0:00:00.530,0:00:01.109
Хорошо.

0:00:01.110,0:00:05.800
Итак, когда я говорю о создании продукта, который пользователи полюбят, я имею ввиду,

0:00:05.800,0:00:09.500
то, как мы делаем что-то, вызывающее[br]интерес у людей.

0:00:09.500,0:00:12.800
То, чему наши пользователи

0:00:12.800,0:00:16.950
будут неосознанно желать

0:00:16.950,0:00:19.590
успеха. И продуции, который мы производим,[br]

0:00:19.590,0:00:21.730
и компаниям, стоящим за этой продукцией

0:00:22.400,0:00:25.340
Мы изучим огромное количество материала

0:00:25.340,0:00:27.259
Попытайтесь не делать слишком много записей

0:00:27.260,0:00:28.290
по большей части, слушайте

0:00:28.290,0:00:30.680
Я покажу ссылку на слайды в своем твиттере

0:00:30.690,0:00:33.100
и по этой ссылке вы сможете получить

0:00:33.100,0:00:35.320
аннотации к слайдам и задать мне вопросы

0:00:35.320,0:00:36.840
И, если мы до них не доберемся

0:00:36.840,0:00:38.570
Я отвечу уже после лекции

0:00:39.120,0:00:41.459
Итак, вы, ребята слушали, и

0:00:41.459,0:00:45.290
слышали о показателе роста,[br]за последние две недели, и

0:00:45.300,0:00:48.500
как по мне, понятие "рост", довольно [br]простое

0:00:48.500,0:00:51.390
Это взаимодействие между двумя концепциями, или переменными

0:00:51.390,0:00:54.430
Коэффициентом конверсии и оттоком [br]пользователей.

0:00:54.430,0:00:55.100
понятно?

0:00:55.100,0:00:57.950
И промежуток между этими двумя понятиями [br]довольно

0:00:57.950,0:01:00.920
неплохо показывает, как быстро вы будете[br]развиваться

0:01:00.920,0:01:04.190
Многие люди, особенно, те, кто [br]занимается бизнесом

0:01:04.190,0:01:06.289
привыкли оценивать это взаимодействие[br]через понятия

0:01:06.290,0:01:08.980
хорошо поддающихся вычислению

0:01:08.980,0:01:09.890
И сегодня, я хочу

0:01:09.890,0:01:10.950
рассказать об этом

0:01:10.950,0:01:12.179
более простым языком.

0:01:12.180,0:01:14.720
Потому что, в стартапе, когда вы [br]взаимодействуете

0:01:14.720,0:01:16.700
со своими пользователями, вы[br]делаете это

0:01:16.700,0:01:19.100
довольно аккуратно, на первых[br]порах

0:01:19.100,0:01:21.200
в связи с этим, я думаю, есть [br]иной способ взглянуть

0:01:21.200,0:01:23.200
на эти вещи, с той точки, что [br]мы создаём что-то своё

0:01:23.200,0:01:25.800
И мы смотрим на большое количество [br]различных примеров этого.

0:01:25.800,0:01:26.190
И как это работает.

0:01:27.500,0:01:31.300
My philosophy behind a lot[br]of things that I teach

0:01:31.300,0:01:33.430
startups is the best way to[br]sort of get to a billion

0:01:33.430,0:01:36.000
dollars is to focus on the[br]values that help you

0:01:36.000,0:01:38.870
get that first dollar, to[br]acquire that first user.

0:01:38.870,0:01:41.800
If you sort of get that[br]right everything else will

0:01:41.800,0:01:43.270
sort of take care of itself.

0:01:43.270,0:01:45.190
It's a sort of faith thing.

0:01:45.190,0:01:49.770
So I came to be a partner at[br]YC by way of being alumni.

0:01:49.770,0:01:51.800
I went to the program in[br]winter of 2006, so

0:01:51.800,0:01:53.890
it's the second ever program[br]and

0:01:53.890,0:01:55.420
I built a product called[br]Wufoo.

0:01:55.420,0:01:57.500
Wufoo's an online forum[br]builder.

0:01:57.500,0:01:58.830
It helps you create contact[br]forums and

0:01:58.830,0:02:01.170
online surveys and simple[br]payment forms.

0:02:01.170,0:02:03.350
It's basically a database[br]that looks like it's

0:02:03.350,0:02:05.890
designed by Fisher Price.

0:02:05.900,0:02:05.810
What's.

0:02:05.810,0:02:09.449
Interesting, though, is that[br]because it was fairly easy

0:02:09.449,0:02:13.290
to use, we're had customers[br]from every industry, market,

0:02:13.290,0:02:15.709
and vertical you can think[br]of, including a,

0:02:15.710,0:02:18.270
a majority of the Fortune[br]500 companies out there.

0:02:18.270,0:02:21.430
Ran the company for five[br]years, and

0:02:21.430,0:02:24.700
then we were acquired by[br]SurveyMonkey in 2013.

0:02:24.700,0:02:26.420
And at the time,

0:02:26.420,0:02:28.500
we were a very interesting[br]acquisition.

0:02:28.500,0:02:30.460
We were only a team of ten[br]people at the time.

0:02:30.460,0:02:32.680
And, while we acquired[br]funding out here in

0:02:32.680,0:02:35.820
Silicon Valley through Y[br]Combinator we actually ran

0:02:35.820,0:02:37.420
the company from Florida.

0:02:37.420,0:02:39.329
We had no office.

0:02:39.330,0:02:40.470
Everyone worked from home.

0:02:41.670,0:02:44.160
And we're an interesting[br]outlier.

0:02:44.160,0:02:47.940
So, each dot here represents[br]a start up that was,

0:02:47.940,0:02:50.400
that exited through IPO or[br]acquisition.

0:02:50.400,0:02:51.760
And we're this outlier to[br]the left.

0:02:51.760,0:02:53.690
The bottom is the funding[br]amount they took.

0:02:53.690,0:02:55.530
And the vertical axis is,

0:02:55.530,0:02:57.440
the valuation of the company[br]at the time.

0:02:57.440,0:02:58.170
To sum it up,

0:02:58.170,0:03:01.684
the average startup rates as[br]about $25 million.

0:03:01.685,0:03:04.930
And they return to their[br]investors about 676%.

0:03:04.930,0:03:09.440
Wufoo raised about $118[br]thousand total.

0:03:09.440,0:03:13.830
And our return to our[br]investors is about 29,000%.

0:03:13.830,0:03:14.810
>> So a lot of people were[br]very interested.

0:03:14.810,0:03:15.800
And are sort of like.

0:03:15.800,0:03:16.395
What makes.

0:03:16.395,0:03:17.510
Wufoo a little bit different[br]or

0:03:17.510,0:03:19.590
how do we run the company[br]very differently, and

0:03:19.590,0:03:20.910
a lot of it was focused on[br]product.

0:03:22.540,0:03:24.648
We weren't interested in[br]building software that,

0:03:24.648,0:03:28.910
I guess that just people[br]wanted to use.

0:03:28.910,0:03:30.700
Right.[br]That reminding you that you

0:03:30.700,0:03:30.989
worked in a cubicle,

0:03:30.990,0:03:33.560
cuz it was database app at[br]its sort of core.

0:03:33.560,0:03:36.560
We wanted a product that[br]people wanted to love.

0:03:36.560,0:03:38.600
That people wanted to have a[br]relationship with, and we

0:03:38.600,0:03:41.920
were actually very fanatical[br]about how we approached.

0:03:41.920,0:03:43.570
This idea.

0:03:43.570,0:03:44.930
At the point where it's[br]almost sort of

0:03:46.920,0:03:48.730
sort of sciency sort of way.

0:03:50.170,0:03:52.440
So what we said was like[br]Okay.

0:03:52.440,0:03:55.300
What's interesting about[br]start-ups in terms of

0:03:55.300,0:03:58.580
us wanting to create things[br]that people love is that,

0:03:58.580,0:04:00.930
love and unconditional sort[br]of feelings are things that

0:04:00.930,0:04:02.919
are difficult for us to do[br]in sort of real life.

0:04:03.990,0:04:06.580
And at start-ups we have to[br]do it sort of at scale.

