that'd be such a useful thing.

I would love that. That was my idea. But I'm non-technical.

So I went to the deep technologists that specialize in

that type of technology.

These are professors

and PhDs in audio signal processing, a subset

of electrical engineering.

And I went to people at Stanford

where they have a group called the Center

for Computer Research and Music

and Acoustics at MIT Media Lab at UC Berkeley,

even at Xerox Park.

And I met with these PhDs and I described it

and I said, I want to build a business

that identifies songs.

Can you please build that technology for me?

And they said, there is no technology that can do that.

And we have no idea how to make it.

See, there were technologies at the time

that could monitor radio stations, like nice clean signals

of sound, no background noises of people talking.

And you only have to ping each radio station once per minute

and a relatively small database of songs

that are played on the radio, but to do it against millions

of songs and with people talking in the background

and noise, and every single time someone shazams a song

to do that at scale

and with noise, that's what was impossible.

So who here has never used Shazam? Never used Shazam.

Okay. There's not that many people.

I see a potential new user though. So what's your name?

I'm Mary. Mary. Nice to meet you, Mary.

Let me show you how Shazam works.

There's nothing you have to do, so don't get nervous.

It's simply an app that you download for free

and when you hear a song

and you want to know what is,

hit the button And that's it.



And it's free, so what do you think?

You might download it. Oh, I'm going to, yeah. Okay.

I love music and I'm always like, what's that song?

Okay, awesome. Well, one more user. Welcome to Shazam Mary.

Thank you. Okay.

See, I mean, when you're an entrepreneur,

you'll do anything to get downloads.

Oops, actually, Mary,

while I think of it, just to make sure you don't forget.

I've got a pair of Shazam socks for you

so you don't forget to download that app.



Oh, maybe you never

know anything to get downloads.

So it all looks pretty simple, right?

But it was not that simple to create.

And the truth is that Shazam almost failed from the very

beginning, and it's all my fault.

See, my original idea for Shazam was not good enough.

My original idea for Shazam was to use what,

what I knew existed these technologies

that monitored radio stations, right?

And so I thought, well,

if these technologies are pinging radio stations in order

to monitor radio play that was used

for royalty distribution, I said, I'll tie into

that information and I'll create a mobile phone service

where you would type in the radio station number

that you're listening to, and then it would tell you

what song is playing on that radio station at that time.

Now it'd be a bit clunky of an experience, right?

Think about those app

experiences that you create, right?

When you're a credit union or a bank or whatever.

So you want it to be simple.

But think about this, you'd have

to type in the radio station number every

single time to find out the song.

And of course you'd have to know the radio station

that you're listening to, and it wouldn't work if

it's not the radio, right?

So bars, clubs, cafes, movie theaters, television,

all these places you hear music.

That's not the radio. It wouldn't work,

but at least we could implement it.

So imagine that we raised millions of dollars,

hired tons of smart people, worked really, really hard,

launch that service.

Do you think Shazam would still exist today? No. Right?

Yeah. And how frustrating would that be, right?

Worked so hard, raise money, do all this hard stuff only

to become irrelevant.

And that's what happened

to the six companies that did do that.

Irrelevant, gone, defunct.

Now, before Shazam almost failed, I almost failed

because I have dyslexia.

And so I always struggled with academics as a kid.

When I was in fifth grade at a public school in San Diego,

there were 60 kids in my grade.

So they said, oh, well

we'll split 'em into these

two classrooms with two different teachers.

We'll put the smart 30 kids in this classroom

and the not so smart 30 kids in this classroom.

And that's where I ended up.

I can remember sitting up against the wall

and I could hear the kids in the smart classroom having

so much fun with their special projects.

Now I worked really, really hard all the way

through middle school, high school,

and I got in to UC

Berkeley Civil Engineering for my degree.

Wow. I was so excited. But you know what?

I got there and everyone was so smart.

Everything was so difficult. I felt so lost.

My first semester at UC Berkeley,

I got a 1.7 grade point average, failing grades.

And you know what they do? They write a letter

and they send it directly to your parents

and it says, your child's gonna be kicked out of UC

Berkeley next semester if they don't get their grades up.

Now, there's one more thing that happened to me as a kid.

And when I was in ninth grade,

my biology teacher wrote a report card to my parents,

and the report card said, your son is defiant.

Your son questions my authority.

He questions the things I teach.

It was written in a negative tone.

But you know what I've been thinking?

Maybe defiance is not such a bad thing.

When I see my son being defiant, I think, hmm,

He has potential.

I mean, I like to make magic shows

as a kid and what is magic?

But defiance, right? Defiance of logic.

So what if you could use defiance

to create magic?

Because that's what we did with Shazam.

