How to use data to make a hit TV show | Sebastian Wernicke

Prevodilac: Milenka Okuka Lektor: Mile Živković

Roy Price is a man that most of you have probably never heard about,

Roj Prajs je čovek za koga većina vas verovatno nije nikad čula,

even though he may have been responsible

iako je možda odgovoran

for 22 somewhat mediocre minutes of your life on April 19, 2013.

za 22 prilično osrednja minuta vašeg života, 19. aprila 2013.

He may have also been responsible for 22 very entertaining minutes,

Možda je takođe odgovoran za 22 veoma zabavna minuta,

but not very many of you.

ali za mali broj vas.

And all of that goes back to a decision

A sve se svodi na odluku

that Roy had to make about three years ago.

koju je Roj morao da donese pre oko tri godine.

So you see, Roy Price is a senior executive with Amazon Studios.

Dakle, vidite, Roj Prajs je viši producent u Amazon Studios.

That's the TV production company of Amazon.

To je Amazonova TV produkcijska firma.

He's 47 years old, slim, spiky hair,

On ima 47 godina, vitak je, ima jež frizuru,

describes himself on Twitter as "movies, TV, technology, tacos."

sebe opisuje na Tviteru kao nekog ko voli "filmove, TV, tehnologiju i takose".

And Roy Price has a very responsible job, because it's his responsibility

A Roj Prajs ima veoma odgovoran posao jer je njegova odgovornost

to pick the shows, the original content that Amazon is going to make.

da izabere serije, originalne sadržine, koje će Amazon da snima.

And of course that's a highly competitive space.

I naravno, to je izuzetno konkurentan prostor.

I mean, there are so many TV shows already out there,

Mislim, već postoji toliko TV serija

that Roy can't just choose any show.

te Roj ne može prosto da izabere bilo koju.

He has to find shows that are really, really great.

Mora da pronađe serije koje su zaista, zaista sjajne.

So in other words, he has to find shows

Drugim rečima, mora da pronađe serije

that are on the very right end of this curve here.

koje su na samom desnom kraju ove krive ovde.

So this curve here is the rating distribution

Dakle, ova kriva ovde predstavlja raspodelu ocena

of about 2,500 TV shows on the website IMDB,

oko 2,500 TV serija sa vebsajta IMDB,

and the rating goes from one to 10,

a ocene se kreću od jedan do deset,

and the height here shows you how many shows get that rating.

a visina krive vam pokazuje koliko je serija imalo tu ocenu.

So if your show gets a rating of nine points or higher, that's a winner.

Pa, ako vaša serija dobije ocenu od devet poena ili više, to je pobednik.

Then you have a top two percent show.

Onda imate seriju iz prvih dva posto.

That's shows like "Breaking Bad," "Game of Thrones," "The Wire,"

To su serije, poput: "Čiste hemije", "Igre prestola", "Žice",

so all of these shows that are addictive,

dakle, sve te serije koje vam uđu pod kožu,

whereafter you've watched a season, your brain is basically like,

gde nakon što ste odgledali prvu sezonu, vaš mozak je u fazonu:

"Where can I get more of these episodes?"

"Gde da nađem još ovih epizoda?"

That kind of show.

Taj tip serije.

On the left side, just for clarity, here on that end,

Na levoj strani, čisto da bude jasno, ovde na kraju,

you have a show called "Toddlers and Tiaras" --

imate seriju koja se zove "Devojčice i dijademe" -

(Laughter)

(Smeh)

-- which should tell you enough

- to vam dovoljno govori

about what's going on on that end of the curve.

o tome šta se dešava na tom kraju krive.

Now, Roy Price is not worried about getting on the left end of the curve,

E sad, Roja Prajsa ne brine da će završiti na levoj strani krive

because I think you would have to have some serious brainpower

jer mislim da morate da posedujete prilično ozbiljne umne moći

to undercut "Toddlers and Tiaras."

da podrijete "Devojčice i dijademe".