0:04:06.580,0:04:07.590
So we've decided to do is

0:04:07.590,0:04:09.510
just start of by just[br]looking at, Okay,

0:04:09.510,0:04:12.750
how does real relationships[br]work in the real world, and

0:04:12.750,0:04:14.390
how can we apply them to[br]sort of how we

0:04:14.390,0:04:17.630
run our business and sort of[br]build our product that way.

0:04:17.630,0:04:20.399
So we'll go over basically[br]these two metaphors,

0:04:20.399,0:04:22.190
find new users as if we're[br]trying to date them and

0:04:22.190,0:04:25.520
existing users as if it were[br]a successful marriage.

0:04:25.520,0:04:28.460
So when it comes to dating,[br]lot of the stuff that

0:04:28.460,0:04:31.630
we uncovered had to do with[br]first impressions.

0:04:33.160,0:04:36.270
all, all of you often talk[br]about your relationships in

0:04:36.270,0:04:37.440
terms of the origin story.

0:04:37.440,0:04:38.950
If I asked you to tell me[br]about the first kiss,

0:04:38.950,0:04:40.760
how you sort of met, how you[br]proposed,

0:04:40.760,0:04:42.530
these are the things that we[br]say over and

0:04:42.530,0:04:43.250
over and over again.

0:04:43.250,0:04:45.580
They're basically the[br]word-of-mouth stories of our

0:04:45.580,0:04:47.590
relationships, and they're[br]the same kinds of

0:04:47.590,0:04:49.989
things that we do with[br]companies.

0:04:49.990,0:04:50.560
Human beings

0:04:50.560,0:04:52.260
are relationship[br]manufacturing creatures.

0:04:52.260,0:04:56.440
We cannot help, but create[br]and anthropomorphize

0:04:56.440,0:04:59.590
the things we interact with[br]over and over again.

0:04:59.590,0:05:01.200
So, whether it's the cars we[br]drive,

0:05:01.200,0:05:04.219
the clothes we wear, the[br]tools and software's we use,

0:05:04.220,0:05:06.470
we eventually prescribe[br]characteristics to it.

0:05:06.470,0:05:07.250
A personality.

0:05:07.250,0:05:08.570
And we expect it to behave a[br]certain way.

0:05:08.570,0:05:11.349
And that's how we sort of[br]interact with it.

0:05:11.350,0:05:13.140
Now first impressions are[br]important for

0:05:13.140,0:05:14.760
the starting of any[br]relationship,

0:05:14.760,0:05:16.420
because it's the one we tell[br]over and over again.

0:05:16.420,0:05:16.960
Right?

0:05:16.960,0:05:19.710
And there's something[br]special about how we

0:05:19.710,0:05:20.909
regard that origin story.

0:05:20.910,0:05:22.300
I'll give an example.

0:05:22.300,0:05:24.100
If you're on a first date[br]with somebody,

0:05:24.100,0:05:26.900
and you're having a nice[br]dinner, and

0:05:26.900,0:05:28.510
you catch them picking their[br]nose.

0:05:28.510,0:05:31.930
You are probably not gonna[br]have another date with them.

0:05:31.930,0:05:33.440
Right, but if you're married[br]to someone for

0:05:33.440,0:05:34.130
about 20 to 30 years,

0:05:34.130,0:05:37.560
and you catch on the Barca[br]Lounger digging for

0:05:37.560,0:05:38.640
gold right?

0:05:38.640,0:05:40.380
You don't immediately like[br]call your lawyer,

0:05:40.380,0:05:42.659
right, and then say like we[br]have a problem here.

0:05:42.660,0:05:45.310
I need to start drawing up[br]papers for divorce.

0:05:45.310,0:05:46.850
You shrug your shoulders and

0:05:46.850,0:05:48.510
say at least he has a heart[br]of gold.

0:05:49.560,0:05:51.470
So solving about.

0:05:51.470,0:05:54.740
First time interactions[br]means that the threshold is

0:05:54.740,0:05:58.150
so much lower in terms of[br]pass fail.

0:05:58.150,0:06:00.590
So, in software, and for

0:06:00.590,0:06:03.250
most products in internet[br]software that we use, like,

0:06:03.250,0:06:05.290
first impressions are pretty[br]obvious.

0:06:05.290,0:06:06.420
And they're the things that[br]you see a lot of

0:06:06.420,0:06:07.890
companies sort of pay[br]attention to in terms of

0:06:07.890,0:06:10.409
what they send their[br]marketing people to work on.

0:06:10.410,0:06:13.830
My argument for people who[br]are very good at product is

0:06:13.830,0:06:16.300
they discover so many other[br]first moments, and

0:06:16.300,0:06:18.890
they make those something[br]memorable.

0:06:18.890,0:06:21.990
Right, the first email you[br]ever get from a piece of

0:06:21.100,0:06:21.760
software.

0:06:21.760,0:06:24.900
What happens when you first[br]log in?

0:06:24.900,0:06:26.479
The links, the advertising,[br]the very first time you

0:06:26.480,0:06:27.730
interact with customer[br]support.

0:06:27.730,0:06:31.940
All of those are[br]opportunities to seduce.

0:06:31.940,0:06:33.890
So how did we think about.

0:06:33.890,0:06:35.849
Sort of like making first[br]moments on there,

0:06:35.850,0:06:38.610
and we actually took this[br]concept from the Japanese.

0:06:38.610,0:06:41.120
They actually have two words[br]for

0:06:41.120,0:06:43.200
how to describe things when[br]you're finished with them

0:06:43.200,0:06:45.539
in terms of saying like is[br]this a quality item.

0:06:45.540,0:06:47.396
And the two words of quality[br]are atarimae hinshitsu and

0:06:47.396,0:06:48.290
miryokuteki hinshitsu.

0:06:48.290,0:06:51.470
And the first one means,[br]taken for

0:06:51.470,0:06:53.200
granted quality, basically[br]functionality.

0:06:53.200,0:06:55.120
And the last one sort of[br]means,

0:06:55.120,0:06:56.550
enchanting quality, right?

0:06:58.100,0:06:59.750
Take for example a pen,[br]right?

0:06:59.750,0:07:04.170
Something has miryokuteki,[br]right, if the weight of

0:07:04.170,0:07:06.220
the pen, the way the ink[br]flows out of it, the way

0:07:06.220,0:07:08.960
it's viewed by the people[br]reading the handwriting from

0:07:08.960,0:07:12.200
the pen is pleasurable both[br]to the user of the pen and

0:07:12.200,0:07:15.109
the people who experience[br]the byproducts of it,

0:07:15.110,0:07:16.840
right, taking it to the sort[br]of next level.

0:07:19.150,0:07:20.150
Start with some examples.

0:07:20.150,0:07:23.429
So this is Wufoo's Login[br]link, and

0:07:23.430,0:07:26.280
it has a dinosaur on it,[br]which I think is awesome.

0:07:26.280,0:07:29.800
But if you hover over it,

0:07:29.800,0:07:32.600
the spec has the added[br]benefit of having a tool tip

0:07:32.600,0:07:33.690
that doesn't explain like[br]how to login or

0:07:33.690,0:07:36.260
what it does, but basically,[br]rawr.

0:07:36.260,0:07:37.800
And.

0:07:37.800,0:07:39.890
What we noticed about this,[br]like in early usability

0:07:39.890,0:07:42.280
studies as like this put a[br]smile on people's faces.

0:07:42.280,0:07:43.630
Like, hands down.

0:07:43.630,0:07:44.170
Right?

0:07:44.170,0:07:45.300
Universally.

0:07:45.300,0:07:46.640
And I think a lot of times[br]when we

0:07:46.640,0:07:48.530
are assessing products, we[br]never think about,

0:07:48.530,0:07:50.190
like, hey what is the[br]emotion on

0:07:50.190,0:07:52.490
the person's face when they[br]interact with this?

0:07:53.580,0:07:55.500
This is Vimeo's log in page[br]this is

0:07:55.500,0:07:57.700
actually a couple integr,[br]iterations ago.

0:07:57.700,0:07:59.540
It's one I find to be the[br]most beautiful, but.

