0:00:01.097,0:00:02.841
Part two of Culture and[br]Team, and

0:00:02.841,0:00:05.708
we have Ben Silverman, the[br]founder of Pinterest, and

0:00:05.708,0:00:08.530
John and Patrick Collison,[br]the founders of Stripe.

0:00:10.510,0:00:13.300
Founders that have obviously[br]sort of, some of the best in

0:00:13.300,0:00:15.190
the world at thinking about[br]culture and

0:00:15.190,0:00:15.940
how they build teams.

0:00:15.940,0:00:18.400
So, there's three areas that[br]we're gonna cover today.

0:00:18.400,0:00:20.690
One will just be sort of[br]general thoughts on

0:00:20.690,0:00:22.980
culture as a follow up to[br]the last lecture.

0:00:22.980,0:00:25.580
And then we're really gonna[br]dig into what happens at

0:00:25.580,0:00:27.120
the founding of these[br]companies.

0:00:27.120,0:00:28.790
And building out the early[br]team.

0:00:28.790,0:00:32.530
And then how that changes[br]and evolves as these

0:00:32.530,0:00:35.940
guys have scaled their[br]companies up to 100 plus.

0:00:35.940,0:00:37.270
I don't even know how many[br]people you have now,

0:00:37.270,0:00:38.380
but quite a lot.

0:00:38.380,0:00:39.520
Very large organizations and

0:00:39.520,0:00:41.240
how you adapt these[br]principles of culture.

0:00:42.910,0:00:46.205
But to start off I just[br]wanna ask a very

0:00:46.205,0:00:47.970
open-ended question which[br]is,

0:00:47.970,0:00:51.010
what are the core pieces of[br]culture that you found to

0:00:51.010,0:00:52.951
be most important in[br]building out your companies?

0:00:52.951,0:00:56.720
>> So what are the most[br]important parts?

0:00:58.130,0:01:01.829
>> It's on.[br]>> Oh, it's on.

0:01:01.829,0:01:02.769
Yeah I mean I think for

0:01:02.769,0:01:04.910
us, like we think about on a[br]few dimensions.

0:01:04.910,0:01:05.780
One is like who do we hire,

0:01:05.780,0:01:07.180
and what do those people[br]value?

0:01:08.710,0:01:11.950
Two is what do we do every[br]day, like why do we do it?

0:01:13.180,0:01:15.460
Three is what do we choose[br]to communicate?

0:01:15.460,0:01:17.480
And then I think the fourth[br]is what we

0:01:17.480,0:01:18.460
choose to celebrate.

0:01:18.460,0:01:19.220
I guess the converse of

0:01:19.220,0:01:20.590
that is like what you choose[br]to punish.

0:01:20.590,0:01:23.880
But in general I think[br]running a company based on

0:01:23.880,0:01:24.490
what you celebrate,

0:01:24.490,0:01:26.360
is more exciting than what[br]you punish.

0:01:26.360,0:01:29.040
But I think those four[br]things kind of make up

0:01:29.040,0:01:30.040
the bulk of it for us.

0:01:35.796,0:01:38.666
>> We've placed a large[br]emphasis on, as Stripe has

0:01:38.666,0:01:41.536
grown and probably more than[br]other companies is,

0:01:41.536,0:01:43.560
transparency internally.

0:01:43.560,0:01:44.940
And I think it's been[br]something that's been

0:01:44.940,0:01:45.505
really valuable for

0:01:45.505,0:01:48.460
Stripe, and also a little[br]bit misunderstood.

0:01:49.590,0:01:53.740
All the things people talk[br]about like hiring really

0:01:53.740,0:01:57.070
great people, or giving them[br]a huge amount of leverage.

0:01:57.070,0:02:00.220
Transparency for us plays[br]into that.

0:02:00.220,0:02:03.335
We think that, if you are[br]aligned at a high

0:02:03.335,0:02:07.340
level about what Stripe is[br]doing, if everyone really

0:02:07.340,0:02:11.078
believes in the mission, and[br]then if everyone has

0:02:11.078,0:02:14.639
really good access to[br]information, and kind of

0:02:14.639,0:02:18.401
has a good picture of the[br]current state of Stripe.

0:02:18.401,0:02:21.175
Then that gets you a huge[br]amount of the way there in

0:02:21.175,0:02:24.230
terms of working[br]productively together.

0:02:24.230,0:02:26.760
And it kind of forgives a[br]lot of the other things that

0:02:26.760,0:02:29.790
tend to break as you, as you[br]grow a start-up.

0:02:29.790,0:02:33.090
As we've grown, we started[br]off two people.

0:02:33.090,0:02:33.980
We're now over 170 people.

0:02:33.980,0:02:37.160
We've put a lot of thought[br]into the tooling that

0:02:37.160,0:02:41.170
goes around transparency,[br]because at 170 people,

0:02:41.170,0:02:44.100
there is so much information[br]being produced,

0:02:44.100,0:02:46.420
that you can't just consume[br]it all as a fire hose.

0:02:46.420,0:02:47.300
And so

0:02:47.300,0:02:49.820
how we use slack, how we use[br]email, things like that.

0:02:49.820,0:02:51.030
We can go into it more[br]later.

0:02:51.030,0:02:52.420
But I think that's one of

0:02:52.420,0:02:54.060
the core things that's[br]helped us work well.

0:02:55.300,0:02:58.180
>> I think culture to some[br]degree is basically kind of

0:02:58.180,0:03:00.790
the resolution to a[br]bandwidth problem.

0:03:00.790,0:03:04.480
In the sense that, maybe[br]when you start out working

0:03:04.480,0:03:06.270
on something, you're sort of[br]coding all the time, but you

0:03:06.270,0:03:08.580
can't code all the things[br]that you think the product

0:03:08.580,0:03:10.440
might need, or the company[br]might need, or whatever, and

0:03:10.440,0:03:13.150
you so you decide to work[br]with more coders, right?

0:03:15.420,0:03:17.870
And so the organization gets[br]larger.

0:03:17.870,0:03:19.590
And maybe, in some idealized[br]world,

0:03:19.590,0:03:21.750
I don't think this actually[br]true, but kind of ideally

0:03:21.750,0:03:23.950
you could be involved in[br]every single decision, and

0:03:23.950,0:03:26.370
every single sort of moment[br]of the company, and

0:03:26.370,0:03:28.820
everything that happens, but[br]obviously you can't, or

0:03:28.820,0:03:30.260
maybe you can if two people.

0:03:30.260,0:03:32.440
But you certainly can't at[br]even like five or

0:03:32.440,0:03:34.530
ten kind of that point comes[br]very quickly,

0:03:34.530,0:03:36.710
then by the time you're 50[br]it's completely hopeless.

0:03:36.710,0:03:39.570
And so culture is kind of[br]how you kind of,

0:03:39.570,0:03:44.110
what the strands are that[br]you sort of want to have,

0:03:44.110,0:03:46.630
the invariance that you want[br]to kind of maintain,

0:03:47.670,0:03:50.840
as you can get specifically[br]involved in sort of

0:03:50.840,0:03:53.338
fewer and fewer decisions[br]over time.

0:03:53.338,0:03:56.192
And I think when you think[br]about it that way it,

0:03:56.192,0:03:59.680
maybe its kind of importance[br]becomes sort of

0:03:59.680,0:04:00.610
self evident, right?

0:04:00.610,0:04:03.030
Because again, like the[br]fraction of things you

0:04:03.030,0:04:07.240
can be involved in directly[br]is diminishing, I mean,

0:04:07.240,0:04:08.130
almost exponentially,

0:04:08.130,0:04:10.430
sort of assuming your head[br]count growth sort of is

0:04:10.430,0:04:14.290
on a curve that looks like[br]one of the great companies.

0:04:14.290,0:04:17.700
And yeah that's super[br]important.

0:04:17.700,0:04:20.648
And again, it manifests[br]itself in a bunch of

0:04:20.648,0:04:21.850
different ways.

0:04:21.850,0:04:25.680
Like for example, in hiring,[br]I think a large part of

0:04:25.680,0:04:28.810
the reason why maybe the[br]first ten people you hire,

0:04:28.810,0:04:29.920
what kind of goes to ship,

0:04:29.920,0:04:32.760
decisions are so important[br]is because you're not just

0:04:32.760,0:04:34.320
hiring those first ten[br]people.

0:04:34.320,0:04:36.330
You're actually kind of[br]hiring 100 people.

0:04:36.330,0:04:37.700
Because you should think of[br]kind of

0:04:37.700,0:04:39.790
each one of those people as[br]bringing along sort of

0:04:39.790,0:04:41.580
another ten people with[br]them, and

0:04:41.580,0:04:45.830
sort of figuring out exactly[br]what 90 people, you would

0:04:45.830,0:04:48.070
like those first ten people[br]to bring along is obviously

0:04:49.340,0:04:53.360
gonna be quite consequential[br]for your company.

0:04:53.360,0:04:55.940
But really briefly I think[br]it's largely about sort of

0:04:57.520,0:04:58.020
abstraction.

0:04:59.280,0:05:00.780
>> So one thing that a lot[br]of speakers in this

0:05:00.780,0:05:03.910
class have touched on is how[br]hiring those first ten

0:05:03.910,0:05:05.670
employees, if you don't get[br]that right,

0:05:05.670,0:05:07.978
the company basically will[br]never recover but

0:05:07.978,0:05:10.180
no one has talked about how[br]to do that so.

0:05:11.180,0:05:14.000
What have the three of you[br]looked for

0:05:15.010,0:05:17.190
when you've hired these[br]initial employees,

0:05:17.190,0:05:19.800
to get the culture of the[br]company right?

0:05:19.800,0:05:20.835
How, how have you found[br]them, and

0:05:20.835,0:05:21.630
what have you looked for?

0:05:24.080,0:05:24.610
>> Sure.[br]So,

0:05:24.610,0:05:28.542
I guess this answer is[br]different for every

0:05:28.542,0:05:32.640
company and I'll say for us[br]it was very inductive.

0:05:32.640,0:05:35.390
So I literally looked for[br]people that I wanted to work

0:05:35.390,0:05:36.960
with and that I thought were[br]talented.

0:05:38.230,0:05:39.960
I think, I've read all these[br]books about culture,

0:05:39.960,0:05:41.130
because when I don't know[br]how to do something,

0:05:41.130,0:05:42.730
I first go read things and

0:05:42.730,0:05:43.920
everyone has all these[br]frameworks.

0:05:43.920,0:05:47.090
And I think one bit, big[br]misconception that someone

0:05:47.090,0:05:49.270
said once is that people[br]think culture is like

0:05:49.270,0:05:51.480
architecture when it's a lot[br]more like gardening.

0:05:51.480,0:05:52.650
You know, you plant some[br]seeds and

0:05:52.650,0:05:54.050
then you pull out weeds that[br]aren't working,

0:05:54.050,0:05:55.570
and they sort of expand.

0:05:55.570,0:05:58.650
So, when we first hired[br]people we hired people that

0:05:58.650,0:05:59.520
were like ourselves.

0:06:01.060,0:06:02.030
I often looked at like three[br]or

0:06:02.030,0:06:04.460
four different things that I[br]really valued in people.

0:06:04.460,0:06:06.700
You know, I looked for[br]people that worked hard and

0:06:06.700,0:06:08.890
seemed high integrity and[br]low ego.

0:06:08.890,0:06:10.490
I looked for people that[br]were creative, and

0:06:10.490,0:06:12.170
I usually meant that they[br]were really curious.

0:06:12.170,0:06:14.430
They had all these different[br]interests.

0:06:14.430,0:06:17.120
Some of our first employees[br]are probably some of

0:06:17.120,0:06:19.280
the quirkiest people I've[br]ever met.

0:06:19.280,0:06:20.250
They were engineers but

0:06:20.250,0:06:22.540
they also have all these[br]crazy hobbies.

0:06:22.540,0:06:24.390
Like one guy made his own[br]board game

0:06:24.390,0:06:26.240
with this elaborate set of[br]rules.

0:06:26.240,0:06:28.240
Another guy was really into[br]magic tricks.

0:06:28.240,0:06:31.210
And he had coded not only[br]like this magic trick on

0:06:31.210,0:06:32.000
the iPhone, but

0:06:32.000,0:06:35.290
he had shot the production[br]video with a preview.

0:06:35.290,0:06:37.400
And I think that, that[br]quirkiness has actually been

0:06:37.400,0:06:38.390
a little bit of a calling[br]card.