We created something so magical, so magical Mary,

that I got 2,000,000,001 downloads, right?

Something so magical that it became a verb, became a verb.

You know, you Google something and then you want to know

what the song is, you Shazam it.

That's every product creator's dream.

And then something that even Google didn't achieve, which is

that many people, and especially Mary will tell you,

they can remember the very first time they used Shazam.

They'll be like, oh yeah, I remember that. I was in a car.

And, my friend said, just look at this. It is an app.

You just push the button and it identifies the song.

It was like magic. It literally was this memorable moment.

So how did we do this?

My friends ask me, they say, Chris, you founded Shazam.

Why in the world would you give a speech?

And this is the reason.

See, I identified in creating Shazam

and coming up with the idea for Shazam, these different ways

of thinking to be innovative.

But then after starting Shazam,

after four years, I stayed on the board

and I joined Google in 2004 when it was a private company

with 2000 employees, six months from going public

as their first mobile partnerships employee.

I was the second witness at the recent Department

of Justice versus Google Antitrust trial.

And then at Google, my eight years at Google,

I saw the way they innovated, the way they thought,

the way Larry and Sergey thought,

I joined the Android team when there was only one Android

phone in the world, the G one on T-Mobile.

And after eight years at Google,

I then joined Dropbox when it was a little startup

with 90 people, duct tape covering the wires on the floors,

and again, seeing the way they think.

And these are the ways to think differently.

They're not the ways we think by default.

To think differently, to bring amazing things to life,

build from basic truths, apply creative persistence,

eliminate friction, and connect to emotions.

So imagine for a second that you work in this restaurant

and you are cooking these delicious dishes.

You have these amazing recipes.

But now imagine that you're the chef in the restaurant

and you have to invent a delicious dish.

How does the chef go about inventing a delicious dish?

Which way is right? Which way is wrong?

How do say that I need to on, you know,

like close to love you?

Well, who would've thought that a chef thinks the same way

as a great innovator?

The great innovators like Steve Jobs, Henry Ford,

even Leonardo Da Vinci.

Because those innovators do what's

called first principles thinking

or what I call build from basic truths.

And a chef, they want to invent a recipe.

They start from the raw ingredients, the basic ingredients.

And that is first principles thinking

or build from basic truths.

You break down the basic truths.

You have to make sure you're not making any assumptions.

You question every single assumption like,

I see there's this radio recognition service.

So we must have to tie into that

to create a music recognition service on mobile phones

because that already exists.

Or I see there's these APIs available from open AI,

and so I must need those in order

to create my new AI feature, right?

That's making assumptions.

You question all the assumptions

and start only from what you know

to be the basic truths, the raw ingredients.

And then you build up the possibilities

for your breakthrough new idea.

So I had this problem in my life that frankly

so many people had at the time, which is

that sometimes I would hear a song

and I'd think, wow, I like that song.

I might be out at a bar or a club or with a friend

and I'd be like, I really love that song.

And the only solution I had was to be able to

ask people, do you know what that song is?

Do you know what that song is?

Most of the time people didn't know,

but once in a while people did know the name of the song.

And if they did, I wrote it down.

I kept a list of my favorite songs.

And when that list got long enough, I made this thing

called a mix tape.

Remember those? Now there was a second problem at the time.

Who here had this phone? That's a lot of people. Wow.

We all thought it was such a great phone, right?

Like, look at my new little Nokia phone.

I'm thinking, at this point, I did finish UC

Berkeley, and I actually went back and did an MBA.

So I'm now doing my MBA at Berkeley

and everyone's starting to carry

mobile phones around with them.

And I'm thinking, you know what?

There's really only three things you

can do with these mobile phones.

You can make phone calls, you can send text messages,

and you could play a game called Snake.

And that was it. And I'm like, if I could come up

with another idea of something to do on these mobile phones,

all these people walking around the world

with mobile phones in their pockets,

just think of the possibilities.

That was way back in the spring of

1999 when someone named Sean Mullins

was in the music charts.

Who remembers that song? Yeah. Wow.

It's really lasted the test of time.

So the traditional thinking, right?

The traditional thinking is what I originally did.

And what these six other companies

that are now defunct did, right?

There's this radio recognition service, we'll tie into

that and we'll make a service as a clunky experience

where you type in the radio station

number and blah blah, right?

That was my traditional thinking. That was my original idea.

But I was laying in bed at this point,

I'm doing a semester abroad for my Berkeley MBA,

I'm in London at London Business School,

I'm living in Notting Hill and I'm laying in bed

and I'm thinking, I'm gonna build this music recognition

business radio recognition.

And I'm thinking, now, hold on.

What could someone do that would leapfrog me?

What could someone do that would make me irrelevant?

That is a great way to think when you're trying

to be innovative in your business.