0:08:00.570,0:08:02.200
It lets you know that when[br]you're starting out on

0:08:02.200,0:08:03.180
this journey with Vimeo,

0:08:03.180,0:08:04.640
that this is gonna be[br]something different.

0:08:05.820,0:08:07.170
They do this all over the[br]app.

0:08:07.170,0:08:09.810
If you search for the word[br]fart, as you scroll up and

0:08:09.810,0:08:12.190
down it makes fart noises as[br]you do this.

0:08:12.190,0:08:12.730
Right?

0:08:12.730,0:08:13.920
There's something different,

0:08:13.920,0:08:15.705
like this site interacts[br]with you.

0:08:15.705,0:08:16.289
>> Mm-hm.[br]>> It's a little bit

0:08:16.290,0:08:17.530
magical, it's a little bit[br]different.

0:08:17.530,0:08:19.510
And it's something that you[br]wanna talk about.

0:08:20.940,0:08:23.600
You don't have to always do[br]it with design.

0:08:23.600,0:08:24.700
This is a sign up form for

0:08:24.700,0:08:26.430
Cork which used to be a[br]social network for

0:08:26.430,0:08:28.700
people who loved to drink[br]wine.

0:08:28.700,0:08:31.190
On it, it says email[br]address,

0:08:31.190,0:08:33.240
it's also your sign-in and[br]has to be legit.

0:08:33.240,0:08:35.500
First name, what your mom[br]calls you, last name,

0:08:35.500,0:08:37.260
what your Army buddies call[br]you.

0:08:37.260,0:08:39.799
Password, something you'll[br]remember, but hard to guess,

0:08:39.799,0:08:42.339
password confirmation, think[br]it a, type it again.

0:08:42.340,0:08:43.558
Think of it as a test.

0:08:43.558,0:08:46.699
It's literally a poem as you[br]fill out the form.

0:08:46.700,0:08:47.540
Right?

0:08:47.540,0:08:50.689
And this is a kind of like[br]thing where you like,

0:08:50.690,0:08:52.110
oh, I like the people behind[br]this.

0:08:52.110,0:08:53.850
I, I, I'm gonna enjoy this[br]experience.

0:08:53.850,0:08:55.970
Now what does it say when[br]you fill out form like this?

0:08:55.970,0:08:56.470
Right.

0:08:56.470,0:08:58.960
On Yahoo, about what the[br]personality.

0:08:58.960,0:08:59.890
Of the site it's gonna be.

0:08:59.890,0:09:02.500
And what's disappointing to[br]me is like,

0:09:02.500,0:09:04.570
Yahoo forces every product[br]and

0:09:04.570,0:09:07.960
service under them to use[br]this exact same login form.

0:09:07.960,0:09:09.450
Flickr, I had thought,

0:09:09.450,0:09:11.280
had one of the best sort of[br]call to actions.

0:09:11.280,0:09:13.699
It was, get in there.

0:09:13.700,0:09:14.200
Right?

0:09:16.200,0:09:17.259
This is Heroku's.

0:09:19.900,0:09:19.650
Sign up page.

0:09:19.650,0:09:21.230
I think this is an older[br]version.

0:09:21.230,0:09:24.340
But, what's remarkable about[br]it is that, what you

0:09:24.340,0:09:28.180
start getting a feel for is[br]like, oh, scaling up.

0:09:28.180,0:09:29.000
My sort of server, and

0:09:29.000,0:09:33.190
back end services is as easy[br]as sort of dragging up and

0:09:33.190,0:09:36.460
down, different sort of nobs[br]and levers.

0:09:36.460,0:09:37.820
It's gonna be beautifully[br]used, and

0:09:37.820,0:09:39.570
it looks fairly easy to[br]scale.

0:09:41.260,0:09:45.460
Since we're in a room full[br]of computer science people,

0:09:45.460,0:09:46.880
I think you'll appreciate[br]this.

0:09:46.880,0:09:47.660
This is Chocolat.

0:09:47.660,0:09:49.750
This is a code editor, and

0:09:49.750,0:09:51.320
they only have one call to[br]action.

0:09:51.320,0:09:55.270
When the time limit is up,[br]they say, everything in

0:09:55.270,0:09:58.500
terms of all the pieces are[br]exactly the same except we

0:09:58.500,0:09:59.599
change the font.

0:09:59.600,0:10:00.600
To comic sans.

0:10:03.670,0:10:06.540
And what they're basically[br]saying is like hey

0:10:06.540,0:10:09.932
we know who our users are,[br]who our real customers are.

0:10:09.932,0:10:12.300
They're gonna be the people[br]who care about this.

0:10:15.470,0:10:17.480
This is Hurl, this is a[br]website for

0:10:17.480,0:10:22.580
checking htp requests, and[br]sometimes the places where

0:10:22.580,0:10:24.650
you get errors are[br]opportunities for

0:10:24.650,0:10:26.526
first moments,

0:10:26.527,0:10:29.390
you had a four of four this[br]is what you get it,

0:10:29.390,0:10:32.400
when we need help oftentimes[br]what we do is where we

0:10:32.400,0:10:39.770
create like really beautiful[br]mark, marketing materials.

0:10:39.770,0:10:42.819
But when you actually need,[br]like, documentation we sort

0:10:42.820,0:10:45.850
of like skimp out on sort of[br]design features.

0:10:45.850,0:10:48.600
And this is a point that,[br]like,

0:10:48.600,0:10:49.829
you see happen over and over[br]again.

0:10:49.830,0:10:51.520
A company that gets this[br]right is MailChimp.

0:10:51.520,0:10:53.470
And what they did was they[br]redesigned all of their

0:10:53.470,0:10:56.350
help guides so that they[br]looked like magazine covers.

0:10:56.350,0:10:57.890
And overnight, basically,

0:10:57.890,0:10:59.920
readership goes up on all[br]these features.

0:10:59.920,0:11:02.790
And customer support for[br]these things that sort of

0:11:02.800,0:11:04.220
help people optimize emails[br]goes down.

0:11:06.350,0:11:08.400
Speaking of documentation,[br]stripe.

0:11:09.810,0:11:13.550
What's interesting about an[br]API company is that there is

0:11:13.550,0:11:14.339
no UX.

0:11:14.340,0:11:16.810
The UX is actually just[br]documentation, right?

0:11:16.810,0:11:19.390
And there's opportunities[br]even in documentation.

0:11:19.390,0:11:20.720
Sort of the enchant and[br]amaze.

0:11:20.720,0:11:21.660
So one of the things that I

0:11:21.660,0:11:24.140
love about them is their[br]examples are wonderful.

0:11:24.140,0:11:25.280
But if you're actually,[br]like,

0:11:25.280,0:11:26.829
sort of logging into the[br]app,

0:11:26.830,0:11:29.340
one of the things that is a[br]super pain for most people,

0:11:29.340,0:11:30.770
when you're dealing with[br]most people's APIs, is,

0:11:30.770,0:11:33.920
like, grabbing your API[br]credentials and keys.

0:11:33.920,0:11:35.610
And what Stripe does is it[br]says oh,

0:11:35.610,0:11:36.610
if you're logged into the[br]app.

0:11:36.610,0:11:37.390
We automatically put

0:11:37.390,0:11:40.110
your API credentials into[br]the examples, so you only

0:11:40.110,0:11:43.900
have to copy paste once when[br]trying to learn their API.

0:11:47.600,0:11:50.500
When Wufoo wanted to launch[br]the third version of our API

0:11:50.500,0:11:51.939
we realized like, Okay, that[br]finally this is good enough

0:11:51.940,0:11:54.700
that we want people to sort[br]of build on top of it.

0:11:54.700,0:11:55.400
We were trying to figure[br]out,

0:11:55.400,0:11:56.740
like, how do we launch this[br]out to the world,

0:11:56.740,0:11:58.550
that sort of has our[br]personality behind it.

0:11:58.550,0:11:59.430
Because a lot of people,

0:11:59.430,0:12:02.000
they usually do things like[br]a programming API contest,

0:12:02.000,0:12:03.960
and they give out iPads and[br]iPhones.

0:12:03.960,0:12:05.830
And it makes you look like[br]everyone else.