0:06:38.390,0:06:40.770
And we find that really[br]creative, quirky people that

0:06:40.770,0:06:42.710
are excited about many[br]disciplines, and are

0:06:42.710,0:06:46.720
extraordinary at one tend to[br]build really great products.

0:06:46.720,0:06:49.220
They tend to be great at[br]collaborating.

0:06:49.220,0:06:50.610
Then the last thing is, we[br]really look for

0:06:50.610,0:06:51.390
people that wanted to,

0:06:51.390,0:06:54.180
they just wanted to build[br]something great, and

0:06:54.180,0:06:56.720
they weren't arrogant about[br]it, but they just felt like,

0:06:56.720,0:06:58.190
it'd be really cool to take[br]a risk and

0:06:58.190,0:07:00.530
build something bigger than[br]themselves.

0:07:01.670,0:07:02.790
And that, in the beginning,[br]is very,

0:07:02.790,0:07:05.250
very easy to select for if[br]you're in our situation.

0:07:05.250,0:07:08.260
We had this horrible office,[br]like, nobody got paid.

0:07:08.260,0:07:10.530
So there was no external[br]reason, other than being

0:07:10.530,0:07:12.810
excited about building[br]something to join.

0:07:12.810,0:07:14.770
In fact, there was every[br]reason not to.

0:07:14.770,0:07:16.410
And that's something,[br]looking back,

0:07:16.410,0:07:17.170
I really, really value,

0:07:17.170,0:07:19.080
because we always knew[br]people were joining for

0:07:19.080,0:07:20.275
the purest reasons, and in[br]fact,

0:07:20.275,0:07:23.680
were willing to forgo other[br]great job opportunities,

0:07:23.680,0:07:27.190
market salary, a clean[br]office,

0:07:27.190,0:07:29.540
good equipment just for the[br]chance to work.

0:07:29.540,0:07:32.540
So, to this day I think a[br]lot of those traits have

0:07:32.540,0:07:33.230
been seeded and

0:07:33.230,0:07:35.050
are embedded in the folks[br]that we look at now.

0:07:37.990,0:07:39.400
>> Yeah the first ten hires[br]are really

0:07:39.400,0:07:43.950
hard because you're making[br]these first ten

0:07:43.950,0:07:46.500
hires at a point where no[br]one's heard of this company,

0:07:46.500,0:07:47.550
no one really wants to work[br]for it.

0:07:47.550,0:07:48.530
You're just these,

0:07:48.530,0:07:51.140
like two weird people[br]working on this weird idea.

0:07:51.140,0:07:53.070
>> And like their friends[br]are telling them not

0:07:53.070,0:07:54.190
to join.

0:07:54.190,0:07:57.320
For our second employee, I[br]think

0:07:57.320,0:07:59.800
he'd accepted the offer or[br]he was just about to, and

0:07:59.800,0:08:02.700
his best friends took him[br]out the night before, and it

0:08:02.700,0:08:06.596
was like a full on assault[br]for, why you should not join

0:08:06.596,0:08:09.250
this company, why this is[br]ruining your life basically.

0:08:10.560,0:08:13.720
And so anyway the guy[br]subsequently

0:08:13.720,0:08:14.750
continued to join.

0:08:14.750,0:08:15.598
And actually one of

0:08:15.598,0:08:17.771
those friends also now works[br]at Stripe.

0:08:17.771,0:08:20.067
But this is what you're up[br]against.

0:08:20.067,0:08:20.934
>> Yeah.

0:08:20.934,0:08:24.362
And I mean it's also hard[br]because no batch of

0:08:24.362,0:08:27.790
ten people will have as[br]great an influence on

0:08:27.790,0:08:31.630
the company as those first[br]ten people.

0:08:31.630,0:08:34.640
I think everyone's[br]impression of recruiting is,

0:08:34.640,0:08:36.429
you open LinkedIn.

0:08:36.429,0:08:38.049
It's sort of like ordering[br]off the dollar menu.

0:08:38.049,0:08:40.188
It's, I want that one, that[br]one, and that one.

0:08:40.190,0:08:42.543
And, and now you have some[br]hires.

0:08:42.543,0:08:46.606
Whereas at least for us, it[br]was very much over a very

0:08:46.606,0:08:50.103
long time period, talking[br]people we knew or

0:08:50.103,0:08:53.780
friends of friends into[br]joining.

0:08:53.780,0:08:56.880
We didn't have huge[br]networks, Pat and

0:08:56.880,0:08:58.100
I were both in college at[br]the time.

0:08:58.100,0:08:59.630
So there were no people that

0:08:59.630,0:09:01.750
we'd really worked with to[br]draw on, and so

0:09:01.750,0:09:05.280
a lot of those early Stripes[br]were people we had heard of,

0:09:05.280,0:09:08.990
friends of friends, and the[br]other interesting thing they

0:09:08.990,0:09:11.665
all had in common is that[br]they were all

0:09:11.665,0:09:16.400
early in their career, or[br]undervalued in some way.

0:09:16.400,0:09:17.290
Cuz when you think about it,

0:09:17.290,0:09:20.430
if someone is a known[br]spectacular quantity,

0:09:20.430,0:09:22.570
then they're probably[br]working in a job and

0:09:22.570,0:09:23.730
very happy with that.

0:09:23.730,0:09:27.720
And so we have to try and[br]find people who were,

0:09:27.720,0:09:30.860
in the case of our designer[br]that we hired, he was 18 and

0:09:30.860,0:09:33.970
in high school and in Sweden[br]at the time.

0:09:33.970,0:09:34.820
In the case of our,

0:09:34.820,0:09:39.040
our CTO, he was in college[br]at the time.

0:09:39.040,0:09:40.030
You know, a lot of these[br]people,

0:09:40.030,0:09:42.820
they were early on in their[br]careers and

0:09:42.820,0:09:46.920
the only way we could, you[br]can relax when constrained.

0:09:48.020,0:09:49.940
You can relax the fact that[br]they're talented or

0:09:49.940,0:09:52.212
relax that it's apparent[br]that they're talented,

0:09:52.212,0:09:56.240
and we, not consciously, but[br]we relaxed the latter.

0:09:57.420,0:10:03.320
>> Yeah, I think finding[br]kind of people who are, or

0:10:03.320,0:10:05.130
just think like a value[br]investor.

0:10:05.130,0:10:08.170
You're looking for the human[br]capital that's

0:10:09.990,0:10:11.900
significantly devalued by[br]the market, you know?

0:10:11.900,0:10:12.940
You probably shouldn't look[br]to

0:10:12.940,0:10:15.850
hire your brilliant friends[br]at Facebook and Google or

0:10:15.850,0:10:18.050
whatever, because they're[br]already discovered.

0:10:18.050,0:10:19.510
You know, if they're wanting[br]to join that's great,

0:10:19.510,0:10:21.919
but they're probably harder[br]to convince.

0:10:23.492,0:10:25.570
John and I spent a little a[br]while yesterday

0:10:25.570,0:10:28.790
afternoon sort of trying to[br]figure out in retrospect,

0:10:28.790,0:10:31.490
what kind of traits our[br]first ten or

0:10:31.490,0:10:34.320
so people had in common that[br]we thought were significant.

0:10:34.320,0:10:37.711
And you know, in general[br]sort of in speaking about

0:10:37.711,0:10:41.182
culture, I sort of want to[br]caveat everything we say

0:10:41.182,0:10:45.785
with, I that sort of advice[br]is very limited experience,

0:10:45.785,0:10:47.785
widely over extrapolated.

0:10:47.785,0:10:50.690
And I think there's a lot of[br]truth to that.

0:10:50.690,0:10:54.130
But for our particular first[br]ten people, the things we

0:10:54.130,0:10:58.350
sort of figured out that[br]seemed to be important were

0:10:58.350,0:11:01.760
they were also very genuine[br]and straight.

0:11:01.760,0:11:03.900
I think that actually[br]matters quite a lot in that

0:11:03.900,0:11:06.880
sort of they're people that[br]sort of that others want to

0:11:06.880,0:11:10.120
work with, that they're[br]people that others trust.

0:11:10.120,0:11:11.810
They sort of have an[br]intellectual honesty in

0:11:11.810,0:11:13.630
how they approach problems[br]and, and so forth.

0:11:14.960,0:11:15.650
They were people who

0:11:15.650,0:11:18.490
really liked getting things[br]finished.

0:11:18.490,0:11:19.290
There's a lot of people who

0:11:19.290,0:11:21.400
are really excited about[br]tons of things.

0:11:21.400,0:11:23.630
Only a subset of those are[br]actually excited about like

0:11:23.630,0:11:24.970
completing things.

0:11:24.970,0:11:26.160
You know, there's a lot of[br]talk about, for

0:11:26.160,0:11:28.940
example hiring people off[br]their GitHub resumes,

0:11:28.940,0:11:31.140
whatever, actually think[br]that doesn't quite ring,

0:11:31.140,0:11:32.620
kind of correct to me.

0:11:32.620,0:11:35.520
In the sense that place is a[br]large premium on sort of

0:11:35.520,0:11:36.680
lots of different things.

0:11:36.680,0:11:38.210
I think it's actually a[br]priori,

0:11:38.210,0:11:40.630
sort of much more[br]interesting to work with

0:11:40.630,0:11:43.750
someone who has spent two[br]years really investing in

0:11:43.750,0:11:44.882
going deep in a particular[br]area.

0:11:44.882,0:11:49.670
And then the third trait[br]they all seem to have in

0:11:49.670,0:11:52.460
common is they just sort of[br]cared a great deal.

0:11:52.460,0:11:54.300
Like, it was offensive to[br]them when something was

0:11:54.300,0:11:55.670
just a little bit off.

0:11:55.670,0:11:56.670
And kind of again,

0:11:56.670,0:11:58.590
in hindsight, there are all[br]these like crazy things we

0:11:58.590,0:12:03.010
used to do that I mean, do[br]in fact seem crazy like we

0:12:03.010,0:12:04.470
probably shouldn't have done[br]them,

0:12:04.470,0:12:08.620
but everyone was always like[br]well, was borderline insane

0:12:08.620,0:12:10.740
in sort of how much they[br]cared about tiny details.

0:12:10.740,0:12:16.340
Like we used to, like every[br]single API request that

0:12:16.340,0:12:19.540
ever generated an error went[br]to all of our inboxes and

0:12:19.540,0:12:20.760
phoned all of us.

0:12:20.760,0:12:23.830
Because it seems terrible to[br]ever have an error that

0:12:25.120,0:12:26.120
didn't go and

0:12:26.120,0:12:28.650
get a resolution from the[br]user standpoint.

0:12:28.650,0:12:31.440
Or we used to like copy[br]everyone else on

0:12:31.440,0:12:32.850
every outgoing email.

0:12:32.850,0:12:35.030
And we'd like point out[br]slight grammar or

0:12:35.030,0:12:37.070
spelling mistakes to each[br]other, because it's terrible

0:12:37.070,0:12:38.910
to ever send an email with a[br]spelling mistake.

0:12:38.910,0:12:41.890
So anyway those were the[br]three traits come

0:12:41.890,0:12:42.660
from that area.

0:12:42.660,0:12:46.466
A genuine, caring a great[br]deal.

0:12:46.466,0:12:51.471
And, so what was my second[br]one?

0:12:51.471,0:12:52.105
>> The other one.

0:12:52.105,0:12:55.442
>> Yeah so.

0:12:55.442,0:12:58.903
Sorry yes completing things[br]like list of three items.

0:13:00.990,0:13:02.730
>> Yeah, I'm only gonna say[br]I just don't think

0:13:02.730,0:13:06.410
there's any wrong place to[br]find people, so when I

0:13:06.410,0:13:09.080
look back at our first few[br]folks that we hired.

0:13:09.080,0:13:10.090
They came from all over the[br]place.

0:13:10.090,0:13:12.159
Like I put up ads on[br]Craigslist.

0:13:12.159,0:13:14.162
I went to random tech talks.

0:13:14.162,0:13:15.642
You know, we met people at,

0:13:15.642,0:13:18.190
we used to throw weekly[br]barbeques at the office.

0:13:18.190,0:13:21.490
It was like bring your own[br]food and drinks.

0:13:21.490,0:13:23.070
And then we would just talk[br]to folks.

0:13:23.070,0:13:24.360
I think every time I ever[br]went and

0:13:24.360,0:13:25.330
got coffee in Palo Alto.

0:13:25.330,0:13:28.060
Like one of you guys was[br]recruiting at

0:13:28.060,0:13:29.570
Coopa because their office[br]was like

0:13:29.570,0:13:32.605
strategically situated next[br]to the best coffee shop.