So what he's worried about is this middle bulge here,

Dakle, njega brine ovo središnje ispupčenje ovde,

the bulge of average TV,

tu su prosečne TV serije,

you know, those shows that aren't really good or really bad,

znate, one serije koje nisu naročito dobre, ni loše,

they don't really get you excited.

zbog njih se ne uzbuđujete previše.

So he needs to make sure that he's really on the right end of this.

Pa mora da se postara da zaista bude na desnom kraju ovoga.

So the pressure is on,

Dakle, prisutan je pritisak

and of course it's also the first time

i, naravno, takođe je prvi put

that Amazon is even doing something like this,

da Amazon uopšte radi nešto slično,

so Roy Price does not want to take any chances.

pa Roj Prajs ne želi da rizikuje.

He wants to engineer success.

Želi da isplanira uspeh.

He needs a guaranteed success,

Potreban mu je zagarantovan uspeh,

and so what he does is, he holds a competition.

pa on organizuje takmičenje.

So he takes a bunch of ideas for TV shows,

Uzima gomilu ideja za TV serije

and from those ideas, through an evaluation,

i od tih ideja, kroz procenu,

they select eight candidates for TV shows,

biraju osam kandidata za TV serije,

and then he just makes the first episode of each one of these shows

a potom samo snimaju prvu epizodu svake od ovih serija

and puts them online for free for everyone to watch.

i postavljaju ih besplatno na internet gde ih svako može gledati.

And so when Amazon is giving out free stuff,

A kada Amazon poklanja nešto,

you're going to take it, right?

uzećete to, zar ne?

So millions of viewers are watching those episodes.

Dakle, milioni gledalaca gledaju ove epizode.

What they don't realize is that, while they're watching their shows,

Međutim, ne shvataju da dok gledaju svoje serije,

actually, they are being watched.

zapravo njih gledaju.

They are being watched by Roy Price and his team,

Posmatraju ih Roj Prajs i njegova ekipa,

who record everything.

koji sve snimaju.

They record when somebody presses play, when somebody presses pause,

Snimaju kad neko pritisne start, kad neko pritisne pauzu,

what parts they skip, what parts they watch again.

koje delove preskaču, koje delove iznova gledaju.

So they collect millions of data points,

Tako su sakupili milione jedinica podataka

because they want to have those data points

jer su im potrebne te jedinice podataka

to then decide which show they should make.

kako bi potom odlučili koju će seriju snimati.

And sure enough, so they collect all the data,

I svakako, sakupili su sve podatke,

they do all the data crunching, and an answer emerges,

usitnili su sve podatke i pojavio se odgovor,

and the answer is,

a odgovor je:

"Amazon should do a sitcom about four Republican US Senators."

"Amazon bi trebalo da snimi sitkom o četiri američka republikanska senatora."

They did that show.

Snimili su tu seriju.

So does anyone know the name of the show?

Da li iko zna naziv te serije?

(Audience: "Alpha House.")

(Publika: "Alpha House.")

Yes, "Alpha House,"

Da, "Alpha House",

but it seems like not too many of you here remember that show, actually,

ali čini se da se nekolicina vas ovde zapravo seća te serije

because it didn't turn out that great.

jer nije ispala naročito dobro.

It's actually just an average show,

Zapravo se radi o prosečnoj seriji,

actually -- literally, in fact, because the average of this curve here is at 7.4,

zapravo - bukvalno, uistinu, jer prosečna vrednost na ovoj krivoj je 7,4,

and "Alpha House" lands at 7.5,

a "Alpha House" se nalazi na 7,5,

so a slightly above average show,

dakle, samo malo iznad prosečne serije,

but certainly not what Roy Price and his team were aiming for.

ali svakako to nije ono što su Roj i njegova ekipa želeli.