0:12:05.830,0:12:09.440
And so, in our company one[br]weird value to

0:12:09.440,0:12:10.740
have it's a quirk of us,

0:12:10.740,0:12:13.570
is that the co-founders are[br]big medieval nuts.

0:12:13.570,0:12:16.510
And we take everyone out to[br]Medieval Times every single

0:12:16.510,0:12:19.130
year, on the anniversary of[br]the founding of the company.

0:12:19.130,0:12:21.760
And so, we said we have to[br]do something in that flavor.

0:12:21.760,0:12:23.650
And so we contacted the guys[br]at armor.com,

0:12:23.650,0:12:26.800
we said can you forward us a[br]custom battle-ax.

0:12:26.800,0:12:27.630
And what we said was if

0:12:27.630,0:12:31.580
you win our programming[br]contest you would win one.

0:12:33.390,0:12:35.800
The result is like people[br]wanted to talk about this.

0:12:35.800,0:12:37.380
It's something that people[br]wanted to work on cuz

0:12:37.380,0:12:39.860
they wanted to be able to[br]say it like I'm programming.

0:12:39.860,0:12:40.420
For a weapon.

0:12:43.269,0:12:46.850
And what's cool is we had[br]over 25 different

0:12:46.850,0:12:50.790
applications created for us[br]of quality and quantity that

0:12:50.790,0:12:53.410
we could not have paid for[br]on the budget and

0:12:53.410,0:12:54.819
the sort of time that we had[br]for this.

0:12:54.820,0:12:56.150
We got things like an iPhone[br]app and

0:12:56.150,0:12:58.720
Android and WordPress[br]plugins, right?

0:12:58.720,0:13:00.850
And all because what we did[br]was we changed how

0:13:00.850,0:13:02.520
people wanna talk about the[br]origin story of

0:13:02.520,0:13:04.779
how they're interacting with[br]one of our services.

0:13:06.100,0:13:08.990
And go like all day long of[br]going over these examples.

0:13:08.990,0:13:11.000
But I'm gonna short cut this[br]by saying,

0:13:11.000,0:13:12.750
you should just subscribe a[br]little bit of details.

0:13:12.750,0:13:16.480
It's just basically tons of[br]screen shots of software

0:13:16.480,0:13:17.370
that's just doing it right.

0:13:17.370,0:13:18.160
That shows that they're[br]being

0:13:18.160,0:13:21.980
conscientious of the user[br]and the customers.

0:13:28.467,0:13:30.799
When it comes to long term[br]relationships or

0:13:30.799,0:13:33.843
marriages, the only research[br]that we ended up having to

0:13:33.843,0:13:37.390
read was the stuff that was[br]done by John Gottman.

0:13:37.390,0:13:39.520
He's been featured in This[br]American Life,

0:13:39.520,0:13:41.490
in Malcolm Gladwell's books.

0:13:41.490,0:13:43.400
He's a marriage researcher[br]up in Seattle.

0:13:43.400,0:13:45.250
And he has an interesting[br]parlor trick that he can do.

0:13:45.250,0:13:48.730
He can watch a video tape of[br]a couple fighting about some

0:13:48.730,0:13:50.320
issue for 15 minutes.

0:13:50.320,0:13:52.900
And predict with an 85%[br]accuracy rate whether that

0:13:52.900,0:13:53.930
couple will be together or

0:13:53.930,0:13:55.630
not, or divorced in four[br]years.

0:13:56.780,0:13:58.770
If he increases that video[br]up to an hour.

0:13:58.770,0:14:00.620
And asks them to also talk[br]about their hopes and

0:14:00.620,0:14:01.540
dreams.

0:14:01.540,0:14:03.166
That prediction rating goes[br]up to 94%.

0:14:03.167,0:14:06.150
They showed these same video[br]tapes to marriage

0:14:06.150,0:14:07.699
counselors, successfully[br]married couples,

0:14:07.700,0:14:11.630
sociologists, psychiatrists,[br]priests, et cetera and

0:14:11.630,0:14:14.680
they can't predict with[br]random chance whether people

0:14:14.680,0:14:16.250
are gonna be together or[br]not.

0:14:16.250,0:14:18.590
So John Gauntman understands[br]something fundamental about.

0:14:19.910,0:14:22.110
How relationships work in[br]the long term.

0:14:22.110,0:14:25.650
And that basically how we[br]fight, even in a short term

0:14:25.650,0:14:29.300
period can indicate sort of[br]the whole system and

0:14:29.300,0:14:30.140
what it's gonna look like.

0:14:30.140,0:14:31.900
And one of the surprising[br]things he discovered,

0:14:31.900,0:14:32.420
it's not that

0:14:32.420,0:14:34.430
successfully married people[br]don't fight at all.

0:14:34.430,0:14:36.430
It turns out everybody[br]fights.

0:14:36.430,0:14:39.890
And we all fight about the[br]exact same things money,

0:14:39.900,0:14:41.300
kids, sex, time and others.

0:14:41.300,0:14:45.262
And others are things like[br]jealousy and in-laws.

0:14:45.263,0:14:47.543
To bring this around, right,[br]you can actually attribute

0:14:47.543,0:14:49.543
every single one of these to[br]problems that you see in

0:14:49.543,0:14:51.931
customer support when you're[br]building out your products.

0:14:53.368,0:14:56.532
Right, so.

0:14:56.532,0:14:58.447
This costs too much.

0:14:58.447,0:15:00.401
I'm having problems with a[br]credit card.

0:15:00.401,0:15:02.260
If you're building a service[br]that helps people deal with

0:15:02.260,0:15:03.189
their clients they're very

0:15:03.190,0:15:05.480
sensitive about anything[br]happening with that.

0:15:05.480,0:15:06.530
Performance.

0:15:06.530,0:15:08.750
How long you're up and how[br]fast.

0:15:10.780,0:15:14.110
Others are I said jealousy[br]and in-laws.

0:15:14.110,0:15:15.960
Right, so that's competition[br]and partnerships.

0:15:15.960,0:15:17.310
So anything weird happening[br]there,

0:15:17.310,0:15:19.890
people are gonna write to[br]you about.

0:15:19.900,0:15:20.520
And the reason I like to[br]think about this in

0:15:20.520,0:15:23.430
terms of customer support is[br]that in every ones that

0:15:23.430,0:15:25.920
are processing of like a[br]conversion funnel,

0:15:25.920,0:15:27.790
customer support is the[br]thing that happens in

0:15:27.790,0:15:29.990
between every one of these[br]steps.

0:15:29.100,0:15:30.720
It's the reason why people[br]don't make it

0:15:30.720,0:15:31.650
further down there.

0:15:31.650,0:15:33.629
It's the thing that prevents[br]conversion from happening.

0:15:35.460,0:15:38.130
Now, as we were thinking[br]through all these ideas and

0:15:38.130,0:15:39.680
as we were building up the[br]company,

0:15:39.680,0:15:42.469
we realized that there's a[br]big problem about how

0:15:42.470,0:15:43.760
everyone sort of starts[br]their company or

0:15:43.760,0:15:46.230
build up their sort of[br]engineering teams.

0:15:46.230,0:15:48.760
And that is that there's a[br]broken feedback loop there.

0:15:48.760,0:15:50.560
People are divorced from the[br]consequences of

0:15:50.560,0:15:51.810
their actions.

0:15:51.810,0:15:53.579
And this is the result of[br]actually the natural

0:15:53.580,0:15:55.600
evolution of how most[br]companies get founded,

0:15:55.600,0:15:57.880
especially by technical[br]co-founders, right.

0:15:57.880,0:16:01.630
Before launch it is a time[br]of bliss, nirvana, and

0:16:01.630,0:16:03.500
opportunity, right?

0:16:03.500,0:16:05.469
Nothing that you do is[br]wrong, right?

0:16:06.660,0:16:09.850
By your hand, which you feel[br]is like god, everything that

0:16:09.850,0:16:12.750
you write, every line of[br]code feels perfect, right,

0:16:12.750,0:16:14.770
and is ingenious to you.

0:16:14.770,0:16:17.829
The thing that happens is[br]after launch reality sort of

0:16:17.830,0:16:20.300
sets in and then all these[br]other tasks sort of

0:16:20.300,0:16:21.579
come into place that we have[br]to deal with.