0:13:32.605,0:13:36.050
But I think that the really[br]good people generally,

0:13:36.050,0:13:38.390
they're generally doing[br]something else, and

0:13:38.390,0:13:39.910
so you have to go seek them[br]out,

0:13:39.910,0:13:42.394
rather than expecting that[br]they're gonna seek you out.

0:13:42.394,0:13:45.000
Triple so, when no one's[br]ever heard of or

0:13:45.000,0:13:47.301
is using the product that[br]you're working on.

0:13:47.301,0:13:52.110
>> Yeah it's probably really[br]important to

0:13:52.110,0:13:55.140
have a great elevator pitch.

0:13:55.140,0:13:56.847
Not even for investors, but

0:13:56.847,0:13:59.979
just because everyone you[br]run into right now is maybe

0:13:59.979,0:14:03.730
six months, a year down the[br]road, a potential recruit.

0:14:03.730,0:14:06.280
And so, the right time to[br]have gotten them

0:14:06.280,0:14:08.200
excited about your company,[br]the right time for

0:14:08.200,0:14:09.620
them to have started[br]following us,

0:14:09.620,0:14:11.820
and be thinking about it if[br]they think about what

0:14:11.820,0:14:15.900
they're going to do next, is[br]as soon as you can start.

0:14:15.900,0:14:18.220
It's gonna take a very long[br]time to recruit people so to

0:14:18.220,0:14:20.190
being able to consistently[br]get people excited about

0:14:20.190,0:14:23.330
what you're doing will pay[br]back dividends later.

0:14:24.650,0:14:28.190
>> Maybe this is a little[br]tangential but John and

0:14:28.190,0:14:29.920
I were also chatting[br]yesterday afternoon sort of

0:14:29.920,0:14:31.850
like a bunch of our friends[br]have sort of

0:14:31.850,0:14:33.862
started companies right out[br]of school.

0:14:33.862,0:14:35.660
We were sort of thinking[br]about,

0:14:35.660,0:14:37.785
what seems to go wrong in[br]those companies.

0:14:37.785,0:14:39.685
And I think something,

0:14:39.685,0:14:43.010
that may be the most common[br]failure mode.

0:14:43.010,0:14:44.680
Since we said of doing[br]something kind of

0:14:44.680,0:14:46.940
overly niche, or overly sort[br]of specific and banded.

0:14:46.940,0:14:50.250
I think maybe it comes from[br]sort of like there's a major

0:14:50.250,0:14:51.930
shift in time horizon.

0:14:51.930,0:14:54.327
As you go from classes to[br]building a start-up.

0:14:54.327,0:14:55.543
Right?[br]A class kind of plays out in

0:14:55.543,0:14:57.193
a quarter or a semester or[br]whatever.

0:14:57.193,0:15:00.116
Whereas a start-up is like a[br]five or ten year thing.

0:15:00.116,0:15:03.163
And I think this is really[br]problematic because it's

0:15:03.163,0:15:05.379
actually quite hard to hire[br]people for

0:15:05.379,0:15:08.009
niche things, in that if you[br]tell somebody,

0:15:08.009,0:15:11.290
look we're going to build a[br]rocket that goes to Mars.

0:15:11.290,0:15:13.170
Like I mean that sounds[br]almost impossible, but

0:15:13.170,0:15:14.470
also sounds fucking awesome.

0:15:14.470,0:15:15.680
Right?[br]And so it's actually pretty

0:15:15.680,0:15:16.730
easy to convince people to[br]work on it.

0:15:16.730,0:15:17.740
Of course, if it's well, you[br]know,

0:15:17.740,0:15:20.050
we're going to build this, I[br]don't wanna single out any

0:15:20.050,0:15:21.710
particular idea, because[br]probably sound like I'm

0:15:21.710,0:15:23.090
picking some actual start up[br]that's doing it.

0:15:23.090,0:15:24.740
But you know, if you pick[br]something pretty narrow,

0:15:24.740,0:15:27.010
something that maybe kind of[br]inductively comes out of

0:15:27.010,0:15:28.450
the kinds of problems you[br]solve as part of a class

0:15:28.450,0:15:30.370
project, that's actually a[br]much harder effort.

0:15:33.460,0:15:36.410
>> One specific question[br]that has come up a lot is,

0:15:36.410,0:15:39.630
how as a relatively[br]inexperienced founder do

0:15:39.630,0:15:41.980
you identify who the really[br]great people are?

0:15:41.980,0:15:43.770
So, you know, you meet[br]people at these barbecues or

0:15:43.770,0:15:44.860
for your friends or[br]whatever, and

0:15:44.860,0:15:45.970
maybe you've worked with[br]them a little bit.

0:15:45.970,0:15:48.480
But what specifically did,

0:15:48.480,0:15:51.350
did you guys do in your[br]processes to identify like,

0:15:51.350,0:15:53.180
you know, what this person's[br]going to be really great?

0:15:53.180,0:15:54.550
Or when did you really get[br]it wrong?

0:15:54.550,0:15:55.950
But what have you learned[br]about how to

0:15:55.950,0:15:58.670
identify raw talent if you[br]can't just say, well they

0:15:58.670,0:16:00.270
worked at Google or[br]Facebook, they must be good?

0:16:03.220,0:16:04.550
>> Well, I mean, you'll[br]never

0:16:04.550,0:16:07.272
like 100% know obviously[br]until you work with folks.

0:16:07.272,0:16:09.260
Which is why the flip side[br]of it is, you know,

0:16:09.260,0:16:11.210
if someone you hired just[br]wasn't a good fit,

0:16:11.210,0:16:12.550
you owe it to the company[br]and

0:16:12.550,0:16:15.650
to them, to tell them how[br]they can improve, and

0:16:15.650,0:16:18.060
if they're not working out[br]to fire them.

0:16:18.060,0:16:19.207
But I think that the[br]generally,

0:16:19.207,0:16:21.350
the question of talent falls[br]into two big buckets.

0:16:21.350,0:16:23.320
Like, one is you have some[br]sense of

0:16:23.320,0:16:25.340
what makes them good at[br]their job, and

0:16:25.340,0:16:28.560
there are some areas where[br]you have taste in that area,

0:16:28.560,0:16:30.080
and there's some where you[br]don't.

0:16:30.080,0:16:30.780
And the ones where you

0:16:30.780,0:16:32.289
don't are actually much more[br]difficult.

0:16:33.570,0:16:36.160
So what we would do is, is[br]we would do a few things.

0:16:36.160,0:16:36.700
Like first,

0:16:36.700,0:16:39.650
before talking to anyone we[br]try to get a sense for like

0:16:39.650,0:16:42.290
what is really world class[br]in that discipline mean.

0:16:42.290,0:16:43.810
And this becomes very[br]important later,

0:16:43.810,0:16:46.140
when you're hiring things[br]like Head of Finance and

0:16:46.140,0:16:47.850
you don't anything about[br]finance.

0:16:47.850,0:16:49.930
Except what was contained in[br]like a library book you

0:16:49.930,0:16:51.800
got about like an[br]introduction to finance,

0:16:51.800,0:16:53.295
or head of marketing.

0:16:53.295,0:16:56.420
So, I always made it a habit[br]of, like, talking to

0:16:56.420,0:16:58.510
people that I knew de facto[br]were world class and

0:16:58.510,0:16:59.020
then asking them,

0:16:59.020,0:17:00.890
specifically, what are the[br]key traits or

0:17:00.890,0:17:02.030
characteristics that you[br]look for?

0:17:02.030,0:17:04.170
What are the questions that[br]you ask, and

0:17:04.170,0:17:05.819
how do you find them?

0:17:05.819,0:17:06.809
And if you're looking for

0:17:06.810,0:17:08.319
the next person that's as[br]good as you, like,

0:17:08.319,0:17:10.699
where are, where is that[br]person working right now,

0:17:10.700,0:17:12.290
and, like, what's, what's[br]her phone number?

0:17:14.420,0:17:17.280
I think that, like, learning[br]what good and bad is during

0:17:17.280,0:17:19.329
the interview process, is[br]extremely expensive.

0:17:19.329,0:17:20.579
It's an expensive use of[br]your time, and

0:17:20.579,0:17:22.138
it's an expensive use of[br]everyone else's time,

0:17:22.140,0:17:25.460
so precalibration of that[br]really matters.

0:17:25.460,0:17:26.950
And then, once you have[br]someone in, sort of,

0:17:26.950,0:17:28.470
the interview process,

0:17:28.470,0:17:30.012
you'll build the process[br]over time,

0:17:30.012,0:17:33.380
to both screen quality.

0:17:33.380,0:17:36.640
And so at Pinterest, you[br]know, we have

0:17:36.640,0:17:38.520
an evolving set of standard[br]questions that we're

0:17:38.520,0:17:40.440
always rotating through, and[br]we're always measuring,

0:17:40.440,0:17:43.360
are these good indicators or[br]bad indicators of quality?

0:17:43.360,0:17:44.900
But the other thing that the[br]questions are meant to do,

0:17:44.900,0:17:46.710
is they're supposed to give[br]a sense for,

0:17:46.710,0:17:47.740
is this the right place for

0:17:47.740,0:17:49.220
that person to come and[br]work?

0:17:49.220,0:17:50.800
And this to the point you[br]guys made about being very

0:17:50.800,0:17:53.250
transparent about what's[br]gonna be easy or hard.

0:17:53.250,0:17:55.460
Really great people wanna do[br]things that are hard,

0:17:55.460,0:17:57.530
they wanna solve tough[br]problems.

0:17:57.530,0:18:01.560
And so, there was a certain[br]brilliance in

0:18:01.560,0:18:03.250
Google setting out these[br]interview questions that

0:18:03.250,0:18:04.410
were thought to be really[br]difficult.

0:18:04.410,0:18:06.250
Because then people who like[br]solving really

0:18:06.250,0:18:10.280
difficult problems, they[br]come out and seek those.

0:18:10.280,0:18:11.780
I think it's really[br]important,

0:18:11.780,0:18:12.880
even as companies get[br]bigger,

0:18:12.880,0:18:14.870
that you don't whitewash the[br]risks.

0:18:14.870,0:18:18.120
I heard that PayPal, you'd[br]go in and

0:18:18.120,0:18:19.940
after interviewing with like[br]Peter Thiel and

0:18:19.940,0:18:22.780
Max Levchin, then they would[br]say by the way like Visa and

0:18:22.780,0:18:23.920
MasterCard wanna kill us,[br]and

0:18:23.920,0:18:26.140
we might be doing something[br]that's illegal.

0:18:26.140,0:18:29.220
But if you succeed, you'll[br]redefine payments.

0:18:29.220,0:18:30.410
Or when they were recruiting[br]for iPhone,

0:18:30.410,0:18:31.906
they didn't even tell people[br]what they were doing.

0:18:31.906,0:18:34.927
They were like you won't see[br]your families for

0:18:34.927,0:18:37.408
three years, but when you're[br]done,

0:18:37.408,0:18:40.970
your kid's kids will[br]remember what you built.

0:18:40.970,0:18:43.890
And I think that's a really[br]good thing in recruiting as

0:18:43.890,0:18:46.100
well that you're very[br]transparent about why you

0:18:46.100,0:18:47.950
think it's an amazing[br]opportunity, but

0:18:47.950,0:18:51.000
you lay out in gory detail[br]why it's gonna be hard.

0:18:51.000,0:18:53.090
And then the right people[br]select in, or

0:18:53.090,0:18:55.580
they select out of that[br]opportunity.

0:18:57.710,0:18:59.010
>> Evidence suggest it's[br]worth a lot of

0:18:59.010,0:18:59.980
people to see their kids,[br]though.

0:19:03.220,0:19:05.610
I feel like one thing you[br]have to do as you try to

0:19:05.610,0:19:12.340
identify talent is have the[br]confidence to interview for,

0:19:12.340,0:19:13.922
in a way that works for you.

0:19:13.922,0:19:17.412
I think you know, if you're,[br]say you're not,

0:19:17.412,0:19:19.410
the world's best engineer[br]and

0:19:19.410,0:19:21.850
you're trying to interview[br]engineering candidates.

0:19:21.850,0:19:23.220
I think it's tempting to try[br]to

0:19:23.220,0:19:24.880
color code what everyone[br]else does.

0:19:24.880,0:19:26.320
And you know, get them to[br]call it on white board.