And so I was thinking that. And then I suddenly thought,

wait, hold on.

What if you don't even need

to know what's playing on the different radio stations?

What if you could just do it from the sound in the air

of the music coming to the microphone of the mobile phone,

despite the background noises of people talking

and cars honking?

I thought, oh my gosh, if you could do that

then you wouldn't even have to type

that radio station number each time.

Wow. And wait, it would work even if it's not the radio.

You could use it in bars, clubs, cafes, movie theaters,

watching TV at home everywhere.

I thought, oh my gosh, that is what I want to do.

That is what I wanna do no matter how hard it is to do.

That was the aha moment of Shazam.

And that was in the fall

of 1999 when Lenny Kravitz was in the chart,

Some great songs from the time.

So I didn't realize it,

but I was building from basic truths.

Because the two basic truths are that every mobile phone

has a microphone, right?

Otherwise you wouldn't be able to talk into them.

And all music is sound. It's audible, right?

Those are the two basic truths from which I had built up a

new possibility, a new breakthrough idea.

So when you're thinking through in your own businesses,

as you're thinking through what, what is the perfect

financial services experience?

How can we make it easier? What could we do?

How can you question the assumptions?

Don't just think about what's available today with APIs.

What could you do if

you had just had a clean slate?

How can you question all the assumptions

and come up with a breakthrough idea for your business?

Now, building from basic truth is actually a form

of defiance because you're defying your own brain.

See, our brains are not wired to do this.

Our brains are actually wired to make assumptions, to think

by analogy, to think from what we know.

If we didn't do that, you wouldn't

have made it to this room today.

How do I know this is my doorknob?

How do I know this elevator is safe?

How do I know I'm in the right hotel?

If you constantly question assumptions,

you would just never get anywhere.

You can only do this when you're trying to come up

with a big breakthrough idea.

That's when you have to go the extra mile.

Question all these assumptions and break down

to the basic truths.

Okay? Now imagine that you are in prison serving a life

sentence and you come up with a breakthrough idea.

I am gonna escape from prison.

Is that one idea enough to get you out?











Well, obviously it's not enough to get you out, right?

I'm gonna say, okay, you're gonna escape from prison.

How are you gonna escape from prison?

I'm gonna make a hole in the wall of my cell.

Oh, how are you gonna do that? I'm gonna steal a pick

axe from the work shed

when the guard's not looking

How are you gonna make sure they don't notice

the pick axe in your cell?

How are you gonna make sure they don't notice the hole

that you're building in the wall of your cell?

Right? Where's that gonna go?

When you go out that hole in the wall,

are you gonna be able to get outta the prison?

Notice all the obstacles you face?

All the problems you face? you're gonna have to come up

with solutions to all of them.

And the way you do that is through creative persistence.

See, people talk about entrepreneurs just being very

persistent, but actually in reality,

it's the creative persistence.

It's coming up with creative solutions

to get past all these barriers.

When I joined Google, there was one product

they had, it was Search.

They didn't even have Gmail.

And they said, Chris, one thing we're searching now, a new

project, Project Whale,

it was called; Code Name Whale.

And they were gonna make it so you can not only search the

web, but you could also search all books.

And so they were scanning books at scale.

But you know what they realized? This is

a very laborious process.

So they employed creative persistence.

They invented a device that automatically turned the pages

of books so they could scan them at scale.

That's creative persistence.

See, when you have a big breakthrough new idea

and you're trying to implement it, you're gonna face so many

obstacles that it's gonna feel

like the system was set up to stop you.

Kind of like a prison. Now, to start off with,

I thought I need a couple co-founders, right?

So I got two co-founders.

I got Philippe. Philippe's on the top of the picture there.

Philippe is from Belgium,

and Philippe has the productivity of 10 super humans.

So he was a great co-founder.

And then I got Dhiraj

and Dhiraj, well he had grown up all over the world,

and he had this like zen-like quality that kept us grounded

through the most stressful of times.

And there were plenty of those, but we were all

non-technical and we're starting a tech company.

We needed a fourth co-founder, a genius that had a chance

of inventing this impossible algorithm.

And remember, we're meeting all these PhDs

and professors that are specialized in audio signal

processing, and they're all saying it's impossible.

I mean, it's so hard, it was such a hard thing

to invent that when we did find this fourth co-founder,

and he did join and he did work

to invent the algorithm for Shazam,

he was writing notes on a whiteboard.

And I took a picture of the whiteboard.

This is the actual picture of him inventing the

algorithm for Shazam.

And I don't understand any of them. Do you?

I mean, you're probably thinking like,

how do you pick this co-founder, right?

How do you pick this perfect employee

to solve this really hard problem, right?