Meanwhile, however, at about the same time,

U međuvremenu, međutim, otprilike istovremeno,

at another company,

u drugoj firmi,

another executive did manage to land a top show using data analysis,

drugi producent je uspeo da isporuči vrhunsku seriju, obradom podataka,

and his name is Ted,

a on se zove Ted,

Ted Sarandos, who is the Chief Content Officer of Netflix,

Ted Sarantos, on je glavni referent sadržaja na Netfliksu,

and just like Roy, he's on a constant mission

i baš kao i Roj, on je stalno na misiji

to find that great TV show,

pronalaženja sjajne TV serije,

and he uses data as well to do that,

i on koristi podatke da bi to postigao,

except he does it a little bit differently.

samo što to on radi malčice drugačije.

So instead of holding a competition, what he did -- and his team of course --

Pa su umesto organizovanja takmičenja, on - i naravno njegova ekipa -

was they looked at all the data they already had about Netflix viewers,

pogledali sve podatke koje su već imali o gledaocima Netfliksa,

you know, the ratings they give their shows,

znate, ocene koje daju serijama,

the viewing histories, what shows people like, and so on.

istoriju gledanja, koje serije ljudi vole i tako dalje.

A onda su koristili te podatke da otkriju

And then they use that data to discover

all of these little bits and pieces about the audience:

sve te sitnice i pojedinosti o publici:

what kinds of shows they like,

koje serije vole,

what kind of producers, what kind of actors.

koje producente, koje glumce.

And once they had all of these pieces together,

I kada su sklopili sve te komadiće,

they took a leap of faith,

otisnuli su se u nepoznato

and they decided to license

i odlučili da odobre,

not a sitcom about four Senators

ne sitkom o četvorici senatora,

but a drama series about a single Senator.

već dramu o jednom senatoru.

You guys know the show?

Znate li, ljudi, tu seriju?

(Laughter)

(Smeh)

Yes, "House of Cards," and Netflix of course, nailed it with that show,

Da, "Kuća od karata" i Netfliks je naravno trijumfovao tom serijom,

at least for the first two seasons.

bar tokom prve dve sezone.

(Laughter) (Applause)

(Smeh) (Aplauz)

"House of Cards" gets a 9.1 rating on this curve,

"Kuća od karata" ima ocenu od 9,1 na ovoj krivoj,

so it's exactly where they wanted it to be.

dakle, tu su i želeli da stignu.

Now, the question of course is, what happened here?

E sad, pitanje je naravno, šta se ovde desilo?

So you have two very competitive, data-savvy companies.

Imate dve veoma konkurentne firme, spretne s podacima.

They connect all of these millions of data points,

One spajaju sve ove milione jedinica podataka,

and then it works beautifully for one of them,

a onda se za jednu sve završi lepo,

and it doesn't work for the other one.

a za drugu ne.

So why?

Zašto?

Because logic kind of tells you that this should be working all the time.

Jer, logika vam govori da bi ovo trebalo stalno da funkcioniše.

I mean, if you're collecting millions of data points

Mislim, ako sakupljate milione jedinica podataka

on a decision you're going to make,

za odluku koju donosite,

then you should be able to make a pretty good decision.

onda bi trebalo da ste u stanju da donesete valjanu odluku.

You have 200 years of statistics to rely on.

Imate da se oslonite na 200-godišnju statistiku.

You're amplifying it with very powerful computers.

Nju ste pojačali izuzetno moćnim kompjuterima.

The least you could expect is good TV, right?

Najmanje što biste očekivali je dobra televizija, zar ne?

And if data analysis does not work that way,

A, ako obrada podataka ne funkcioniše tako,

then it actually gets a little scary,

onda zapravo postaje malčice zastrašujuće

because we live in a time where we're turning to data more and more

jer živimo u vremenu u kom se sve više okrećemo podacima

to make very serious decisions that go far beyond TV.

da bismo doneli veoma ozbiljne odluke koje sežu mimo televizije.

Does anyone here know the company Multi-Health Systems?

Da li iko ovde zna za firmu Multi-Health Systems?