0:19:26.320,0:19:28.750
And do other engineering,[br]interviewee things.

0:19:28.750,0:19:30.140
You know in the case of[br]Stripe,

0:19:30.140,0:19:31.990
hiring our first engineer.

0:19:31.990,0:19:35.730
We flew the guy out, and we[br]spent a weekend coding with

0:19:35.730,0:19:37.760
him, and you know looked[br]over his shoulder and

0:19:37.760,0:19:40.830
kind of, it was the only way[br]we could really tell and

0:19:40.830,0:19:45.070
get ourselves confident[br]that, that person was good.

0:19:45.070,0:19:47.720
And I think you can actually[br]extend that to other roles,

0:19:47.720,0:19:49.310
where again you're not an[br]expert, and

0:19:49.310,0:19:52.500
that you know I'm no[br]business development guru.

0:19:52.500,0:19:54.170
But when we interview people[br]for

0:19:54.170,0:19:55.360
business development roles,

0:19:55.360,0:19:56.962
we'll ask them to do a[br]project, you know?

0:19:56.962,0:19:58.780
Where they talk about how[br]they would

0:19:58.780,0:20:01.080
improve an existing[br]partnership that Stripe has.

0:20:01.080,0:20:02.810
Or which new partnerships[br]they would go out and do.

0:20:02.810,0:20:04.140
And again, you know,

0:20:04.140,0:20:06.520
even though it's not my[br]domain area, I am actually

0:20:06.520,0:20:09.390
confident enough that I can[br]judge those pretty well.

0:20:09.390,0:20:12.160
And I think people often[br]have this imposter syndrome

0:20:12.160,0:20:14.580
when it comes to[br]interviewing for roles.

0:20:14.580,0:20:16.135
Yeah.[br]I think a pretty specific,

0:20:16.135,0:20:18.483
just kind of just tactical[br]thing, and to do for

0:20:18.483,0:20:21.313
again the first 10 people,[br]is to work with them as much

0:20:21.313,0:20:23.620
as you can before committing[br]to hiring them.

0:20:23.620,0:20:24.370
I mean, once you

0:20:24.370,0:20:26.070
reach a certain scale it's[br]kind of impractical,

0:20:26.070,0:20:27.700
because it's a huge time[br]commitment on their side.

0:20:27.700,0:20:29.630
And ultimately I mean it[br]just would be unskilled,

0:20:29.630,0:20:31.590
would be expensive from your[br]side.

0:20:31.590,0:20:33.210
But it's really worth it

0:20:33.210,0:20:34.330
to get the first ten people[br]right.

0:20:34.330,0:20:35.180
And so, for

0:20:35.180,0:20:38.120
a majority, maybe in all of[br]the first ten people,

0:20:38.120,0:20:39.680
we worked with them in some[br]capacity.

0:20:39.680,0:20:41.480
Usually for a week in[br]advance, and

0:20:41.480,0:20:43.777
it's pretty hard to fake it[br]for a week.

0:20:45.010,0:20:49.020
It tends to be pretty clear,[br]quite quickly.

0:20:49.020,0:20:52.840
And another thing I thought[br]of you know, to the question

0:20:52.840,0:20:54.830
of sort of how do you know[br]like who is great or

0:20:54.830,0:20:55.870
who is good enough or[br]whatever.

0:20:55.870,0:20:57.740
And people always talk about[br]this notion of

0:20:57.740,0:21:00.350
the ten x person, that[br]whatever that skill set is.

0:21:00.350,0:21:02.210
I don't really know what ten[br]x means,

0:21:02.210,0:21:05.550
I think maybe the slightly[br]more kind of helpful or like

0:21:05.550,0:21:09.180
intuitive version of that[br]is, is this person probably,

0:21:09.180,0:21:12.090
the best out of all of their[br]friends at what they do.

0:21:13.560,0:21:15.600
And so you know, it's a[br]little bit sensitive to you

0:21:15.600,0:21:18.290
know, how well they chose[br]friends, how many they have.

0:21:18.290,0:21:20.150
But for me at least, I find[br]that kind of

0:21:20.150,0:21:21.550
a more helpful way to think[br]about it, like

0:21:21.550,0:21:23.209
is this the best engineer[br]this engineer knows.

0:21:24.510,0:21:25.230
And the other thing I

0:21:25.230,0:21:26.570
think that's actually[br]probably just worth

0:21:26.570,0:21:28.320
mentioning in all of those[br]kind of first ten people, or

0:21:28.320,0:21:31.550
even more generally on the[br]culture and hiring topic.

0:21:31.550,0:21:34.650
I think everyone sort of[br]doesn't realize until they

0:21:34.650,0:21:36.620
go through it themselves how[br]it important it is,

0:21:36.620,0:21:40.210
in large part because the,[br]like in life and the media

0:21:40.210,0:21:42.580
and everything, people focus[br]way too much on founders.

0:21:42.580,0:21:43.780
And they're like, here we[br]are.

0:21:43.780,0:21:44.430
And so kind of,

0:21:44.430,0:21:46.760
we're re-enforcing the sort[br]of structural narrative that

0:21:46.760,0:21:48.660
like, stripe is about John[br]and Patrick.

0:21:48.660,0:21:50.510
And Pinterest is about Ben[br]and so forth.

0:21:50.510,0:21:51.830
Whereas, I mean, obviously,

0:21:51.830,0:21:54.320
sort of the vast majority of[br]what our companies do

0:21:54.320,0:21:56.860
like 99.9% is being down by[br]people who are not us.

0:21:56.860,0:21:57.490
Right?

0:21:57.490,0:21:58.930
And I think, I mean, that's[br]kind of,

0:21:58.930,0:22:02.410
it's obvious when you say[br]it, but it's at a very much,

0:22:02.410,0:22:03.840
not just the macro narrative[br]and

0:22:03.840,0:22:04.970
you know, companies are[br]abstract.

0:22:04.970,0:22:06.670
So you kind of need to[br]associate them with people.

0:22:06.670,0:22:08.230
But I think it's worth[br]bearing in mind that like

0:22:08.230,0:22:12.010
for, you know, Apple, you[br]know everything was

0:22:12.010,0:22:14.840
you know, Steve Jobs was a[br]like tiny, tiny detail at

0:22:14.840,0:22:17.240
the end, right, or Google[br]was that way, and so forth.

0:22:17.240,0:22:19.590
>> So don't screw it up, is[br]that what you're saying?

0:22:19.590,0:22:20.430
>> Something like that.

0:22:20.430,0:22:22.870
>> I think, you know, one[br]other thing I'll mention is

0:22:22.870,0:22:25.360
that I think referencing[br]people is really important,

0:22:25.360,0:22:30.270
and referencing people is[br]just what it sounds like.

0:22:30.270,0:22:31.780
You're basically asking[br]people who have real,

0:22:31.780,0:22:34.890
material working experience,[br]for their honest opinion.

0:22:34.890,0:22:38.220
And we do that really[br]aggressively, and

0:22:38.220,0:22:38.740
what we're trying to

0:22:38.740,0:22:40.900
figure out is what's this[br]person like to work with?

0:22:42.590,0:22:44.070
We're not trying to validate[br]like whether they

0:22:44.070,0:22:45.260
told the truth on their[br]resume,

0:22:45.260,0:22:47.400
cuz we assume that they're[br]telling the truth.

0:22:47.400,0:22:50.740
So, very standard questions[br]that I'll ask somebody who's

0:22:50.740,0:22:53.460
in an interview, I might[br]say, hey,

0:22:53.460,0:22:55.500
we both know Jonathan in[br]common.

0:22:55.500,0:22:57.240
I'm gonna talk to him, you[br]know,

0:22:57.240,0:22:59.090
in a couple weeks, cause[br]we're both social friends.

0:22:59.090,0:23:02.560
Like if I asked him, what[br]you're the best at, or

0:23:02.560,0:23:03.690
what you would be most proud[br]of, or

0:23:03.690,0:23:06.230
what you were kind of[br]working to improve.

0:23:06.230,0:23:08.770
What would he or she say?

0:23:08.770,0:23:10.760
Because it sort of tests the[br]level of self awareness, and

0:23:10.760,0:23:12.900
creates a bit of social[br]accountability.

0:23:12.900,0:23:15.550
And then I'll try to ask[br]something that makes the

0:23:15.550,0:23:18.170
question which is typically[br]very soft feel a little bit

0:23:18.170,0:23:20.300
more quantitative, and then[br]calibrate that over time.

0:23:20.300,0:23:21.890
So you know,

0:23:21.890,0:23:24.680
in evaluating this person or[br]this dimension, do you think

0:23:24.680,0:23:26.110
this is the top 1% of people[br]you've ever worked with?

0:23:26.110,0:23:28.090
The top 5% and the top 10%?

0:23:28.090,0:23:31.410
And it forces a scarcity[br]that gives a materially

0:23:31.410,0:23:34.110
different reference, then if[br]you just say hey,

0:23:34.110,0:23:36.060
like what's the best thing[br]about John?

0:23:36.060,0:23:37.680
He told me you know, he's[br]good at these things.

0:23:37.680,0:23:38.250
Can you validate?

0:23:38.250,0:23:38.840
And you're like yeah,

0:23:38.840,0:23:40.220
sure, Cuz they don't gain[br]anything.

0:23:40.220,0:23:43.132
So I think that's just a[br]tool that people should

0:23:43.132,0:23:43.920
take seriously.

0:23:43.920,0:23:44.559
>> Yeah.

0:23:44.559,0:23:45.149
>> Yeah, and

0:23:45.149,0:23:48.168
referencing isn't obviously[br]easy to begin with, but

0:23:48.168,0:23:51.400
it does actually prove[br]really useful over time.

0:23:51.400,0:23:54.160
And you just have to, people[br]I think especially for

0:23:54.160,0:23:56.350
named references, people[br]really want to be nice.

0:23:56.350,0:23:58.370
So you have to do things[br]like create an artificial

0:23:58.370,0:24:00.600
scarcity by saying you know,[br]where would you rank this

0:24:00.600,0:24:03.230
person amongst the people[br]you worked within this role?

0:24:03.230,0:24:03.780
Or, you know,

0:24:03.780,0:24:05.730
you just splitz out on the[br]enchant of how to reference,

0:24:05.730,0:24:07.630
but usually I aim to spend,[br]you know,

0:24:07.630,0:24:09.470
15 minutes on the phone with[br]that person.

0:24:09.470,0:24:12.520
Not just let them say, yeah[br]this person is awesome, and

0:24:12.520,0:24:13.720
then hang up.

0:24:13.720,0:24:16.190
>> That's also a tremendous[br]source of new recruits.

0:24:16.190,0:24:16.800
>> Yeah.

0:24:16.800,0:24:18.530
>> Those references are also[br]a tremendous source of

0:24:18.530,0:24:19.120
new recruits.

0:24:20.900,0:24:22.690
What have each of you[br]learned about,

0:24:22.690,0:24:25.770
once you hire these first[br]people and they joined?

0:24:25.770,0:24:28.300
What have you done to make[br]them effective quickly?

0:24:28.300,0:24:31.280
To get them sort of, you[br]know,

0:24:31.280,0:24:34.190
to the right cultural place,[br]you know?

0:24:34.190,0:24:36.600
Cuz sort of hiring is[br]usually very difficult, but

0:24:36.600,0:24:38.430
then not as difficult as[br]making them

0:24:38.430,0:24:39.150
happy and effective.

0:24:39.150,0:24:39.860
So what do you do

0:24:39.860,0:24:41.500
with these early employees[br]to accomplish that?

0:24:44.640,0:24:46.860
>> Well, ways that it's[br]changed when we've gone from

0:24:46.860,0:24:48.790
really small to bigger.

0:24:48.790,0:24:49.800
When we first started,

0:24:49.800,0:24:52.970
we were generally hiring[br]because we needed

0:24:52.970,0:24:54.700
that person, like, a long[br]time ago.

0:24:54.700,0:24:57.090
And so their whole[br]on-boarding was like,

0:24:57.090,0:24:57.690
here's your computer,

0:24:57.690,0:24:58.880
we already set up your[br]environment,

0:24:58.880,0:24:59.500
don't worry about it,

0:24:59.500,0:25:01.690
this is the problem that we[br]have to solve together.

0:25:01.690,0:25:04.810
And then because of the[br]nature of the start ups,

0:25:04.810,0:25:07.380
we were all in this, like,[br]tiny, two bedroom apartment.