And I thought, I had noticed

that I was interviewing all these PhDs

and professors from MIT, Media Lab, Stanford Karma Center

for Computer Research and Music

and Acoustics, Berkeley, Xerox Park.

And as I interviewed them, I realized they're all geniuses.

And I didn't understand a word they were saying.

So how am I gonna pick the smartest one that has a potential

to invent this impossible algorithm?

And then I thought, I know creative persistence.

I'm not gonna pick them.

I'm gonna get Professor Julius Smith from

Stanford to pick them.

So we made Professor Julius Smith from Stanford,

an advisor to Shazam.

Professor Julius Smith had invented the algorithms

that are in every electronic keyboard in the world.

Professor Julius Smith had been the head

of audio technologies reporting directly to Steve Jobs.

Professor Julius Smith had been hired by the US military

to detect submarines under the water using only sound.

He was worldwide eminent in music

and acoustic signal processing.

And so we collected a bunch of names of

PhDs from all these institutions and made a list.

And in our first meeting with Julius as an advisor,

we sat down in his living room in Palo Alto, California.

And we said, Julius, can you please rank the five

smartest people on our list?

Because he knew them all.

And here's the actual notes from this meeting in case you're

looking to hire an audio signal processing expert,

the number one rank on the list was Avery Wang.

Avery has four degrees from Stanford in Mathematics,

Electrical Engineering, Audio Signal Processing,

Computational Neuroscience.

And he joined as the fourth co-founder of Shazam.

Now you're probably thinking, look,

it's just a button You push, the button it gives,

tells you the name of the song.

I mean, nowadays, AI does all this stuff. Why is it so hard?

Why are all these professors

and PhDs saying it's impossible?

I mean, why? What's so difficult about it?

See, it turns out that in pattern recognition,

and pattern recognition includes eyeball recognition,

voice recognition, fingerprint

recognition, music recognition.

That's all pattern recognition. In pattern recognition,

when you combine two challenges at the same time,

that's when it becomes impossible

or certainly extremely difficult.

And those two challenges are noise,

like people talking in the background

and scale a huge database of songs.

So if we had to invent an algorithm

that had only 10,000 songs in the database,

but people talking in the background, okay,

we could maybe do that

or maybe clean signals from the radio

against millions of songs.

Okay, maybe we could do that.

But you combine noise and scale

and that's when it becomes such a

difficult, difficult thing.

For example, you walk into an airport, right?

And you know, you stand in front of that camera

to travel internationally.

It looks at your face and compares you

to everyone in the world and then opens the gate

and lets you through, right?

But notice that you have

to stand right in front of the camera.

You can't just stand anywhere in the airport.

'cause that would be noise and scale.

So if I said to you, let's use your own brains

to do pattern recognition, right?

So you would look at this picture

and I'd say, okay, try to find Tom Cruise.

So you look in the picture and then all the other people in

the picture represent the noise.

And then you, your brain finds Tom Cruise.

Now do it here. That's noise and scale for your brain.

So not only did we have to deal with noise,

but we had to deal with noise cancellation technologies

built into mobile phones that emphasize the human voice

and deemphasize all other sounds, including music.

Audio filtering, where they would,

cellular networks would chop the top in the bottom

of the sound spectrum and only send over the cellular

networks, the portion that carried human voice

for efficiency reasons.

Speed, variation, where the DJs would, you know, play music

and then change the speed of one of the records, right?

So they can match the beats

and then pitch correction if they're using CDs

or digital music and they change the speed of the music,

but then they also change the pitch so it doesn't start

to sound like the chipmunks.

That's a non-linear transformation of the music

reverberation that echo effects you get in big

rooms. Sound distortion,

if the speakers report quality, we had to make sure

that the algorithm we invented works

across all these things.

So to give you a sense for what it was like, I'm gonna play

for you now 15 seconds of a song.

And now I'm gonna play that exact same 15 seconds of

that exact same song

as we recorded on a mobile phone in a bar.

Now you can see why those professors are going,

I don't know how you're gonna do this.

Right? And we worked on it for months and months

and frankly almost gave up.

In fact, there were times that Avery

is thinking, you know what?

I'm gonna have to tell Chris that, Chris

your startup is just not gonna work

because I can't invent this technology.

And so finally one day,

Avery was working at a cafe in Palo Alto, California,

and he came up with a breakthrough, a breakthrough idea,

a novel invention.

It turned out to be the very first AI invention in history

that would become a mass consumer product.

And what it was, there was a three dimensional graph

of sound of spectral peaks, energy peaks

and sound graphed both against time,

but also against frequencies.

And then combinatorial, hashing the same technique

that Google uses to make a very fast index search.

That breakthrough idea, that invention

happened on the very same day

that Coldplay released their very first song,

An album that I listened to over and over

and over while we built Shazam.