No one. OK, that's good actually.

Niko. U redu, to je zapravo dobro.

OK, so Multi-Health Systems is a software company,

U redu, Multi-Health Systems je softverska firma

and I hope that nobody here in this room

i nadam se da niko iz ove prostorije

ever comes into contact with that software,

nikada neće doći u dodir s tim softverom

jer ako dođete, to će značiti da ste u zatvoru.

because if you do, it means you're in prison.

(Laughter)

(Smeh)

If someone here in the US is in prison, and they apply for parole,

Ako je neko, ovde u SAD-u, u zatvoru i prijavi se za uslovnu,

then it's very likely that data analysis software from that company

onda će verovatno softver te firme za obradu podataka

will be used in determining whether to grant that parole.

da koriste kako bi utvrdili da li da vam odobre uslovnu.

So it's the same principle as Amazon and Netflix,

Dakle, isti je princip kao kod Amazona i Nefliksa,

but now instead of deciding whether a TV show is going to be good or bad,

ali sad umesto odlučivanja o tome da li će serija da bude dobra ili loša,

you're deciding whether a person is going to be good or bad.

odlučuje se da li će osoba da bude dobra ili loša.

And mediocre TV, 22 minutes, that can be pretty bad,

A prosečna televizijska 22 minuta mogu da budu vrlo loša,

but more years in prison, I guess, even worse.

ali dodatne godine u zatvoru su, valjda, još gore.

And unfortunately, there is actually some evidence that this data analysis,

I nažalost, zapravo imamo neke dokaze da ova obrada podataka,

despite having lots of data, does not always produce optimum results.

uprkos velikom broju podataka, ne daje uvek optimalne rezultate.

And that's not because a company like Multi-Health Systems

A to nije zato što firma, poput Multi-Health Systems

doesn't know what to do with data.

ne zna šta da radi s podacima.

Even the most data-savvy companies get it wrong.

Čak i firme koje su stručnjaci za podatke, greše.

Yes, even Google gets it wrong sometimes.

Da, čak i Gugl ponekad pogreši.

In 2009, Google announced that they were able, with data analysis,

Godine 2009, Gugl je najavio da su u stanju, koristeći obradu podataka,

to predict outbreaks of influenza, the nasty kind of flu,

da predvide epidemiju influence, gadnog oblika gripa,

by doing data analysis on their Google searches.

koristeći obradu podataka s Guglovih pretraga.

And it worked beautifully, and it made a big splash in the news,

I savršeno je funkcionisalo i dospeli su u glavne vesti,

including the pinnacle of scientific success:

uključujući i vrhunac naučnog uspeha:

a publication in the journal "Nature."

objavu u magazinu "Nejčer".

It worked beautifully for year after year after year,

Savršeno je funkcionisalo godinu za godinom,

until one year it failed.

sve dok jedne godine nije zatajilo.

And nobody could even tell exactly why.

A niko nije čak mogao da tačno kaže zašto.

It just didn't work that year,

Prosto nije funkcionisalo te godine

i, naravno, to je ponovo bila glavna vest,

and of course that again made big news,

including now a retraction

uključujući sada i povlačenje

of a publication from the journal "Nature."

članka iz časopisa "Nejčer".

So even the most data-savvy companies, Amazon and Google,

Dakle, čak i stručnjaci za podatke, Amazon i Gugl,

they sometimes get it wrong.

ponekad pogreše.

And despite all those failures,

I uprkos svim tim neuspesima,

data is moving rapidly into real-life decision-making --

podaci sve brže postaju deo odlučivanja u stvarnom životu -

into the workplace,

na poslu,

law enforcement,

u policiji,

medicine.

medicini.

So we should better make sure that data is helping.

Dakle, trebalo bi da se postaramo da nam podaci budu korisni.