0:25:08.770,0:25:09.760
All of the other things,

0:25:09.760,0:25:11.250
like building personal[br]relationships,

0:25:11.250,0:25:12.680
spending time together,[br]getting to know one another,

0:25:12.680,0:25:13.810
it just all happened.

0:25:13.810,0:25:14.990
Kind of automatically like,

0:25:14.990,0:25:16.145
you didn't have to do[br]anything.

0:25:16.145,0:25:19.730
The one thing that I would[br]add to that is, though,

0:25:19.730,0:25:21.420
we would always try to[br]remind people of like,

0:25:21.420,0:25:22.600
where we wanted to go some[br]day.

0:25:22.600,0:25:24.370
Cuz it's really easy when[br]you first join to drop

0:25:24.370,0:25:26.380
somebody into a problem, and[br]then they think the whole

0:25:26.380,0:25:28.520
world is like this little[br]problem in front of them.

0:25:28.520,0:25:29.080
We'd always say,

0:25:29.080,0:25:31.320
you know, hey, some day we[br]wanna do for

0:25:33.130,0:25:34.830
Discovery what Google did[br]for search.

0:25:34.830,0:25:37.300
This is our plan for trying[br]to get that done.

0:25:37.300,0:25:40.690
Now as the company, grows I[br]think that process has to

0:25:40.690,0:25:42.300
get a little bit more[br]formalized.

0:25:42.300,0:25:45.660
And so we spend a lot of[br]time thinking, and honestly

0:25:45.660,0:25:48.770
trying to constantly refine[br]what is that person's

0:25:48.770,0:25:51.560
experience look like from[br]the day they came in,

0:25:51.560,0:25:53.240
to their very first[br]interview,

0:25:53.240,0:25:54.960
through 30 days after they[br]joined.

0:25:54.960,0:25:57.590
Like do they have somebody[br]whose name they know?

0:25:57.590,0:25:58.730
Do they know who their[br]manager is?

0:25:58.730,0:26:01.100
Have they sat down with[br]people on their team?

0:26:01.100,0:26:03.070
Do they know what the[br]general architecture of

0:26:03.070,0:26:05.700
the company is, and what the[br]top priorities are?

0:26:05.700,0:26:08.390
And we have a program and[br]it's a week long, and

0:26:08.390,0:26:10.370
then there are some function[br]specific programs that

0:26:10.370,0:26:11.740
go in deeper.

0:26:11.740,0:26:13.638
And that's something that's[br]always been refined.

0:26:13.638,0:26:15.190
And the alpha metrics that[br]we look on that,

0:26:15.190,0:26:17.400
it's when we ask people like[br]what did you think

0:26:17.400,0:26:19.090
afterwards and then 30 days[br]afterwards.

0:26:19.090,0:26:21.030
And then we also ask their[br]peers and their manager,

0:26:21.030,0:26:23.330
like hey, is this person[br]kind of up to speed?

0:26:23.330,0:26:24.090
Do you feel like we've done

0:26:24.090,0:26:25.290
a good job making them[br]productive?

0:26:25.290,0:26:28.330
And if we haven't, that's a[br]key A, that, that team

0:26:28.330,0:26:30.760
shouldn't be hiring anymore[br]people, because they're not

0:26:30.760,0:26:32.720
doing a good job bringing in[br]new people.

0:26:33.740,0:26:35.655
And B, that we need to[br]retool that.

0:26:35.655,0:26:37.520
And so I think those things[br]are important.

0:26:37.520,0:26:40.450
I just wouldn't discount[br]just how important it is to

0:26:40.450,0:26:42.380
get to know the person as a[br]person.

0:26:42.380,0:26:43.840
Like what are their[br]aspirations?

0:26:43.840,0:26:44.730
What's their working style?

0:26:44.730,0:26:47.250
How do they like to be[br]recognized?

0:26:48.450,0:26:51.810
Do they really prefer being[br]in total silence?

0:26:51.810,0:26:53.070
Are they a morning person or[br]night person?

0:26:53.070,0:26:56.400
Like knowing those things I[br]think, it just

0:26:56.400,0:26:58.270
demonstrates that you care[br]about them individually as

0:26:58.270,0:27:00.320
well as collectively, like[br]what you goals are.

0:27:00.320,0:27:04.540
>> I think there are two[br]things that are important at

0:27:04.540,0:27:06.680
any stage through the[br]implementation of a change.

0:27:06.680,0:27:08.330
The first is to get them up[br]and

0:27:08.330,0:27:10.480
running, doing real work[br]quickly.

0:27:10.480,0:27:12.560
Because that's only when you[br]can find the problems,

0:27:12.560,0:27:15.110
that's how progress is[br]measured,

0:27:15.110,0:27:16.340
is how much real work[br]they're doing.

0:27:16.340,0:27:19.000
And so, you know, when we[br]have engineers start,

0:27:19.000,0:27:20.800
we'll try to get them[br]committing on the first day.

0:27:20.800,0:27:21.620
When we have people you[br]know,

0:27:21.620,0:27:23.150
people in business roles[br]start,

0:27:23.150,0:27:26.750
we will have them in real[br]meetings on the first day,

0:27:26.750,0:27:28.390
on what they're meant to be[br]working on.

0:27:28.390,0:27:30.260
But I think sometimes[br]there's a tendency to

0:27:30.260,0:27:32.430
be tentative, and help ease[br]people in.

0:27:32.430,0:27:36.930
We're much more of the push[br]people off the cliff school.

0:27:36.930,0:27:38.820
And then I think the second,[br]is to try and

0:27:38.820,0:27:41.690
start quickly giving people[br]feedback, and

0:27:41.690,0:27:43.710
especially feedback on how[br]to adapt to the culture.

0:27:43.710,0:27:46.020
Because when you think about[br]it, like, you know,

0:27:46.020,0:27:49.490
if you have built a strong[br]culture as, you know,

0:27:49.490,0:27:52.310
all the companies up here[br]are trying to,

0:27:52.310,0:27:55.080
then it's going to take some[br]adapting for the person.

0:27:55.080,0:27:58.210
It's not going to be[br]necessarily easy, you know?

0:27:58.210,0:27:59.150
One thing we have at Stripe,

0:27:59.150,0:28:02.210
is just the culture was a[br]lot more written.

0:28:02.210,0:28:03.050
And so, you know?

0:28:03.050,0:28:04.480
You have people who are[br]right next to

0:28:04.480,0:28:05.790
each other each with[br]headphones on, and

0:28:05.790,0:28:07.360
just like IM'ing away to[br]each other.

0:28:07.360,0:28:08.170
And for you know,

0:28:08.170,0:28:09.490
a lot of people coming in[br]who hadn't worked in

0:28:09.490,0:28:11.210
an environment like that,[br]it's, it's sort of hard.

0:28:11.210,0:28:12.050
>> In normal places.

0:28:12.050,0:28:12.680
>> Exactly.[br]Yeah.

0:28:13.980,0:28:16.520
And so, everything from kind[br]of high level how you're

0:28:16.520,0:28:19.780
doing at your job, to kind[br]of minor cultural pointers.

0:28:19.780,0:28:22.530
The more feedback you can[br]give them,

0:28:22.530,0:28:23.440
the better they'll do.

0:28:23.440,0:28:24.150
And it's unnatural.

0:28:24.150,0:28:24.660
Right?[br]Because it's

0:28:24.660,0:28:27.850
unnatural to be constantly[br]telling people that

0:28:27.850,0:28:28.980
they're doing a good or bad[br]job.

0:28:28.980,0:28:30.870
You don't do that,[br]hopefully,

0:28:30.870,0:28:34.210
in your normal life, you're[br]more restrained.

0:28:34.210,0:28:35.570
But when you have employees,

0:28:35.570,0:28:38.470
that's kind of what you owe[br]them for them to do well.

0:28:38.470,0:28:40.170
>> So I think this is sort[br]of a good transition too,

0:28:40.170,0:28:41.609
as your companies have[br]scaled.

0:28:43.490,0:28:46.420
What are the biggest changes[br]you've had to make to your

0:28:46.420,0:28:47.560
hiring processes?

0:28:47.560,0:28:50.140
And also how you sort of[br]manage and

0:28:50.140,0:28:53.020
run the team, as you've gone[br]from, you know,

0:28:53.020,0:28:54.980
two to 10 to 100 to 1000[br]employees?

0:28:59.010,0:28:59.950
>> There a lot of changes.

0:29:01.650,0:29:03.550
I mean, I think one thing[br]that we try to do on

0:29:03.550,0:29:07.040
the team side is, the goal[br]is to make the teams feel as

0:29:07.040,0:29:09.130
autonomous and nimble as[br]possible within

0:29:09.130,0:29:11.140
the constraints of having a[br]large organization.

0:29:11.140,0:29:14.020
And that means over time,[br]that we're always trying to

0:29:14.020,0:29:16.430
make it feel like a start-up[br]of of many start-ups,

0:29:16.430,0:29:18.690
rather than this monolithic[br]organization with a set of

0:29:18.690,0:29:21.170
foreign processes that cut[br]horizontally through it.

0:29:21.170,0:29:22.290
And it's easier said than[br]done.

0:29:22.290,0:29:24.280
I think that, like we're not[br]all the way there.

0:29:24.280,0:29:27.830
But one of our goals is that[br]each team has, they control

0:29:27.830,0:29:29.772
the resources that they need[br]to get things done,

0:29:29.772,0:29:32.960
they know what the most[br]important thing is, and

0:29:32.960,0:29:34.150
how it's measured.

0:29:34.150,0:29:35.320
And in that way,

0:29:35.320,0:29:37.490
then the management problem[br]becomes somewhat tractable,

0:29:37.490,0:29:39.410
otherwise it feels[br]completely impossible if

0:29:39.410,0:29:41.610
you can't decompose it into[br]autonomous units.

0:29:41.610,0:29:43.609
Like you just look at it,[br]and you're like, oh my gosh,

0:29:43.609,0:29:45.752
communication complexity is[br]increasing geometrically.

0:29:45.752,0:29:46.810
Management complexity,

0:29:46.810,0:29:48.250
like it's never gonna work,[br]and so

0:29:48.250,0:29:50.411
you have to sort of create[br]these abstracted units, or

0:29:50.411,0:29:52.247
that's at least what we're[br]gonna try to do.

0:29:52.247,0:29:54.741
And at Pinterest in[br]particular, the real

0:29:54.741,0:29:57.811
challenge of creating those[br]abstracted units is.

0:29:57.811,0:29:59.296
We want to actually have

0:29:59.296,0:30:02.419
units that encompass a[br]super-strong designer,

0:30:02.419,0:30:06.060
a super-strong leader in[br]engineering, often a writer,

0:30:06.060,0:30:08.830
often sometimes like a[br]community manager.

0:30:08.830,0:30:10.890
We want them all to be[br]self-contained.

0:30:10.890,0:30:12.190
And so that makes the[br]problem really hard,

0:30:12.190,0:30:14.860
but that's kind of core to[br]our philosophy of how we

0:30:14.860,0:30:15.480
build products.

0:30:15.480,0:30:17.560
We try to put people[br]together that have all of

0:30:17.560,0:30:18.520
these different disciplines.

0:30:18.520,0:30:20.056
They're curious about lots[br]of things.

0:30:20.056,0:30:21.680
And then we anchor them on a[br]single project and then we

0:30:21.680,0:30:24.115
try to remove barriers to[br]let them go fast.

0:30:24.115,0:30:25.940
When we find new barriers,[br]we sit down and

0:30:25.940,0:30:28.180
we think, how do we speed[br]that up?

0:30:28.180,0:30:30.850
I think hiring is a little[br]bit different.

0:30:30.850,0:30:33.710
And I think the biggest[br]change and

0:30:33.710,0:30:35.200
the biggest asset that you[br]get as you get more

0:30:35.200,0:30:38.890
people is referrals become[br]like more and more and

0:30:38.890,0:30:40.660
more and more the life blood[br]depending on

0:30:40.660,0:30:41.998
the network of all the[br]people that you bring in.

0:30:41.998,0:30:43.380
So one of the really lucky,

0:30:43.380,0:30:46.087
and in hindsight, great[br]decisions that we made, was

0:30:46.087,0:30:48.794
actually the 14th or 15th[br]person that we hired was

0:30:48.794,0:30:51.888
a professional recruiter and[br]she had worked at start-ups.

0:30:51.888,0:30:54.580
She had worked at big[br]companies like Apple.