So now we had this invention, we filed a patent,

but now we need money.

We need lots of money. We need seven

and a half million dollars in venture capital money

to build this business.

But you have to keep in mind, the .com bubble had just

popped and like we're pitching to venture capital firm

after venture capital firm after venture capital firm.

We were based in London.

And they just kept saying, no, no, no.

We pitched over 100 venture capital firms

that just turned us down.

I can still remember pitching

to one big giant venture capital firm, very famous

that you all know, but I won't say who.

And I showed the head of the firm,

the demo of Shazam on this little old Nokia phone.

And he said, I don't see why anyone would ever use that.

And I thought, I've got to prove this guy wrong.

I've got to prove him wrong. Now, he didn't invest,

but we did raise seven

and a half million dollars from three venture capital firms,

one of them being Richard Branson version

group's venture firm.

And now we can embark on building the business.

And the first thing you need,

if you're gonna build a music recognition

business, is the music.

Now, there were no digital music

databases available at that time.

There was no iTunes, no Apple Music, no Spotify

music was in the form of CDs,

and we did not have the CDs.

I mean, we're a startup. What are we gonna do?

Buy one of every single cd.

So I had an idea, I thought, I know.

So we went and met you see, this is in London.

And the two big record stores in London at the time were HMV

and Virgin Megastore on Oxford Street.

So we went and we met with the executives

of Virgin Megastore,

and I said to them, our investor is Richard Branson,

can we have access to your CDs?

And they said no.

And I was like, oh no, what are we gonna do?

So then we went and we went to this, to this company

that was the distributor of CDs.

They were a warehouse an hour outside of London.

And we met with those executives,

I can still remember their grizzly gray beers.

And we said to them, the world is changing.

The world is going to digital.

Wouldn't it be great to have one digital copy

of all that music?

Because we can build it for you for free.

And if you're thinking what I'm thinking,

the next thing you know, we had a partnership

where we didn't pay a single dollar,

but we hired 30 18-year-old kids

that worked in three eight hour shifts, nine months a year,

24 hours a day, 24 hours a day, going

to the racks of their warehouse,

picking the CDs off the racks, taking them over to computers

that we had wrote, built custom software

that would take the music off the CDs.

And then they had to type the name of every artist,

every album, title, and every song.

Because that information does not reside on the CD.

24 hours a day for nine months.

That's creative persistence right?

Now, when I first came up with the idea for Shazam,

I was like, I thought I'm starting a music recognition

business, but I'm not sure I realized that really

what I'm starting is a search engine,

not a search engine of the web,

but a search engine of music.

Because every time you Shazam a song,

you're giving us a few seconds of sound.

And then we have to search and compare

that against every few seconds

of every single song in milliseconds

and do a big search.

So we knew there was this company

that had started just a couple years

before us called Google,

and they had built their own search engine.

They had pioneered a model called a Beowulf cluster,

where they had taken a whole bunch of PCs, commoditized PCs,

wired them all together, and then built custom software

that they pioneered and created

to turn all those PCs into one giant supercomputer,

one giant search engine.

And so we decided to do the same thing.

We even got a tour of the data center

where those Google servers sat,

and then we hired a really smart engineer

to build our own Beowulf Cluster, write all

that really hard code

to turn all those PCs into one giant supercomputer.

Now, I don't know if anyone here might be thinking,

yeah, but Google became a much more successful

company than Shazam.

But whose search engine do you think is better looking?

That was the spring of 2002.

So think about where we are.

We are a year ahead of the iPod, right?

Let alone iTunes. There's no market for digital music.

People didn't buy digital individual songs.

We were seven years ahead of the iPhone,

eight years ahead of the iPhone app store.

So we actually had this beautiful colorful PowerPoint

presentation that we showed the venture capitalists

that anticipated all the things we'd do

as mobile phones evolved,

but we had no idea how far away we

were from being able to do it.

After you Shazam a song, you'll be able to read the lyrics

of the song, couldn't do that.

And then you'll see a picture, colorful picture

of the album cover, couldn't do that.

And then you’ll be able to download the song to your phone.

You couldn't do that. And then you'd be able

to share it with a friend.

There was no Twitter or Facebook.

And then you'll be able to share it

with a friend and then you'd be able

to add it to a playlist.

There was no Spotify or Apple Music.

I mean, we had all these ideas, they're in the presentation.

We were years away from being able to do any of it.

So the way that Shazam launched in the summer of 2002 was

as a telephone number that you called,

you took out your phone if you were in the United Kingdom,

our first market, you took out your phone

and you dialed a four digit number that we secured

by partnering with each

of the four major mobile operators in the United Kingdom.

2-5-8-0, the only four digits

that are in a straight line on your mobile phone keypad.