0:16:23.330,0:16:25.720
Now what technical[br]co-founders wanna do is get

0:16:25.720,0:16:29.240
back to that initial state[br]and so what we often do,

0:16:29.240,0:16:31.380
and what we often see is[br]that companies start

0:16:31.380,0:16:34.240
siloing off all these other[br]things that actually is

0:16:34.240,0:16:37.410
what makes a start up or a[br]company sort of real, right?

0:16:37.410,0:16:39.540
And have other people do[br]them.

0:16:39.540,0:16:42.589
To, in our minds, these[br]other tasks are inferior.

0:16:42.590,0:16:43.660
Right?[br]And we have other people in

0:16:43.660,0:16:44.310
the company do them.

0:16:45.822,0:16:47.580
And so for us, what we're[br]trying to figure out

0:16:47.580,0:16:49.590
is how do we change software[br]development so that we

0:16:49.590,0:16:53.970
inject some values that we[br]don't talk about enough?

0:16:53.970,0:16:55.970
Responsibility,[br]accountability,

0:16:55.970,0:16:57.780
humility, modesty.

0:16:57.780,0:17:01.300
Right, and we called this,[br]like a lot of other people,

0:17:01.300,0:17:05.720
we had an acronym, Support[br]Driven Development, and

0:17:05.720,0:17:09.900
it's very similar to TDD or[br]other agile practice.

0:17:09.900,0:17:12.589
It's a way of creating high[br]quality software, but

0:17:12.589,0:17:13.819
it's super simple.

0:17:13.819,0:17:15.182
You don't need like, a[br]SCRUM,

0:17:15.182,0:17:17.608
you don't need a bunch of[br]post-it notes, all you

0:17:17.608,0:17:19.909
have to do is make everyone[br]do customer support.

0:17:21.880,0:17:22.800
And what you end up having[br]is

0:17:22.800,0:17:23.990
you fix the feedback loop,[br]right?

0:17:23.990,0:17:25.450
The people who build the[br]software are the ones

0:17:25.450,0:17:26.240
supporting it and

0:17:26.240,0:17:28.940
you get all these sort of[br]nice benefits as a result.

0:17:30.320,0:17:33.340
So one of them is, support[br]responsible developers and

0:17:33.340,0:17:35.120
designers and people who[br]build the stuff,

0:17:35.120,0:17:36.600
they give the very best[br]support.

0:17:36.600,0:17:38.600
Now we're not the first[br]person to think of this.

0:17:38.600,0:17:41.480
Paul English was a big[br]proporter of this at Kayak,

0:17:41.480,0:17:44.750
and what he did was install[br]a red customer support phone

0:17:44.750,0:17:46.640
line in the middle of the[br]engineering floor.

0:17:46.640,0:17:49.950
And they were just regular[br]customer support calls.

0:17:49.950,0:17:53.890
And people would ask him[br]often times,

0:17:53.890,0:17:55.560
why would you pay engineers[br]$120,000 or

0:17:55.560,0:17:59.389
more to do something that[br]you can pay other people

0:17:59.390,0:18:02.890
a fraction of to handle in[br]like a call center?

0:18:02.890,0:18:04.560
And he says, well, after the[br]second or

0:18:04.560,0:18:06.929
third time that phone rings[br]and the engineer gets

0:18:06.930,0:18:09.990
the same problem they stop[br]what they're doing, they fix

0:18:09.990,0:18:13.940
the bug, and we stop getting[br]phone calls about it.

0:18:13.940,0:18:16.670
It, it's a way of having Q-A[br]in a sort of

0:18:16.670,0:18:17.840
nice elegant solution.

0:18:19.710,0:18:23.820
Now, John Gotman talks about[br]the reason that we often

0:18:23.820,0:18:26.600
break up with one another as[br]due to four major causes and

0:18:26.600,0:18:27.389
their warning signs.

0:18:27.390,0:18:29.300
He calls them the four[br]horsemen, right?

0:18:29.300,0:18:32.830
Criticism, contempt,[br]defensiveness,

0:18:32.830,0:18:33.679
and stonewalling.

0:18:33.680,0:18:38.590
Now, criticism is basically[br]people starting to focus,

0:18:38.590,0:18:40.350
not just on the specific[br]issue at hand, but

0:18:40.350,0:18:41.447
on the overarching issues.

0:18:41.448,0:18:45.780
Like, you never, right,[br]listen to your users or

0:18:45.780,0:18:48.480
you never think about us all[br]the time.

0:18:48.480,0:18:48.980
Right?

0:18:50.390,0:18:53.880
Contempt is when someone is[br]purposely trying to

0:18:53.880,0:18:55.170
insult somebody.

0:18:55.170,0:18:59.620
Defensiveness is not trying[br]to take accountability

0:18:59.620,0:19:01.810
trying to make excuses for[br]the actions.

0:19:01.810,0:19:03.532
And stonewalling is[br]basically shutting down.

0:19:03.532,0:19:07.120
Stonewalling, to John[br]Gotman, is, is one of

0:19:07.120,0:19:09.250
the worst things that we can[br]do in a relationship.

0:19:09.260,0:19:10.593
Hold up.

0:19:14.588,0:19:17.711
And oftentimes, you know, we[br]don't worry much about this

0:19:17.711,0:19:19.970
in customer se, criticism[br]and contempt.

0:19:19.970,0:19:20.770
Right?[br]Defensiveness,

0:19:20.770,0:19:22.530
you see this all the time,[br]all the times in companies,

0:19:22.530,0:19:23.790
especially as they get[br]older.

0:19:23.790,0:19:25.320
But stonewalling, this is[br]something I

0:19:25.320,0:19:27.230
see happen with start-ups[br]all the time.

0:19:28.720,0:19:30.100
You get a bunch of customer[br]support sort of

0:19:30.100,0:19:32.379
coming in and you just think[br]I don't need to answer it.

0:19:32.380,0:19:33.130
I don't need to respond.

0:19:33.130,0:19:34.420
Right?[br]And

0:19:34.420,0:19:36.410
that act of just not even[br]getting back to

0:19:36.410,0:19:38.290
them is one of the worst[br]things you can do.

0:19:38.290,0:19:40.159
And it's probably some of[br]the biggest causes of

0:19:40.160,0:19:42.340
churn in the early stages of[br]start-ups.

0:19:43.580,0:19:45.429
This is how support worked[br]out with Wufoo.

0:19:45.430,0:19:48.250
When we were acquired we had[br]about 500,000 users on

0:19:48.250,0:19:51.230
the system, 5 million people[br]used Wufoo forms and

0:19:51.230,0:19:52.940
reports, whether they knew[br]it or not, and

0:19:52.940,0:19:56.100
all those people got support[br]from the same ten people,

0:19:56.100,0:19:56.639
and usually it was

0:19:56.640,0:20:00.200
only one person dedicated[br]support a day, or any shift.

0:20:00.200,0:20:02.180
Results in about 400 issues[br]a week.

0:20:02.180,0:20:04.110
That's about 800 emails.

0:20:04.110,0:20:05.928
But a response time from 9[br]a.m.

0:20:05.928,0:20:06.440
to 9 p.m.[br]was

0:20:06.440,0:20:07.865
between seven to 12 minutes.

0:20:07.865,0:20:08.980
Right?[br]And from 9 p.m.

0:20:08.980,0:20:11.130
to midnight it was an hour.

0:20:11.130,0:20:13.300
And on the weekend it would[br]be no longer than 24 hours.

0:20:13.300,0:20:16.200
And we carried this up all[br]the way up to the scale.

0:20:18.300,0:20:19.780
What a lot of people forget[br]about, and

0:20:19.780,0:20:23.200
often talk about with[br]Airbnb, is how, like oh,

0:20:23.200,0:20:24.870
they did this interesting[br]thing where they had,

0:20:24.870,0:20:26.699
went up to New York and[br]offered, like,

0:20:26.700,0:20:28.280
professional photographers,[br]and the founders would go

0:20:28.280,0:20:30.400
out there and actually take[br]pictures of

0:20:30.400,0:20:31.860
the people's apartments to[br]help them sell more.