0:30:54.580,0:30:57.672
But she knew where that[br]pipeline breaks down,

0:30:57.672,0:30:59.812
knew the early indicators,[br]and

0:30:59.812,0:31:03.460
taught everyone not just how[br]to screen for talent, but

0:31:03.460,0:31:06.949
how scalably to identify the[br]people that are gonna be

0:31:06.949,0:31:09.900
culturally, really good for[br]the company.

0:31:09.900,0:31:12.076
And I think looking back on[br]that,

0:31:12.076,0:31:15.350
it's something that I[br]personally really value.

0:31:16.770,0:31:17.894
>> You know, there's just a[br]huge amount of

0:31:17.894,0:31:18.495
stuff here, right?

0:31:18.495,0:31:19.632
In that I guess,

0:31:19.632,0:31:23.408
this is all under kinda the[br]rubric of managing growth.

0:31:23.408,0:31:27.737
And either your company[br]fails very quickly or all of

0:31:27.737,0:31:31.210
your problems become in some[br]way about managing growth.

0:31:32.800,0:31:35.030
One thing that I think tends[br]to take people by surprise,

0:31:35.030,0:31:36.550
and certainly took me by[br]surprise,

0:31:36.550,0:31:39.330
is kind of how quickly just[br]like the time

0:31:39.330,0:31:40.370
horizon has changed.

0:31:40.370,0:31:44.720
In that, in your first month[br]you're largely thinking

0:31:44.720,0:31:48.660
about things maybe one month[br]ahead, right?

0:31:48.660,0:31:50.168
In that, maybe that's kind[br]of

0:31:50.168,0:31:52.721
what your development road[br]map is oriented around,

0:31:52.721,0:31:54.820
even in terms of who you're[br]working with.

0:31:54.820,0:31:55.370
Maybe it's like,

0:31:55.370,0:31:58.280
informal relationships where[br]they haven't fully committed

0:31:58.280,0:32:00.160
to remain full time or[br]whatever, right?

0:32:00.160,0:32:02.372
And then, the more time goes[br]by,

0:32:02.372,0:32:05.058
I think that kind of has a[br]reciprocal, or

0:32:05.058,0:32:08.220
corresponding increase in[br]the time horizons.

0:32:08.220,0:32:09.920
After one year, you're kind[br]of thinking one year ahead.

0:32:09.920,0:32:13.675
After four years, you're[br]thinking four years ahead.

0:32:13.675,0:32:16.870
But that increases very[br]quickly, right?

0:32:16.870,0:32:17.879
And so after you know,

0:32:17.879,0:32:20.440
it's after one month, it's[br]again, super short term.

0:32:20.440,0:32:23.180
And then just 11 months[br]later you should now

0:32:23.180,0:32:25.650
actually be thinking and[br]planning a year ahead, and

0:32:25.650,0:32:27.250
thinking about human[br]structures on

0:32:27.250,0:32:28.162
that time horizon.

0:32:28.162,0:32:31.315
Thinking about the stuff Ben[br]talked about, like where you

0:32:31.315,0:32:34.530
want to be going long term[br]and things like that.

0:32:34.530,0:32:36.773
I think that also plays into[br]your hiring and

0:32:36.773,0:32:38.775
the kinds of people you[br]hire, right?

0:32:38.775,0:32:40.499
In that in the really early[br]days you kind of

0:32:40.499,0:32:42.322
have to hire people who will[br]be productive,

0:32:42.322,0:32:43.520
essentially immediately.

0:32:44.670,0:32:47.812
You don't have the luxury of[br]hiring people who are really

0:32:47.812,0:32:50.496
promising, but they're not[br]quite gonna be up to

0:32:50.496,0:32:52.420
speed for another year or[br]two.

0:32:52.420,0:32:54.090
They have to be able to[br]contribute immediately.

0:32:54.090,0:32:57.330
But after two or three[br]years, now it starts to get

0:32:57.330,0:32:59.550
much more reasonable to make[br]those investments and

0:32:59.550,0:33:01.110
in fact, if you're not[br]making those investments,

0:33:01.110,0:33:03.380
you're probably being much[br]too short term.

0:33:03.380,0:33:06.060
And so I think that's really[br]important.

0:33:06.060,0:33:07.880
And a lot of it also just[br]comes from

0:33:09.420,0:33:11.355
all these problems in some[br]sense are easy,

0:33:11.355,0:33:14.240
like how do you build good[br]social bonds between people?

0:33:14.240,0:33:17.795
I mean, we all do it every[br]day, right?

0:33:17.795,0:33:21.355
It's kind of, how do you[br]make it systematic and

0:33:21.355,0:33:22.968
effective at scale?

0:33:22.968,0:33:26.597
It's always a very severely[br]imperfect approximation of

0:33:26.597,0:33:29.620
what you would ideally do if[br]you were small, and

0:33:29.620,0:33:32.566
what hacks can you pull to[br]make it work as well as

0:33:32.566,0:33:35.460
you possibly can at a larger[br]scale.

0:33:35.460,0:33:36.804
Like a rapidly growing[br]company,

0:33:36.804,0:33:38.244
say growing headcount here,[br]two or

0:33:38.244,0:33:40.412
three x a year, it is a very[br]unnatural thing.

0:33:40.412,0:33:43.327
And so it's, what's the[br]least bad way of sort of

0:33:43.327,0:33:45.370
managing that period of[br]growth?

0:33:45.370,0:33:46.620
Human organizations aren't[br]designed for

0:33:46.620,0:33:52.240
it and I think it's worth[br]being quite systematic about

0:33:52.240,0:33:53.360
thinking about ways to do[br]that.

0:33:53.360,0:33:56.650
But realizing that you[br]probably can't do

0:33:56.650,0:33:59.420
much better than adequation.

0:33:59.420,0:34:01.464
You know, for Stripe, it's[br]things like we

0:34:01.464,0:34:03.120
have three meals every day[br]at sort of

0:34:03.120,0:34:06.008
long tables where everyone[br]can sit together, right?

0:34:06.008,0:34:09.190
And if you think of net, how[br]much more total human

0:34:09.190,0:34:12.168
interaction happens as a[br]result of having these kind

0:34:12.168,0:34:15.232
of randomly mixed meals,[br]it's vast, right?

0:34:15.233,0:34:17.900
And it's kind of a whole[br]list of things like that.

0:34:17.900,0:34:20.260
But, I think that's kind of[br]the general framework.

0:34:22.020,0:34:23.500
>> One thing I'm really[br]curious about you guys value

0:34:23.500,0:34:24.053
transparency.

0:34:24.053,0:34:26.577
How have you scaled it over[br]time?

0:34:26.577,0:34:28.320
I know we think about[br]figuring it out all

0:34:28.320,0:34:29.318
the time, just curious.

0:34:29.318,0:34:33.404
>> So you know, start-ups, I[br]can't remember who it was,

0:34:33.404,0:34:37.576
defined it as a, start-up is[br]an organization that's not

0:34:37.576,0:34:41.510
yet stuck with all of these[br]principle agent problems.

0:34:41.510,0:34:43.616
That at most large[br]companies,

0:34:43.616,0:34:47.261
what is locally optimal for[br]you is very frequently not

0:34:47.261,0:34:50.204
what is globally optimal for[br]the company.

0:34:50.204,0:34:56.850
And so there are, as a[br]consequence of that,

0:34:56.850,0:34:58.480
probably a lot of ways in[br]which a start-up can

0:34:58.480,0:35:00.270
work differently to a big[br]company.

0:35:00.270,0:35:04.130
At a big company, a lot of[br]the things that are good for

0:35:04.130,0:35:07.460
you, well, you couldn't do[br]them in

0:35:07.460,0:35:09.640
a completely transparent[br]environment, right?

0:35:09.640,0:35:11.350
Because people would think[br]less of you or

0:35:11.350,0:35:13.180
you're doing things you're[br]not supposed to.

0:35:13.180,0:35:15.005
But because everyone is kind[br]of sailing in the, or

0:35:15.005,0:35:16.916
rowing in the same direction[br]at a start-up, you can

0:35:16.916,0:35:19.022
actually just make all of[br]the information transparent.

0:35:19.022,0:35:23.369
And so, I guess I mentioned[br]earlier, Stripe used to BCC

0:35:23.369,0:35:27.626
every other person at Stripe[br]on basically every single

0:35:27.626,0:35:31.789
outgoing email unless you[br]opted out of it, because we

0:35:31.789,0:35:35.230
thought that would be much[br]more efficient.

0:35:35.230,0:35:37.470
You wouldn't need to have as[br]many meetings,

0:35:37.470,0:35:39.370
if you just kind of keep[br]abreast of what's happening.

0:35:39.370,0:35:42.590
And over time we've sort of[br]built an increasingly

0:35:42.590,0:35:45.700
intricate framework of[br]mailing lists and

0:35:45.700,0:35:48.680
we now have a program for[br]generating Gmail filters.

0:35:48.680,0:35:51.390
And for like a pretty rocky[br]patch where 50 people or

0:35:51.390,0:35:54.840
so, to Ben's point of like,[br]asking people how they're

0:35:54.840,0:35:56.380
getting on after the first[br]couple of days.

0:35:56.380,0:35:58.671
They're reported terribly[br]because like they couldn't

0:35:58.671,0:36:00.722
even find all the emails[br]that people were sending to

0:36:00.722,0:36:03.710
them and they were missing[br]things and everything.

0:36:03.710,0:36:05.080
>> Gmail broke at one stage.

0:36:05.080,0:36:06.584
>> Oh, right, at one point[br]Gmail broke because we

0:36:06.584,0:36:07.838
were just like sending too[br]much email.

0:36:07.838,0:36:10.804
It is hard to scale, because[br]I mean,

0:36:10.804,0:36:15.210
you might contact somebody[br]outside of the company with

0:36:15.210,0:36:18.657
like some great idea, and[br]maybe the person

0:36:18.657,0:36:22.776
sitting across the way from[br]you thinks that's like

0:36:22.776,0:36:26.818
the stupidest idea they've[br]ever heard, right?

0:36:26.818,0:36:29.124
And you're kind of subject[br]to the scrutiny of

0:36:29.124,0:36:31.369
the entire organization, to[br]some degree,

0:36:31.369,0:36:33.080
with all of your[br]communication.

0:36:33.080,0:36:35.125
Like kind of the challenging[br]side of it.

0:36:35.125,0:36:38.110
Then the good side is people[br]are much more informed about

0:36:38.110,0:36:38.930
what's happening.

0:36:40.212,0:36:45.312
I guess I don't feel that I[br]can give much

0:36:45.312,0:36:50.510
stronger endorsement of it[br]than it has worked so far.

0:36:50.510,0:36:52.216
>> That's a pretty good[br]endorsement.

0:36:52.216,0:36:55.550
>> Yeah, I'm actually really[br]curious how or whether it'll

0:36:55.550,0:36:57.940
work when we're 5,000 people[br]or something like that.

0:36:57.940,0:36:59.950
If we're ever at that scale.

0:36:59.950,0:37:01.110
>> I think the two things[br]that have

0:37:01.110,0:37:04.400
helped us scale it are one,[br]changing the tools and

0:37:04.400,0:37:06.060
two, developing the culture[br]around it.

0:37:06.060,0:37:08.010
And so on the tools front,[br]you know,

0:37:08.010,0:37:10.760
used to be the case, or the[br]infrastructure, it

0:37:10.760,0:37:13.670
used to be the case that you[br]could keep abreast by what's

0:37:13.670,0:37:15.700
happening in the company by[br]reading all the email.

0:37:15.700,0:37:18.480
Now we have weekly all-hands[br]sometimes on the deck,

0:37:18.480,0:37:20.720
and we actually have to put[br]all this work into

0:37:20.720,0:37:22.830
developing a deck to, to[br]communicate to people what's

0:37:22.830,0:37:25.640
going on in the company,[br]since there's so much more.

0:37:25.640,0:37:29.820
And the second is on the[br]cultural side, so much

0:37:29.820,0:37:33.660
information being available[br]internally, you have to

0:37:33.660,0:37:35.950
develop cultural norms[br]around how it's treated.

0:37:35.950,0:37:38.680
Obvious things like the fact[br]that a lot of

0:37:38.680,0:37:40.590
it's confidential to Stripe.

0:37:40.590,0:37:44.260
But even less obvious things[br]like when emailing someone

0:37:44.260,0:37:46.190
or talking in Slack or

0:37:46.190,0:37:48.860
IRESE, when that is viewed[br]by now 170 people,

0:37:48.860,0:37:52.524
it's pretty easy to get[br]stage fright.