And you dialed 2-5-8-0 kind

of like 4-1-1 or 9-1-1.

And it initiated a voice phone call into our interactive

voice response system, which automatically answered the

phone and said, hold your phone to the music.

You held your phone up in the air, we recorded 15 seconds

of sound, we terminated the phone call,

but we had your caller id.

And so a second later, we sent you a text message

with the name of the song.

And that was Shazam. Pretty cool, huh?

Thank you. That was in the summer

of 2002 when Moby was in the charts.

Some of my favorite songs from the time.

Creative persistence. Because you have

to defy barriers.

You have to defy these barriers, right?

Because when you have a big new innovation,

a big new idea you're trying to implement, there will be

so many obstacles, so many barriers that you have

to get past as if the system was set up

to stop you, like a prison.

See, there's something that we all do, everyone here,

including myself, which is that, sometimes

we have some idea, we encounter a barrier

or an obstacle, and then you know what?

We respect it like an authority,

and we say it can't be done.

Oh, it can't be done. And instead we have to defy it,

defy it, and find a creative way past it.

And that's how you implement creative persistence.

Okay, so in our lives, in our personal lives

and in our professional lives,

we all encounter problems, right?

Sometimes those problems are small problems

and sometimes those problems are big problems.

And when they're like big problems,

I compare them to like a boulder.

But can you always spot a boulder coming at you?

Okay, you're thinking, where's he going

with this whole pebble boulder thing, right?

So I was thinking the other day,

we all know that we wanna eliminate friction, right?

We all know we wanna make things simple.

And, when you think about all the things you can do,

right on apps and AI

and stuff, there's so many possibilities

to make things simple for your customers,

for your suppliers, your partners, your employees, right?

You can make things simple. Every executive knows that.

We all know it. And we all look at Apple

and we say, gosh, Apple, look at that company.

It makes so much money, commands such big

margins, and it's worth so much money.

How do they do it? Oh, well, that's how they do it.

They, they make things simple

and they eliminate friction.

They delight customers. That's how they do it.

And so we all know it. We all know it.

We all know that's the secret.

And yet there's still so much friction out there, right?

There's still so much friction out there.

I mean, I'll give you one example.

My son has a Minecraft account,

and I just wanted to change the email address

that he uses for his Minecraft account.

You do that on microsoft.com,

and I still haven't succeeded.

That's how hard it is.

I mean, there's still so much friction out there

and I’m thinking, well, why is this? Why is it

that we all know

that friction, we need to eliminate it.

And yet there's still so much friction.

And I came up with the answer, this is the answer.

The answer is that often we see a piece

of friction in our business

and we think, ah, the user just has to click here.

You know, they just have to click here.

They just have to wait 24 hours for it to process, right?

And we think it's not, that's not a big deal.

It's just a pebble. We think it's a pebble,

but actually it's a boulder.

And that's the problem.

We think it's a pebble, but actually it's a boulder.

So I think of friction as anything that someone has to think

or is anything someone has to do.

So for example, imagine you're renewing

your passport online, right?

So it goes, okay,

please enter your known traveler number.

Ah, what is my known traveler number?

You have to know something, friction.

And then it says, make sure when you submit a passport photo

that there's a quarter inch to the right of your head,

a quarter inch to the left of your head, half an inch

above your head in a white background,

or we're gonna reject the picture.

You have to learn something that's friction.

And then it says, do you want the regular passport renewal

service or the expedited passport renewal service?

You have to choose something that's friction.

And then it says, your new passport will be ready

in 4 to 12 weeks.

4 to 12 weeks. You have

to interpret something that's friction.

And then it says, don't worry, your passport

might be ready in time for your international flight.

Might be ready? I need

my passport for the international flight!

Right? Concern is friction, right?

Initiate the process here, register here,

input your last three home addresses.

Wait this amount of time. And if you're wondering whether

your passport's almost ready, just log in

and you can check the status.

Monitoring is friction.

See, what we need to do is really need to reduce effort

for others, not for ourselves.

And sometimes we have to move mountains

to eliminate this friction.

And I saw this first hand when I joined Dropbox.

See, when I joined Dropbox, they told me a story.

They said, Chris, we are not the first cloud service.

There were predecessors, many, many predecessors.

There were two that called xDrive

and iDrive that had raised

millions of dollars in venture capital.

And they were nearly identical to Dropbox.

Just like Dropbox, you downloaded a piece

of software onto your computer

and you could put your files in the cloud.

The one difference with xDrive

and iDrive was that with those services,

when you wanted something in the cloud,

you had one click upload.

And when you wanted something back from the cloud,

you had one click download.

And I guarantee you that xDrive and iDrive

and all the venture capitalists that gave them millions

of dollars, all said, that's just one click.