0:20:31.860,0:20:34.389
Focusing on the stories[br]around conversion.

0:20:34.390,0:20:36.440
What most people don't[br]realize is a lot of

0:20:36.440,0:20:38.783
times when I saw Joe in the[br]early days of Airbnb,

0:20:38.783,0:20:41.301
he had a phone, sort of head[br]set stuck to his head all

0:20:41.301,0:20:44.720
the time, because he was[br]doing phone support nonstop.

0:20:46.286,0:20:48.307
Churn is a story we don't[br]like to talk about often,

0:20:48.307,0:20:49.280
all the time.

0:20:49.280,0:20:52.540
Airbnb's sort of growth[br]really started picking up

0:20:52.540,0:20:53.440
once they figured out

0:20:53.440,0:20:56.300
how to match capacity to the[br]demand, or the phone

0:20:56.300,0:20:58.730
calls that they were getting[br]into their support system.

0:21:00.580,0:21:02.960
At Wufoo we actually[br]constantly did experiments

0:21:02.960,0:21:05.550
around support, because[br]we're so obsessed with it.

0:21:05.550,0:21:08.540
One experiment we did was,[br]we heard Kathy Sierra

0:21:08.540,0:21:10.860
do a talk about there's a[br]disconnect between the

0:21:10.860,0:21:14.580
motions that we have when we[br]need help, and sort of.

0:21:14.580,0:21:16.620
The content and the[br]reactions we get from

0:21:16.620,0:21:18.429
people when we get help from[br]them, especially online.

0:21:18.430,0:21:22.780
Because they just don't see[br]all those nonverbal cues.

0:21:22.780,0:21:25.440
So she said unless there's[br]face recognition on the web,

0:21:25.440,0:21:26.420
we're just always going to[br]be

0:21:26.420,0:21:27.980
disconnected from our users.

0:21:27.980,0:21:28.760
Our feeling was,

0:21:28.760,0:21:31.110
like, well we're not face[br]recognition experts but

0:21:31.110,0:21:32.699
I think there's another way[br]of getting empathy.

0:21:32.700,0:21:35.710
So, as form builders, we[br]added a drop-down and

0:21:35.710,0:21:37.590
what we said was like, hey,[br]what's your emotional state?

0:21:39.640,0:21:41.180
And our hypothesis was

0:21:41.180,0:21:42.300
that no one was gonna fill[br]this out.

0:21:42.300,0:21:45.790
We basically thought, oh[br]okay, you know what the,

0:21:45.790,0:21:48.909
this is gonna be pretty a[br]lame experiment, but

0:21:48.910,0:21:49.930
we'll see how it sort of[br]goes.

0:21:49.930,0:21:51.125
And it turned out the[br]Emotional State

0:21:51.125,0:21:53.770
drop-down field was filled[br]out 75.8% of the time.

0:21:55.300,0:21:57.374
The browser type drop-down[br]filled just in

0:21:57.374,0:22:00.184
comparison was filled out[br]78.1% of the time.

0:22:00.184,0:22:00.816
All right?

0:22:00.816,0:22:03.337
So people were basically[br]telling us, for

0:22:03.337,0:22:06.506
my technical support issue[br]how I feel about this

0:22:06.506,0:22:10.360
problem is just as important[br]as, like, all the technical

0:22:10.360,0:22:13.220
details you need to sorta[br]figure out how to debug it.

0:22:14.480,0:22:16.580
Now we didn't prioritize[br]things or

0:22:16.580,0:22:18.379
triage things by emotion,[br]right, and for

0:22:18.380,0:22:19.860
the most part, people didn't[br]game the system.

0:22:19.860,0:22:22.820
One of the interesting[br]byproducts of it was that we

0:22:22.820,0:22:24.450
noticed that people started[br]being nicer to

0:22:24.450,0:22:25.850
us in the customer support.

0:22:25.850,0:22:27.270
It was something sort of[br]subconscious.

0:22:27.270,0:22:28.590
We just were thinking like,[br]wow,

0:22:28.590,0:22:29.870
users are so much better[br]now.

0:22:29.870,0:22:31.429
What's going on?

0:22:31.430,0:22:33.710
And we went back and looked[br]at the data and

0:22:33.710,0:22:36.910
we did some text analysis[br]and we realized is that, oh,

0:22:36.910,0:22:39.790
when it comes to only[br]communicating with

0:22:39.790,0:22:43.620
people over written words,[br]like email, there's

0:22:43.620,0:22:45.750
only three ways that you[br]show strong emotions, right?

0:22:45.750,0:22:49.560
Exclamation marks curse[br]words, and all caps.

0:22:49.560,0:22:52.139
And sure enough on all three[br]of those metrics,

0:22:52.140,0:22:52.840
they've gone down.

0:22:52.840,0:22:55.700
In sort of the way people[br]were talking to us in

0:22:55.700,0:22:56.340
the customer support.

0:22:56.340,0:22:58.780
Once people had a simple[br]outlet for their emotions,

0:22:58.780,0:23:00.340
right, people would be a lot[br]more rational,

0:23:00.340,0:23:02.996
and it made our jobs a lot[br]more pleasant as a result.

0:23:02.997,0:23:08.860
The other byproduct that is[br]awesome is that

0:23:08.860,0:23:11.280
you actually build better[br]software when you do this.

0:23:11.280,0:23:12.310
Far better software.

0:23:12.310,0:23:14.900
This is actually backed up[br]by tons of research.

0:23:14.900,0:23:16.189
Jared Spool, a user[br]interface engineer, which is

0:23:16.190,0:23:18.745
sort of the big players in[br]this space says like,

0:23:18.745,0:23:21.399
there's a direct correlation[br]to how much time we

0:23:21.400,0:23:24.810
spend directly exposed to[br]users and

0:23:24.810,0:23:27.800
how good our designs sort of[br]get.

0:23:27.800,0:23:28.800
He says it has to come in[br]this specific way.

0:23:28.800,0:23:30.680
It has to be direct[br]exposure, right?

0:23:30.680,0:23:31.710
It can't be something where

0:23:31.710,0:23:34.370
someone generates a report[br]or through a graph, you

0:23:34.370,0:23:37.700
have to be interacting with[br]them somewhat real time.

0:23:37.700,0:23:39.290
It has to be a minimum of[br]every six weeks.

0:23:39.290,0:23:41.180
And it has to be for at[br]least two hours.

0:23:41.180,0:23:43.720
Otherwise your software will[br]get worse over time.

0:23:46.340,0:23:48.860
Our developers, our people[br]who are on,

0:23:48.860,0:23:51.200
on Wufoo were getting[br]exposed to our users four to

0:23:51.200,0:23:52.820
eight hours every single[br]week.

0:23:54.280,0:23:56.570
And what it does is it[br]changes the way you sort of

0:23:56.570,0:23:57.437
build software.

0:23:57.437,0:24:00.149
Jared Spool has another way[br]of talking about how

0:24:00.150,0:24:01.380
we build products.

0:24:01.380,0:24:02.650
Right?

0:24:02.650,0:24:06.780
Let's imagine that this[br]represents all the knowledge

0:24:06.780,0:24:09.899
needed to sort of use your[br]app on a spectrum.

0:24:09.900,0:24:11.770
Right?[br]This is like no knowledge.

0:24:11.770,0:24:12.340
Right?[br]And this is

0:24:12.340,0:24:13.780
all the knowledge needed.

0:24:13.780,0:24:14.300
Right?[br]And

0:24:14.300,0:24:16.280
these two lines are pretty[br]much your interactions with

0:24:16.280,0:24:17.610
users what you're trying to[br]get them to.

0:24:17.610,0:24:20.270
This is currently where[br]their knowledge point is and

0:24:20.270,0:24:22.460
this is the target knowledge[br]point that you're trying to

0:24:22.460,0:24:25.250
get them to, to understand[br]that to use your app.

0:24:25.250,0:24:27.102
The gap between those is[br]called the knowledge gap,

0:24:27.102,0:24:28.497
Jared Spool called it, Spool[br]calls.

0:24:28.497,0:24:31.846
And what's interesting about[br]this is there's only two

0:24:31.846,0:24:34.280
ways, right, to sort of fix[br]this.