0:37:52.524,0:37:55.970
And it's, it's pretty easy[br]for what you thought was

0:37:55.970,0:37:59.110
a reasonable proposal, you[br]get this drive-by criticism.

0:37:59.110,0:38:01.890
And you're now less likely[br]to share in the future.

0:38:01.890,0:38:05.200
And so, we've had to create[br]norms around when it's

0:38:05.200,0:38:06.650
reasonable to jump into[br]discussions and

0:38:06.650,0:38:08.010
how that interaction works.

0:38:08.010,0:38:11.110
Because people are on the[br]stage so much more.

0:38:11.110,0:38:12.300
>> I'm pretty sure it's not[br]good,

0:38:12.300,0:38:14.530
but, not to put her on the[br]spot but

0:38:14.530,0:38:16.180
Emily interned at Stripe[br]this summer.

0:38:16.180,0:38:17.140
I'm curious, like,

0:38:17.140,0:38:19.480
as an intern, what you[br]thought of it?

0:38:19.480,0:38:20.920
>> I think, overall, it's[br]great.

0:38:20.920,0:38:22.351
I think, like, my first[br]week,

0:38:22.351,0:38:24.259
I spent most of the time[br]reading Hackpad and

0:38:24.259,0:38:26.670
getting caught up on what[br]the company was doing.

0:38:26.670,0:38:28.870
It can often be quite[br]distracting,

0:38:28.870,0:38:31.460
from your own work, as there[br]are oftentimes other parts

0:38:31.460,0:38:33.330
of the company you're really[br]interested in.

0:38:33.330,0:38:35.050
>> Hackpad, by the way, is[br]like Google Docs, but

0:38:35.050,0:38:35.760
with a news feed and

0:38:35.760,0:38:37.600
that way you can just like[br]see all the documents.

0:38:37.600,0:38:38.860
>> Yeah, and you're[br]encouraged to make

0:38:38.860,0:38:40.229
everything public,[br]everything you work on.

0:38:41.410,0:38:43.310
But overall, it gets you[br]spun up really quickly.

0:38:43.310,0:38:45.172
And we also have things[br]called spin ups where

0:38:45.172,0:38:47.990
every single leader of a[br]team at the company,

0:38:47.990,0:38:51.280
whether it's sys or it's[br]product gives like a 30

0:38:51.280,0:38:53.360
minute talk on what their[br]team is currently doing and

0:38:53.360,0:38:55.120
how you can contribute, if[br]you're interested.

0:38:55.120,0:38:56.925
>> Do you think email[br]transparency was net good?

0:38:56.925,0:38:59.982
>> Yes I remember having a[br]hard time understanding what

0:38:59.982,0:39:02.410
I should and should not[br]subscribe too.

0:39:02.410,0:39:05.480
And the first week having[br]2,000 emails in my inbox and

0:39:05.480,0:39:07.480
then by the end there are[br]three of four teams you

0:39:07.480,0:39:11.940
actually want that[br]information coming in from.

0:39:11.940,0:39:13.903
>> All right, audience[br]questions.

0:39:13.903,0:39:14.971
Yes.

0:39:14.971,0:39:17.974
>> So this question for[br]Patrick and John.

0:39:17.974,0:39:21.766
Is it your experience that[br]your early hires begin to

0:39:21.766,0:39:22.383
grow and

0:39:22.383,0:39:26.900
evolve into leadership roles[br]as the company scales?

0:39:26.900,0:39:28.421
And a question for

0:39:28.421,0:39:33.330
Ben, how is the difference,[br]what difference is today?

0:39:33.330,0:39:36.660
How is a difference in your[br]initial bid run of

0:39:36.660,0:39:39.630
the product and the audience[br]would be?

0:39:39.630,0:39:41.080
>> All right, we'll go first[br]on.

0:39:41.080,0:39:42.858
The people, actually all[br]three of you are welcome to

0:39:42.858,0:39:44.050
answer this, have the people[br]that you

0:39:44.050,0:39:46.960
hired early been able to[br]grow into leadership roles?

0:39:49.200,0:39:52.470
>> In the Stripes case, yes,[br]in that quite a number of

0:39:52.470,0:39:55.410
the first ten people are in[br]leadership roles now.

0:39:55.410,0:39:59.610
I think that's one thing[br]that organizations, again

0:39:59.610,0:40:02.548
it's an unnatural skill that[br]they have to get good at,

0:40:02.548,0:40:05.800
is realizing that people[br]don't necessarily come out

0:40:05.800,0:40:09.140
of the womb being good at[br]managing or being good at

0:40:09.140,0:40:13.420
leadership and being able to[br]develop that in people, and

0:40:13.420,0:40:16.510
being able to help people[br]progress as

0:40:16.510,0:40:18.880
they spend a number of years[br]at the company.

0:40:18.880,0:40:22.450
It's a lot of work at a time[br]when everyone is running

0:40:22.450,0:40:25.730
around with their hair on[br]fire, but it's also damaging

0:40:25.730,0:40:27.130
if the company can't develop[br]that skill.

0:40:30.160,0:40:30.740
>> Yeah, I think for

0:40:30.740,0:40:33.280
us the answer is some yes[br]and some no.

0:40:36.220,0:40:37.320
I think one of the big[br]benefits of

0:40:37.320,0:40:38.690
working at a start-up is[br]that you can be

0:40:38.690,0:40:40.310
handed a challenge that no[br]one else would be

0:40:40.310,0:40:43.340
crazy enough to give you the[br]opportunity to take on.

0:40:43.340,0:40:45.160
And that could managing[br]people.

0:40:45.160,0:40:46.970
It could be taking on a[br]project.

0:40:46.970,0:40:49.520
But the implicit contract[br]with that is that if

0:40:49.520,0:40:51.780
your going to ask somebody[br]to take a really big risk on

0:40:51.780,0:40:54.720
that, it shouldn't be like[br]one-way through the door and

0:40:54.720,0:40:55.380
if you don't succeed,

0:40:55.380,0:40:57.545
otherwise it creates fright[br]to give it a shot.

0:40:57.545,0:41:00.730
So we have some folks that[br]are managing large teams

0:41:00.730,0:41:03.800
that started as individual[br]programmers, where they were

0:41:03.800,0:41:05.680
engineering and they said[br]hey, I would love to try.

0:41:05.680,0:41:07.220
I'd love to try leading a[br]project and

0:41:07.220,0:41:07.970
then leading a group and

0:41:07.970,0:41:08.940
then taking responsibility[br]for

0:41:08.940,0:41:10.576
management, taking a group.

0:41:10.576,0:41:11.710
And we've had other folks[br]that tried it and

0:41:11.710,0:41:14.130
they were like, I'm really[br]glad I tried it because I

0:41:14.130,0:41:15.750
never wanna do that again.

0:41:15.750,0:41:17.740
We try to make sure that,[br]for those people, like you

0:41:17.740,0:41:20.660
can have just as much impact[br]at a company through your

0:41:20.660,0:41:22.510
individual contributions as[br]an engineer or designer.

0:41:22.510,0:41:23.130
You don't have to manage.

0:41:24.810,0:41:25.640
But it's really hard to

0:41:25.640,0:41:27.102
predict until you give[br]people a shot.

0:41:27.102,0:41:30.800
And so my strong preference[br]is that you give as many

0:41:30.800,0:41:34.140
people a shot as possible,[br]in the few areas where you

0:41:34.140,0:41:38.070
really feel like there's too[br]much learning curve relative

0:41:38.070,0:41:39.630
to the business objective[br]you're trying to achieve.

0:41:39.630,0:41:42.750
That's when you look for[br]somebody who might be able

0:41:42.750,0:41:44.470
to walk in and really[br]execute well in that job.

0:41:46.420,0:41:48.910
Are you going to answer the[br]other question?

0:41:48.910,0:41:50.100
>> Oh, so the question was

0:41:50.100,0:41:52.170
how has the vision changed[br]since we initially started?

0:41:52.170,0:41:59.364
>> Yes,[br]>> Sure,

0:41:59.364,0:42:04.260
well I think on the vision,[br]when we first started,

0:42:04.260,0:42:06.760
I think we started hiring[br]very inductively.

0:42:06.760,0:42:07.300
We were like, oh, we're

0:42:07.300,0:42:08.420
gonna build this really cool[br]tool.

0:42:08.420,0:42:09.740
People are gonna enjoy it.

0:42:09.740,0:42:10.690
I love collecting things.

0:42:10.690,0:42:12.670
Maybe other people like[br]collecting things.

0:42:12.670,0:42:14.100
And what we didn't expect,

0:42:14.100,0:42:15.790
that kind of revealed itself[br]early on,

0:42:15.790,0:42:19.750
was that looking at other[br]people's collections turns

0:42:19.750,0:42:21.460
out to be this really[br]amazing way to discover

0:42:21.460,0:42:23.730
things that you didn't know[br]you were looking for.

0:42:23.730,0:42:25.839
It becomes kind of a[br]solution to a problem that

0:42:25.839,0:42:27.790
a lot of other technologies[br]don't have.

0:42:27.790,0:42:30.980
And so, over the last couple[br]years especially,

0:42:30.980,0:42:33.340
we've poured enormous kind[br]of technical and design

0:42:33.340,0:42:36.600
resources into building out[br]recommendations products,

0:42:36.600,0:42:38.486
search products, feed[br]products.

0:42:38.486,0:42:41.520
Leveraging the unique data[br]that we have which

0:42:41.520,0:42:43.505
are these pins that were all[br]hand-picked by someone and

0:42:43.505,0:42:45.190
hand-categorized.

0:42:45.190,0:42:47.980
And then on the audience[br]side, the first big

0:42:47.980,0:42:51.140
surprise was truthfully when[br]we first started,

0:42:51.140,0:42:52.610
we didn't really know if[br]anyone would use it.

0:42:52.610,0:42:55.890
And we were just happy that[br]anybody that wasn't related

0:42:55.890,0:42:56.570
to us and

0:42:56.570,0:42:59.930
obligated by familial[br]relationship would use it.

0:42:59.930,0:43:03.190
And so one of the biggest[br]surprises has been how many

0:43:03.190,0:43:05.662
people and how diverse those[br]groups of people have been.

0:43:05.662,0:43:07.235
I think that's been one of

0:43:07.235,0:43:09.554
the things that's really[br]exciting.

0:43:09.554,0:43:11.290
And, the funny thing is,

0:43:11.290,0:43:14.460
that is often like the[br]company goes farther along,

0:43:14.460,0:43:16.162
your aspirations therefore[br]get bigger.

0:43:16.162,0:43:19.440
So there's this gap that[br]always exists,

0:43:19.440,0:43:20.840
I tell my team, between[br]where we are and

0:43:20.840,0:43:21.910
where I think we should be,[br]and

0:43:21.910,0:43:24.100
even though objectively,[br]we're much farther along,

0:43:24.100,0:43:26.390
I feel like the gap has[br]widened even farther.

0:43:26.390,0:43:30.290
But I think that's a really[br]common trait among

0:43:30.290,0:43:32.795
people who found companies.

0:43:32.795,0:43:34.197
>> Yes?

0:43:34.197,0:43:35.348
>> You said that like,

0:43:35.348,0:43:37.986
while selling a vision of[br]the company you had to

0:43:37.986,0:43:38.731
describe and

0:43:38.731,0:43:41.340
go into detail that how hard[br]it is going to be.

0:43:41.340,0:43:43.290
And like they won't see[br]their families for

0:43:43.290,0:43:45.650
three years but then they[br]will also get a thing that

0:43:45.650,0:43:47.610
their grandchildren can be[br]proud of.

0:43:47.610,0:43:49.050
But how do you really know?

0:43:49.050,0:43:51.010
Both of you know that the[br]second part is

0:43:52.180,0:43:53.680
not guaranteed but the first[br]part,

0:43:53.680,0:43:55.200
if they're not gonna see[br]their family for

0:43:55.200,0:43:57.380
the next year, it's gonna be[br]a different show.

0:43:57.380,0:43:59.800
So like, how can you be[br]authentic by selling that

0:43:59.800,0:44:02.514
vision of the company or is[br]it about giving high returns

0:44:02.514,0:44:04.992
for the high risk they are[br]taking when the number of

0:44:04.992,0:44:08.331
equity that already[br]>> Great question.

0:44:08.331,0:44:09.565
So the question is,

0:44:09.565,0:44:12.090
most start-ups are not the[br]iPhone.

0:44:12.090,0:44:13.810
You can't guarantee that[br]people's grandchildren

0:44:13.810,0:44:15.820
are gonna remember this[br]because most start-ups fail.