I mean, it's so easy, it's just a pebble. But you know what?

That one click wasn't just a pebble, it was a boulder.

And they got no users at all.

Dropbox said, the Dropbox founders were these like

smart MIT engineers,

and they said, you know what? That little one click upload,

one click download, that's not a pebble, that's a boulder.

And we're gonna move mountains to eliminate it.

So they did two really,

really hard things, two really hard things.

First, they hacked into the file

and folder system of Mac OS, PC Linux.

And they made it

so they can make this little folder

that had a Dropbox icon on it.

And it looked just like all the other folders,

but it was a magic folder

because if you stuck your files in that folder,

they were also in the cloud.

There was no one click upload, no one click download.

If they were in the folder, they were in the cloud.

And then they realized, wait, you might

change one of those documents.

You might edit the PowerPoint or edit the word.

And so they had to build synchronization so that as soon

as you did that, it synchronized that file with the cloud.

And when the cloud had the same file

that you had on your computer,

that was an extremely hard problem.

Because there's so many things

that can break when you're doing synchronization.

You know, you shut your computer

or the wifi goes out, right?

When you're synchronizing a file.

And I'm sorry, but we just corrupted your document

and you lost all your work forever, right?

Clearly not acceptable.

So they had to solve synchronization.

And this is something that was so hard that even Microsoft

and Google hadn't done it.

They were the first. And by solving synchronization

and by solving, hacking into the file

and folder system, they created a magic folder

that sat on your computer.

If you put a file in it, it was also in the cloud.

And you know what they said to me? They said, Chris,

they said, Chris, our predecessors, xDrive

and iDrive, our predecessors had one click

upload download, right?

We have zero clicks.

If it's in the folder, it's in the cloud.

And the difference between zero

and one is infinite.

I will never forget the moment they told me that

the difference between zero and one is infinite.

And that's why Dropbox was the first service in the cloud

to rocket to hundreds of millions of users.

And eventually Google and Microsoft, of course, replicated

that technology as well.

Now, as a part of the Shazam story that I left out,

we thought we had done everything.

I mean, we're like, wow, we invented the algorithm

that everyone told us was impossible.

We built a search engine like aBeowulf cluster, just like

what Google built.

We built a music database by hiring 30 18 year-olds

and having them working 24 hours a day

for nine months, right?

We got a 2-5-8-0

number from all the major mobile operators,

the short code that no company had ever

gotten a voice short code before.

Not 1-800-FLOWERS. Not United Airlines, not Domino's Pizza.

But we did. We did everything. We did everything.

And we thought, you know what? We're gonna launch

and we're gonna be a huge smashing success.

And it didn't happen.

We got barely any users, barely any users.

Has anyone here ever had

a really bad year?

How about two really bad years?

'cause we had six. Six consecutive years

of burning through cash, struggling,

almost going bankrupt, having almost no users

and having no way to figure out how to get them. Six years.

It's a miracle that we even survived that time.

And the reason it was six years is

because it was six years from when we launched the service

in the summer of 2002 Until,

well, when the app store came out on the iPhone in the

summer of 2008, six years.

And you know what? It's all

due to friction.

I mean, think about it. You had to know something with

that old Shazam 2-5-8-0,

it wasn't just an app on your phone.

You have to know something that's friction.

And then if you wanted to buy the song

that you just identified,

you couldn't just click download on iTunes

or add to Spotify playlist.

You had to go in your car, drive to the record store,

show the person the text messages

and say, I wanna buy that cd.

That's a lot of friction, right?

And then also, we had to make money. Back then

there was no selling digital music

or advertising on mobile phones at that time.

And so our only business model we could do was to do

what directory inquiries 4-1-1did at the time

charge your mobile phone bill

50 cents every time you use Shazam.

And we share that with a mobile phone company.

But if you charge people 50 cents every time they Shazam a

song that is friction.

And it's only when the app store enabled us to eliminate all

that friction, that Shazam rocketed to hundreds

of millions of users.

So when you think in your own businesses,

in your own businesses, where can you find a piece

of friction, right?

Find somebody that looks just like a pebble,

but is actually a boulder and remove it.

Create a Shazam button like experience

where you just push something

and you get the user, the customer, the supplier,

the partner, gets a delightful solution instantly.

Eliminating friction is about defying efforts.

It's about looking in

and experience end to end from the beginning all the way

to the end, searching for that little bit of effort,

that little pebble that might actually be a boulder.

And then moving mountains to eliminate it. Okay?

Now imagine for a second that you inherit a piece of land

and you think, oh gosh, a little piece

of land out in the countryside.

I'm gonna build my dream house.

What will it be like building your dream house?

And how will it be different?