0:24:34.280,0:24:36.129
That gap represents how[br]intuitive your

0:24:36.130,0:24:38.100
app is, right?

0:24:38.100,0:24:41.210
You either get the user to[br]increase their

0:24:41.210,0:24:44.800
knowledge or you decrease[br]the amount of knowledge

0:24:44.800,0:24:45.871
that's needed to use your[br]application.

0:24:45.872,0:24:48.450
And oftentimes, as engineers[br]and people who build and

0:24:48.450,0:24:50.699
work on products, we think[br]let's add new features.

0:24:51.800,0:24:52.790
And new features only

0:24:52.790,0:24:55.149
means let's increase the[br]knowledge gap.

0:24:57.870,0:25:00.700
So for us we actually[br]focused a lot on

0:25:00.700,0:25:01.792
the other sort of direction.

0:25:01.792,0:25:03.530
And so what that meant we[br]spent a lot of time,

0:25:03.530,0:25:06.870
30% of our engineering time[br]was spent on internal tools

0:25:06.870,0:25:09.101
to help with our customer[br]support stuff.

0:25:09.102,0:25:10.200
But oftentimes it

0:25:10.200,0:25:12.750
was spent helping people[br]help themselves.

0:25:12.750,0:25:15.870
Things like frequently asked[br]questions, tool tips.

0:25:15.870,0:25:17.649
Things like, if you just[br]click the help link,

0:25:17.650,0:25:20.510
right, instead of taking you[br]to the generic help sort of

0:25:20.510,0:25:21.480
documentation page,

0:25:21.480,0:25:24.150
you go to the specific page[br]where you're looking at

0:25:24.150,0:25:25.540
is going to be most, sort[br]of,

0:25:25.540,0:25:27.960
appropriate for what you're[br]working on.

0:25:27.960,0:25:29.700
We redesigned our[br]documentation over and

0:25:29.700,0:25:31.870
over again, AB tested it[br]constantly.

0:25:31.870,0:25:34.400
One iteration of our[br]documentation reduced

0:25:34.400,0:25:36.290
customer support by 30% over[br]night.

0:25:37.330,0:25:38.900
It's one of those things[br]where,

0:25:38.900,0:25:41.399
like, overnight, all the[br]people that work on

0:25:41.400,0:25:43.960
the product immediately had[br]30% less work to do.

0:25:46.800,0:25:48.490
Now what happens if you have[br]everyone work on

0:25:48.490,0:25:49.630
customer support constantly,[br]and

0:25:49.630,0:25:52.220
thinking about it in terms[br]of a remarkable way?

0:25:52.220,0:25:53.430
Well, I talked a lot about,

0:25:53.430,0:25:56.290
in the very beginning growth[br]is a function of

0:25:56.290,0:25:58.000
conversion and churn.

0:25:58.000,0:26:01.140
This is Wufoo's growth curve[br]for the first five years.

0:26:01.140,0:26:02.270
Right?

0:26:02.270,0:26:03.389
What's interesting is that[br]we

0:26:03.390,0:26:08.200
paid no money on[br]advertising, on marketing.

0:26:08.200,0:26:10.720
All of it was done by word[br]of mouth growth.

0:26:10.720,0:26:11.800
Right?

0:26:11.800,0:26:13.480
And the interaction between,[br]like,

0:26:13.480,0:26:16.460
new users and downgrades are[br]this.

0:26:18.570,0:26:20.939
It's so slight what it[br]takes, that gap,

0:26:20.940,0:26:21.800
making that sort of work.

0:26:21.800,0:26:24.710
And what a lot of people[br]keep forgetting is that

0:26:24.710,0:26:26.970
there's almost no difference[br]between an increase in

0:26:26.970,0:26:30.370
conversion rate, 1% increase[br]and a 1% decrease churn.

0:26:30.370,0:26:32.840
They do exactly the same[br]thing to your growth.

0:26:32.840,0:26:35.709
However, the ladder is[br]actually much easier to do.

0:26:35.710,0:26:38.380
It's much cheaper to do, in[br]your apps, and

0:26:38.380,0:26:39.940
a lot of the times we[br]neglect this,

0:26:39.940,0:26:42.100
neglect this to way far[br]along.

0:26:42.100,0:26:42.960
Right?[br]And we usually have our

0:26:42.960,0:26:46.900
B team works on these sort[br]of projects and services.

0:26:46.900,0:26:47.770
This is actually not the[br]graph that we track most of

0:26:47.770,0:26:49.250
the time at Wufoo, it's not[br]even the one I'm proud of.

0:26:49.250,0:26:51.855
This is the one I'm proud[br]of.

0:26:51.855,0:26:54.980
Cuz even though we have this[br]sort of nice, awesome curve

0:26:54.980,0:26:58.400
of growth, this is what, how[br]loud is this scale?

0:26:58.400,0:27:00.340
Keep the company small, have[br]an awesome culture.

0:27:01.680,0:27:02.920
And that required doing a[br]lot of

0:27:02.920,0:27:08.796
these things to help people[br]sort of do what they need.

0:27:08.796,0:27:13.709
So John Gotman noticed there[br]was a different type of

0:27:13.709,0:27:18.978
behavior for relationships[br]and why people divorced.

0:27:18.978,0:27:21.938
Basically there would be[br]some subset of people who

0:27:21.938,0:27:24.404
would stay together 10, 15[br]years and

0:27:24.404,0:27:26.870
then all of a sudden they[br]divorce and

0:27:26.870,0:27:29.899
there was none of the other[br]indicators which sort of

0:27:29.899,0:27:32.320
show that this is what was[br]gonna happen.

0:27:32.320,0:27:34.670
And I was looking through[br]the data and

0:27:34.670,0:27:35.868
I realized oh, there's no[br]passion,

0:27:35.868,0:27:38.194
there's no fire between[br]these people, right?

0:27:38.194,0:27:39.678
When it comes to[br]relationships they

0:27:39.679,0:27:42.590
kinda follow the second law[br]of thermodynamics, right?

0:27:42.590,0:27:44.505
In a closed energy system[br]things tend to run down so

0:27:44.506,0:27:46.662
you have to constantly be[br]putting energy and

0:27:46.662,0:27:47.720
effort back into it.

0:27:47.720,0:27:50.430
Now the way a lot of people[br]sort of think about showing

0:27:50.430,0:27:54.400
people that I care about you[br]in products and in companies

0:27:54.400,0:27:56.870
is to do things like let's[br]have a blog, right?

0:27:56.870,0:27:58.239
Lets' have a newsletter?

0:27:59.460,0:28:01.870
The thing is, we look at[br]these rates and basically it

0:28:01.870,0:28:03.959
was such a small percentage[br]of our active users that it

0:28:03.960,0:28:06.290
was like, most of our users[br]have no idea all

0:28:06.290,0:28:08.659
the awesome stuff that we're[br]doing for them.

0:28:08.660,0:28:09.980
So we built a new tool.

0:28:09.980,0:28:11.540
We called it the Wufoo[br]system, and

0:28:11.540,0:28:13.570
what it allowed us to do was[br]just time stamp every new

0:28:13.570,0:28:15.990
feature that we're building[br]for users and

0:28:15.990,0:28:17.540
then every time they would[br]log in, we would look at

0:28:17.540,0:28:20.830
the difference between their[br]log in time or last log in

0:28:20.830,0:28:22.580
time and the new features[br]that were implemented.

0:28:22.580,0:28:24.120
And when you had this[br]message show up,

0:28:24.120,0:28:25.300
hey since you've been gone

0:28:25.300,0:28:27.800
here's all the awesome stuff[br]that Wufoo did for you.

0:28:27.800,0:28:29.840
Hands down this was the most[br]talked about feature I've

0:28:29.840,0:28:32.584
ever had every time I went[br]out to talk to users.

0:28:32.585,0:28:34.860
Right they'd say like dude I[br]love that since you've be,

0:28:34.860,0:28:37.490
since you've gone thing even[br]though I pay the same amount

0:28:37.490,0:28:39.510
every single month you guys[br]are doing something for

0:28:39.510,0:28:42.110
me almost every week and[br]it's totally awesome and

0:28:42.110,0:28:43.780
makes me feel, I'm getting[br]maximum value.