Now, personally I've seen a lot of this struggle with data myself,

Sad, lično sam video mnogo ove borbe s podacima

because I work in computational genetics,

jer radim na polju računarske genetike,

which is also a field where lots of very smart people

što je takođe oblast gde mnogo veoma pametnih ljudi

are using unimaginable amounts of data to make pretty serious decisions

koristi nezamislivu količinu podataka da bi doneli veoma ozbiljne odluke,

like deciding on a cancer therapy or developing a drug.

poput odlučivanja o terapiji za rak ili o razvoju leka.

And over the years, I've noticed a sort of pattern

I vremenom sam primetio nešto nalik obrascu

or kind of rule, if you will, about the difference

ili nekakvom pravilu, ako hoćete, o razlici

between successful decision-making with data

između uspešnog odlučivanja pomoću podataka

and unsuccessful decision-making,

i neuspešnog odlučivanja

and I find this a pattern worth sharing, and it goes something like this.

i smatram da ovaj obrazac vredi deliti, a radi se o sledećem.

So whenever you're solving a complex problem,

Dakle, kad god rešavate složen problem,

you're doing essentially two things.

u suštini radite dve stvari.

The first one is, you take that problem apart into its bits and pieces

Prvo razlažete dati problem na sastavne komadiće i delove

so that you can deeply analyze those bits and pieces,

kako biste podrobno analizirali te komadiće i delove

and then of course you do the second part.

i potom, naravno, prelazite na drugi deo.

You put all of these bits and pieces back together again

Sastavljate ponovo sve ove komadiće i delove

to come to your conclusion.

da biste došli do zaključka.

And sometimes you have to do it over again,

A ponekad to morate da uradite više puta,

but it's always those two things:

ali uvek se radi o ove dve stvari:

taking apart and putting back together again.

rastavljanju i ponovnom sastavljanju.

And now the crucial thing is

E sad, ključno je

that data and data analysis

da su podaci i obrada podataka

is only good for the first part.

jedino korisni u prvom delu.

Data and data analysis, no matter how powerful,

Podaci i obrada podataka, ma koliko moćni bili,

can only help you taking a problem apart and understanding its pieces.

jedino vam mogu pomoći da razložite problem i da razumete njegove delove.

It's not suited to put those pieces back together again

Neodgovarajući su za ponovno sastavljanje tih delova

and then to come to a conclusion.

i za dolaženje do zaključka.

There's another tool that can do that, and we all have it,

Postoji drugo oruđe koje može to da uradi i svi ga imamo,

and that tool is the brain.

a to oruđe je mozak.

If there's one thing a brain is good at,

Ako je mozak u nečemu dobar,

it's taking bits and pieces back together again,

to je ponovno sastavljanje komadića i delova,

even when you have incomplete information,

čak i kad imate nepotpunu informaciju,

and coming to a good conclusion,

i dolaženje do dobrog zaljučka,

especially if it's the brain of an expert.

naročito ako se radi o mozgu stručnjaka.

And that's why I believe that Netflix was so successful,

I zato verujem da je Netfliks bio uspešan

because they used data and brains where they belong in the process.

jer su koristili podatke i mozak tamo gde im je i mesto u procesu.

They use data to first understand lots of pieces about their audience

Koristili su podatke kako bi prvobitno shvatili gomile stvari o svojoj publici

that they otherwise wouldn't have been able to understand at that depth,

koje u suprotnom ne bi bili u stanju da razumeju tako podrobno,

but then the decision to take all these bits and pieces

ali potom je odluka da uzmu sve te komadiće i delove

and put them back together again and make a show like "House of Cards,"

i da ih ponovo sastave i naprave seriju, poput "Kuće od karata",

that was nowhere in the data.

toga nije bilo u podacima.

Ted Sarandos and his team made that decision to license that show,

Ted Sarandos i njegova ekipa su odlučili da odobre tu seriju,

which also meant, by the way, that they were taking

što je takođe značilo, usput, da su preuzimali

a pretty big personal risk with that decision.

prilično veliki lični rizik tom odlukom.