Well, obviously if it's your own house now

you're really gonna care, right?

You're not just buying a house,

you're building your dream home.

You're gonna be like, oh my gosh, I want those windows

and skylights

so the sun rays will come in right

where I make coffee in the morning

and I want those hardwood floors that are so rustic,

and I want that doorknob on the front door that's so easy

to lock and so easy to unlock.

You're gonna care about every little detail

'cause you're emotionally connected,

personally connected to what you're creating.

And we, the founders of Shazam, were emotionally connected

and personally connected to creating something amazing.

We cared so deeply as did all their

early employees of Shazam.

We cared about the name, the color, the touch, the feel,

the speed, the features.

We cared about everything.

But something very scary happened.

Giant companies replicated our technology

and launched their own services.

And it terrified me. I thought

we were gonna be made irrelevant.

Phillips created a technology

and then they offered to license it

to any company in the world that was willing

to build their own music database

and their own search engine.

Sony bought that technology

and made a music recognition service preloaded on

every Sony Erickson phone.

Google made their own music recognition service

and put it right in the search app on every Android phone.

That's 80% of smartphones in the world. But you know what?

We ended up with 90% market share

of music recognition around the world.

And why? 'cause we cared so much.

We built all these amazing features, synchronized lyrics,

offline Shazam, auto Shazam, localized concerts

and charts super fast

and amazing music coverage to make sure

that we delighted our users.

And because we delighted our users,

the users loved Shazam.

When we came out, it was an app.

We had huge, hugely high ratings even compared

to giant companies with enormous, enormous resources.

And then Apple did something they very rarely do.

They made a public announcement. One year

after the app store, they announced, they said,

we're gonna announce the most popular apps in the world.

There was Google, there was Facebook, there was Skype,

and there was little old Shazam.

Now, I don't know if any of you would do this, but I did.

I went to my computer and I thought, what was the name of

that venture capitalist guy again?

And I sent him an email.

It turns out people will use this.

That was the summer of 2009

Connecting to emotions is making it personal about thinking about the

work that you do, thinking about

how it impacts other people's lives.

'cause we all need money. We all need accounts.

And it's critical to the way we live.

And then feeling the impact of what you do.

So when I show you this old 1997 Apple television

commercial, ask yourself,

how did these people impact the world?

Here's to the crazy ones, the misfits,

the rebels, the troublemakers,

the round pegs in the square holes.

The ones who see things differently.

They're not fond of rules

and they have no respect for the status quo.

You can quote them, disagree with them, glorify

or vilify them about the only thing you can't do is ignore

Them. Because they change things.

They push the human race forward.

While some may see them as the crazy ones, we see genius

because the people who are crazy enough

to think they can change the world are the

ones who do.

The people who are crazy enough

to think they can change the world are the ones who do.

So think how you can make it personal.

Connect to emotions to really, really,

think how you can defy your brain.

How you can defy your brain by building from basic truths,

questioning the assumptions, and building from basic truths

from the core things that you know to be true.

Think how you can defy those barriers, right?

Defy those barriers. Questioning the authority.

Implement creative persistence, right?

By finding your way past all those obstacles, right?

And think how you can eliminate friction by looking

for those little pebbles that are,

could actually be boulders.

And to find that effort.

And then how do you power yourself

through all these really hard things?

You make it personal, right? You connect to emotions.

So when people hear the Shazam story, they often ask me,

Chris, did you ever wanna give up?

I mean, it was 18 years, 18 years from starting the company

until Apple bought it.

$140 million

of venture capital raised six years of almost going bankrupt.

Did you ever want to give up?

And I think about it, and I think, no,

I never wanted to give up.

I always thought one day we're gonna get there.

I know one day we're gonna get there.

So I'm gonna finish by showing

you one minute of this concert.

And as you watch this concert, I'd like you

to look at the emotions in the audience,

the emotions in the audience,

and think to yourself, how can you use your own emotions

to overcome the obstacles in your life?

So the most rewarding moment of Shazam for me

was not the day that we invented this algorithm

that everyone said was impossible.

And the most rewarding moment of Shazam for me

was not even the day, 18 years later

when Apple acquired Shazam.

When every venture capitalist had said to me, Chris,

Apple just doesn't make big acquisitions.

The most rewarding moment of Shazam for me was actually not

that long ago when I was in this little grocery store in San

Francisco called Buy Right.

And I was waiting in line to buy my groceries.

And suddenly the person in front of me said to the cashier,

they said, Hey, what's that song?

And the cashier said, I don't know, let's Shazam it.

And then these two strangers who didn't even know each other

start Shazaming a song together, inches in front of my face.

And I didn't say anything, but I just thought to myself,

Wow, Thank you very much.

Thank you.

Thank you. Thank you so much.