0:44:15.820,0:44:17.180
How do you convince people[br]to sort of

0:44:17.180,0:44:19.055
make sacrifices to join a[br]start-up?

0:44:21.510,0:44:25.584
>> I think, part of the way[br]in which it resonates with

0:44:25.584,0:44:29.660
people is because it's not[br]guaranteed, right?

0:44:29.660,0:44:31.720
If it were guaranteed then[br]it'd be boring.

0:44:31.720,0:44:36.784
And so it's that there is[br]the prospect of affecting

0:44:36.784,0:44:42.219
this outcome, but nothing[br]more than that potential.

0:44:42.219,0:44:45.236
To the not seeing their[br]families or

0:44:45.236,0:44:49.709
kids, start-ups often do[br]involve longer hours in

0:44:49.709,0:44:52.945
the beginning, but I think,[br]well.

0:44:52.945,0:44:55.758
I, I think that that[br]particular story is

0:44:55.758,0:44:58.160
probably somewhat[br]overstated.

0:44:58.160,0:44:59.650
I guess, it was, I think

0:44:59.650,0:45:02.490
Scott Forestall was trying[br]to recruit these people.

0:45:02.490,0:45:05.830
But I mean, even though[br]startups,

0:45:05.830,0:45:06.790
especially in the earlier[br]days,

0:45:06.790,0:45:08.830
tend to involve somewhat[br]longer working hours.

0:45:08.830,0:45:10.950
I think it was kind of this[br]tendency to,

0:45:10.950,0:45:11.715
to exaggerate it.

0:45:11.715,0:45:12.320
And sort of,

0:45:12.320,0:45:14.550
I mean it's like the startup[br]version of fishing.

0:45:14.550,0:45:17.460
Like, every startup thinks[br]they worked even more insane

0:45:17.460,0:45:18.570
hours than, you know,

0:45:18.570,0:45:20.530
the next one back in the[br]early days.

0:45:20.530,0:45:23.450
Like, we literally never[br]sleep or slept rather,

0:45:23.450,0:45:24.810
for two years.

0:45:24.810,0:45:27.330
So I don't know.

0:45:27.330,0:45:29.980
I think that realistically[br]for most people,

0:45:29.980,0:45:32.350
it's not that big a[br]sacrifice, right?

0:45:32.350,0:45:35.710
You're, maybe on average,[br]being really realistic about

0:45:35.710,0:45:37.842
it, you'll work two hours[br]more on average per day.

0:45:37.842,0:45:39.570
It's certainly a sacrifice,[br]but

0:45:39.570,0:45:44.770
it is not forgoing all you[br]know, pleasure and

0:45:44.770,0:45:47.820
enjoyment for the next half[br]decade.

0:45:47.820,0:45:49.450
>> Yeah, I mean, I think[br]even the iPhone

0:45:49.450,0:45:53.600
wasn't the iPhone before it[br]got done, right?

0:45:53.600,0:45:55.890
I mean, no smart person[br]you're hiring is

0:45:55.890,0:45:57.890
under the illusion that you[br]have a crystal ball into

0:45:57.890,0:46:00.170
the future that only you[br]have, and

0:46:00.170,0:46:01.550
that joining is a guaranteed[br]thing.

0:46:01.550,0:46:03.590
And in fact, if you're[br]telling him that and

0:46:03.590,0:46:06.520
they select in, you probably[br]shouldn't hire them.

0:46:06.520,0:46:07.452
Because they,

0:46:07.452,0:46:10.394
they didn't pass like a[br]basic intelligence test

0:46:10.394,0:46:12.992
about uncertainty in the[br]future.

0:46:12.992,0:46:15.436
But I, but I think it's fair[br]to say, like you know,

0:46:15.436,0:46:17.883
what's exciting and where[br]you think you can go, and

0:46:17.883,0:46:20.920
where it's going to be hard[br]and chart your best plan.

0:46:20.920,0:46:24.144
And then tell them why their[br]role in it can be

0:46:24.144,0:46:26.930
instrumental, because it is.

0:46:26.930,0:46:29.090
You know I really liked what[br]you said.

0:46:29.090,0:46:31.340
You know, if you tell people[br]like hey we're gonna go

0:46:31.340,0:46:33.750
to Mars, it attracts the[br]best people and

0:46:33.750,0:46:36.250
then you're incrementally[br]closer to getting to Mars.

0:46:36.250,0:46:37.960
And they know that going in.

0:46:37.960,0:46:39.630
What I would discourage[br]doing is

0:46:39.630,0:46:40.720
just whitewashing all of[br]that.

0:46:42.010,0:46:45.250
And if people kind of are[br]joining because they want

0:46:45.250,0:46:47.200
sort of oh, I want all this[br]certainty and guarantee of

0:46:47.200,0:46:49.520
working at Google, plus like[br]the perk of working in

0:46:49.520,0:46:52.560
a small start up and more[br]email, and transparency.

0:46:52.560,0:46:53.440
Like that's, that's a[br]really,

0:46:53.440,0:46:54.365
really negative sign.

0:46:54.365,0:46:57.210
And for

0:46:57.210,0:47:00.330
example, when I interview[br]people they'll often say oh,

0:47:00.330,0:47:01.620
I'm really passionate at[br]what you're doing, and

0:47:01.620,0:47:03.030
then I'm like well, what[br]else are you interviewing?

0:47:03.030,0:47:05.340
And then they'll just list[br]seven companies that have

0:47:05.340,0:47:07.580
nothing to do with each[br]other except they're sort of

0:47:07.580,0:47:08.861
at the same stage we're at.

0:47:08.861,0:47:09.916
They're like, you know,

0:47:09.916,0:47:11.471
I love the problem of[br]discovery, so

0:47:11.471,0:47:15.060
I'm interviewing at Stripe,[br]Dropbox, Airbnb, Uber.

0:47:15.060,0:47:17.200
Now I'm also putting into my[br]resumé into Google X,

0:47:17.200,0:47:18.730
into that part of the[br]division.

0:47:18.730,0:47:19.960
And that's a sign that[br]they're probably not

0:47:19.960,0:47:21.420
being authentic with which[br]they care about.

0:47:21.420,0:47:21.940
>> Yeah.

0:47:21.940,0:47:24.660
>> And those folks often,[br]when things get really hard,

0:47:24.660,0:47:26.855
they won't stick it out and[br]work through it.

0:47:26.855,0:47:28.480
Because they were really[br]signing up for

0:47:28.480,0:47:31.520
an experience, not for[br]achieving a goal.

0:47:31.520,0:47:34.650
>> I think the other thing[br]that motivates people

0:47:34.650,0:47:37.650
a great deal, in addition to[br]the prospect of sort of

0:47:37.650,0:47:40.797
them you know, affecting[br]some outcome is just sort of

0:47:40.797,0:47:42.680
the personal development[br]angle.

0:47:42.680,0:47:43.340
And that a start up,

0:47:43.340,0:47:47.170
just because it's much, kind[br]of more lightly staffed.

0:47:48.920,0:47:50.280
It's much less forgiving,[br]right?

0:47:50.280,0:47:51.038
In that, like,

0:47:51.038,0:47:53.099
even if you're the best[br]person in the world,

0:47:53.099,0:47:54.249
if you're not going to.

0:47:54.249,0:47:55.687
Well, whether or not you're[br]the best or

0:47:55.687,0:47:57.550
the worst person in the[br]world, you're probably not

0:47:57.550,0:47:59.835
going to significantly alter[br]Google's trajectory, right?

0:47:59.835,0:48:02.860
Whereas if you sort of,

0:48:02.860,0:48:05.810
really wanted to benchmark[br]yourself, and

0:48:05.810,0:48:08.880
see how much a contribution[br]and an impact you can make.

0:48:08.880,0:48:10.990
And I think that prospect is[br]quite compelling to a lot of

0:48:10.990,0:48:13.350
the best people, then a[br]start up is

0:48:13.350,0:48:18.120
a much better place to, to[br]go test that.

0:48:18.120,0:48:18.660
>> All right.

0:48:18.660,0:48:20.850
>> Last question, who'd you[br]want?

0:48:20.850,0:48:22.280
Last question.

0:48:22.280,0:48:23.670
So this question is for

0:48:23.670,0:48:25.880
Ben, but you guys can all[br]answer it.

0:48:25.880,0:48:28.590
But how has your user base[br]effected your

0:48:28.590,0:48:29.540
hiring strategies?

0:48:29.540,0:48:32.759
So Pinterest is a site[br]that's used like 80% by

0:48:32.759,0:48:33.569
women, so

0:48:33.569,0:48:36.700
how did that effect your[br]initial hiring decisions?

0:48:36.700,0:48:38.020
>> How has your user base[br]effected your

0:48:38.020,0:48:38.560
hiring strategy?

0:48:40.340,0:48:41.690
>> Yes, so, You know,

0:48:41.690,0:48:44.040
conventional wisdom is like[br]you only hire people that

0:48:44.040,0:48:46.820
religiously use your product[br]every single day.

0:48:46.820,0:48:48.010
And that probably works[br]really well if

0:48:48.010,0:48:50.272
you're making an API,[br]probably amazing.

0:48:50.272,0:48:54.930
For us, we screen for people[br]that are ambitious and

0:48:54.930,0:48:58.780
excited about the vision of[br]cracking discovery online.

0:48:58.780,0:49:00.900
And they have to know[br]exactly how

0:49:00.900,0:49:02.720
our service works and have[br]to have used it.

0:49:02.720,0:49:04.490
But they may not be a[br]lifelong user.

0:49:04.490,0:49:06.770
And that, for us it's this[br]great opportunity because we

0:49:06.770,0:49:08.700
can be like, what is the[br]barrier that's

0:49:08.700,0:49:10.270
preventing you from using[br]it?

0:49:10.270,0:49:12.100
Come join, remove that[br]barrier,

0:49:12.100,0:49:15.303
and help us get closer to[br]that vision.

0:49:15.303,0:49:17.860
I don't know, there's a lot[br]of,

0:49:19.000,0:49:20.420
if you read any sort of[br]book, there's all this

0:49:20.420,0:49:23.050
startup wisdom that sounds[br]like really reasonable.

0:49:23.050,0:49:25.200
But it's only useful if it[br]works in

0:49:25.200,0:49:27.060
your particular[br]circumstance.

0:49:27.060,0:49:28.340
And so for us, we've had to[br]sort of

0:49:28.340,0:49:31.360
broaden the lens a little[br]bit in looking broadly about

0:49:31.360,0:49:33.169
people that are ambitious[br]about the mission.

0:49:33.169,0:49:34.620
That care about the product[br]and

0:49:34.620,0:49:36.290
our approach to building[br]products,

0:49:36.290,0:49:38.509
even if from day one, they[br]weren't our earliest users.

0:49:40.040,0:49:44.140
>> The one thing I'll just[br]tack on to that is that when

0:49:44.140,0:49:45.830
we talked about the fact[br]that, you know,

0:49:45.830,0:49:48.500
it's really hard to hire for[br]those early employees.

0:49:48.500,0:49:49.760
And you know, you have[br]people who have

0:49:49.760,0:49:50.880
other good options.

0:49:52.230,0:49:54.890
You're very much at the ugly[br]duckling stage.

0:49:54.890,0:49:56.714
Finding people who are[br]passionate about your

0:49:56.714,0:49:58.399
product can be a great way[br]to find people.

0:49:58.399,0:50:01.888
Because there you kind of[br]have an unnatural advantage

0:50:01.888,0:50:03.443
over other companies.

0:50:03.443,0:50:04.784
And so I know, for

0:50:04.784,0:50:09.258
sure, in Stripe's case, we[br]definitely, we hired I think

0:50:09.258,0:50:13.690
it was four Stripe users in[br]the very early days.

0:50:13.690,0:50:16.150
And you know, those were[br]people who,

0:50:16.150,0:50:18.540
who we probably couldn't[br]have gotten otherwise.

0:50:18.540,0:50:21.196
I'm sure it was the same in[br]Pinterest's case,

0:50:21.196,0:50:23.830
where you'll get all this[br]benefit of working at

0:50:23.830,0:50:25.630
Pinterest and hey, it's[br]Pinterest.

0:50:25.630,0:50:27.170
And you get to work where[br]you pin.

0:50:27.170,0:50:29.469
Yeah.

0:50:29.469,0:50:30.618
>> Thank you guys very much[br]for coming in today.

0:50:30.618,0:00:00.000
>> Thank you.