And Amazon, on the other hand, they did it the wrong way around.

A Amazon, s druge strane, oni su pogrešili u postupku.

They used data all the way to drive their decision-making,

Vodili su se podacima sve vreme donošenja odluke,

first when they held their competition of TV ideas,

prvo kada su organizovali takmičenje u TV idejama,

then when they selected "Alpha House" to make as a show.

potom kada su odabrali da snimaju seriju "Alpha House".

Which of course was a very safe decision for them,

Što je naravno bila veoma bezbedna odluka za njih

because they could always point at the data, saying,

jer su uvek mogli da pokažu na podatke i kažu:

"This is what the data tells us."

"Tako su nam podaci rekli."

But it didn't lead to the exceptional results that they were hoping for.

Međutim to nije dovelo do izvanrednih rezultata kojima su se nadali.

So data is of course a massively useful tool to make better decisions,

Dakle, podaci su svakako izuzetno korisno oruđe da bolje odlučujete,

but I believe that things go wrong

ali verujem da stvari kreću po zlu

when data is starting to drive those decisions.

kada su podaci vodeći u odlučivanju.

No matter how powerful, data is just a tool,

Ma koliko moćni, podaci su samo oruđe,

and to keep that in mind, I find this device here quite useful.

a da biste imali to na umu, smatram da je ovo sredstvo ovde korisno.

Many of you will ...

Mnogi od vas će...

(Laughter)

(Smeh)

Before there was data,

Pre podataka,

this was the decision-making device to use.

ovo je bio uređaj koji ste koristili u odlučivanju.

(Laughter)

(Smeh)

Many of you will know this.

Mnogima je ovo poznato.

This toy here is called the Magic 8 Ball,

Ova igračka se zove magična osmica

and it's really amazing,

i zaista je izvanredna

because if you have a decision to make, a yes or no question,

jer ako treba da odlučite o nečemu, o pitanju sa da ili ne,

all you have to do is you shake the ball, and then you get an answer --

sve što je potrebno je da protresete kuglu i potom dobijate odgovor -

"Most Likely" -- right here in this window in real time.

"najverovatnije" - baš tu u ovom prorezu, u realnom vremenu.

I'll have it out later for tech demos.

Kasnije ću je podvrći tehničkim probama.

(Laughter)

(Smeh)

Now, the thing is, of course -- so I've made some decisions in my life

Sad, radi se, naravno - doneo sam neke odluke u svom životu

where, in hindsight, I should have just listened to the ball.

za koje se kasnije ispostavilo da je trebalo da poslušam kuglu.

But, you know, of course, if you have the data available,

Međutim, znate, naravno, ako su vam podaci dostupni,

you want to replace this with something much more sophisticated,

želećete da zamenite ovo nečim daleko prefinjenijim,

like data analysis to come to a better decision.

poput obrade podataka, da biste doneli bolju odluku.

But that does not change the basic setup.

Ali to ne menja osnovnu postavku.

So the ball may get smarter and smarter and smarter,

Dakle, kugla može da postaje sve pametnija i pametnija,

but I believe it's still on us to make the decisions

ali verujem da je odlučivanje i dalje na nama,

if we want to achieve something extraordinary,

ako želimo da postignemo nešto izuzetno

on the right end of the curve.

na desnom kraju ove krive.

And I find that a very encouraging message, in fact,

I za mene je to zapravo veoma ohrabrujuća poruka,

that even in the face of huge amounts of data,

da čak i kad ste suočeni s ogromnom količinom podataka,

it still pays off to make decisions,

i dalje se isplati odlučivati,

to be an expert in what you're doing

biti stručnjak u onome što radite

and take risks.

i preuzimati rizike.

Because in the end, it's not data,

Jer, naposletku, neće vas podaci

it's risks that will land you on the right end of the curve.

već rizici smestiti na desni kraj ove krive.

Thank you.

Hvala vam.

(Applause)

(Aplauz)