0:28:47.200,0:28:48.200
The other thing that we did[br]in

0:28:48.200,0:28:50.300
addition to having everyone[br]support the people that

0:28:50.300,0:28:52.500
paid their pay check is have[br]them say thank you.

0:28:52.500,0:28:56.790
And this was a large part[br]due to us injecting sort of

0:28:56.790,0:28:59.280
humility and modesty in to[br]sort of the equation.

0:29:01.200,0:29:03.270
Every single Friday we would[br]get together and

0:29:03.270,0:29:05.350
we'd just write simple hand[br]written thank you cards to

0:29:05.350,0:29:06.290
our users.

0:29:07.340,0:29:10.606
And I know there's tons of[br]people who would not be sort

0:29:10.606,0:29:13.799
of excited about doing this,[br]but it was a ritual that

0:29:13.799,0:29:16.485
made sort of all the[br]difference in terms of,

0:29:16.485,0:29:19.822
like, having a team that was[br]very tightly neat,

0:29:19.822,0:29:21.226
tightly knit, also.

0:29:21.226,0:29:23.190
And working on stuff that[br]they really cared about.

0:29:23.190,0:29:26.130
They always constantly knew[br]what the mission was for,

0:29:26.130,0:29:27.960
and why we sort of did what[br]we did.

0:29:27.960,0:29:30.100
These aren't fancy thank you[br]cards, right?

0:29:30.100,0:29:30.810
They're just simple,

0:29:30.810,0:29:32.860
like handwritten stuff on an[br]index card.

0:29:32.860,0:29:33.990
We threw in a sticker, and

0:29:33.990,0:29:37.140
slapped on a dinosaur on the[br]front of it.

0:29:37.140,0:29:41.720
And, what's interesting is[br]we started this practice as

0:29:41.720,0:29:45.830
a result of the early days[br]of starting Wufoo.

0:29:45.830,0:29:48.390
Chris, Ron, and I were[br]talking, and we're trying to

0:29:48.390,0:29:50.103
figure out, what are we[br]gonna do to sort of show

0:29:50.103,0:29:52.139
users that we appreciate[br]them around Christmas.

0:29:52.139,0:29:55.341
And he, Chris came up with[br]this idea where he said

0:29:55.341,0:29:58.824
hey guys, so a couple years[br]ago my mom, like, made me

0:29:58.825,0:30:01.831
write thank you notes to all[br]of my relatives.

0:30:01.831,0:30:03.554
For my Christmas gifts.

0:30:03.555,0:30:05.440
And I didn't really like to[br]do it but

0:30:05.440,0:30:07.750
the following year all my[br]presents were super good.

0:30:07.750,0:30:12.170
So, I think we should try[br]this for our business and

0:30:12.170,0:30:13.770
see how it goes.

0:30:13.770,0:30:17.320
So, that first year we wrote[br]handwritten Christmas cards

0:30:17.320,0:30:19.750
to all of our users that[br]first year.

0:30:19.750,0:30:21.360
Second year rolls around,[br]and

0:30:21.360,0:30:22.679
we have too many customers,[br]like, and

0:30:22.680,0:30:24.670
it's still just the three[br]founders.

0:30:24.670,0:30:26.920
And we were going like we're[br]kinda screwed,

0:30:26.920,0:30:28.650
I don't know what we're[br]gonna do.

0:30:28.650,0:30:30.940
And we read a book called[br]the ultimate question, and

0:30:30.940,0:30:32.540
in it he talks about hey,

0:30:32.540,0:30:35.490
just focus on your most[br]profitable users.

0:30:35.490,0:30:36.510
And just tend them and

0:30:36.510,0:30:38.330
take care of them, then[br]it'll work out.

0:30:38.330,0:30:38.909
So we're like, all right,

0:30:38.910,0:30:41.000
that, that makes sense,[br]that's scalable.

0:30:41.000,0:30:41.770
So that year we

0:30:41.770,0:30:43.459
only write to our highest[br]paying customers.

0:30:44.480,0:30:47.630
And the January rolls around[br]that second year and

0:30:47.630,0:30:50.540
one of our long time loyal[br]users writes us and

0:30:50.540,0:30:55.620
he's basically like, hey[br]guys I, I really loved

0:30:55.620,0:30:58.290
that Christmas card you sent[br]me the first year and I just

0:30:58.290,0:31:01.000
wanted you to know I haven't[br]received my second card yet

0:31:01.000,0:31:02.520
and I'm just looking forward[br]to it.

0:31:02.520,0:31:04.400
I know you didn't forget[br]about me.

0:31:06.500,0:31:07.399
Thanks a lot.

0:31:07.400,0:31:11.930
So we're like, fuck, because[br]the best way to sort of

0:31:11.930,0:31:13.890
exceed expectations is not[br]to send any to begin with,

0:31:13.890,0:31:15.500
so we were like sort of in[br]this conundrum.

0:31:15.500,0:31:17.940
And what we decided, after[br]thinking about it for

0:31:17.940,0:31:20.270
a while, is that we need to[br]stop doing it,

0:31:20.270,0:31:22.400
you know, just one time a[br]year.

0:31:22.400,0:31:23.760
It needs to be something[br]that's part of the culture.

0:31:23.760,0:31:25.800
Happens every, every, every[br]sort of week, even.

0:31:25.800,0:31:27.950
And even though we'll never[br]catch up to all of

0:31:27.950,0:31:29.500
our customers.

0:31:29.500,0:31:29.710
Just the practice of

0:31:29.710,0:31:31.490
doing it will make all the[br]difference.

0:31:33.300,0:31:34.310
I talked a lot

0:31:34.310,0:31:37.320
about a bunch of like[br]lovey-dovey stuff and

0:31:37.320,0:31:39.470
sort of like touchy-feely[br]things that I think a lot of

0:31:39.470,0:31:41.450
engineers don't like to[br]think about too often.

0:31:41.450,0:31:45.570
And so I'll end on some sort[br]of hard business

0:31:45.570,0:31:47.320
data or research.

0:31:47.320,0:31:49.159
There's an article that was[br]put out by

0:31:49.160,0:31:51.220
the Harvard Business Review[br]several years ago by

0:31:51.220,0:31:53.800
Michael Treacy and Fred[br]Wiersema and in it they

0:31:53.800,0:31:55.560
talk about the discipline of[br]market leaders.

0:31:55.560,0:31:57.360
They say there's only three[br]ways that you

0:31:57.360,0:32:01.629
achieve market dominance and[br]depending on how you want to

0:32:01.630,0:32:03.300
achieve that market[br]dominance you have

0:32:03.300,0:32:05.870
to organize your company in[br]a very specific way.

0:32:05.870,0:32:09.350
Best price, best product and[br]best overall solution.

0:32:09.350,0:32:09.860
If you want to be

0:32:09.860,0:32:12.270
the best price out there you[br]focus on logistics.

0:32:12.270,0:32:13.680
A Wal-Mart, an Amazon.

0:32:13.680,0:32:15.390
If you want to be the best[br]product out there you

0:32:15.390,0:32:16.540
focus on R&D.

0:32:16.540,0:32:17.460
Apple's usually

0:32:17.460,0:32:19.490
a quintessential example of[br]that.

0:32:19.490,0:32:21.400
Best overall solution is[br]about

0:32:21.400,0:32:21.930
being customer intimate.

0:32:21.930,0:32:23.410
And this is the path that[br]you see followed by

0:32:23.410,0:32:25.950
luxury brands and[br]hospitality industry.

0:32:27.190,0:32:30.400
What I love about this path[br]towards market dominance is

0:32:30.400,0:32:34.750
that the third one is the[br]only one that

0:32:34.750,0:32:37.620
everyone can do at any stage[br]of their company.

0:32:37.620,0:32:40.459
Requires almost no money to[br]get started with.

0:32:40.460,0:32:41.540
Usually just co.

0:32:41.540,0:32:44.780
Requires a little bit of[br]humility and some manners.

0:32:44.780,0:32:48.590
And as a result you can[br]achieve the success as any

0:32:48.590,0:32:50.189
other people in sort of your[br]market.

0:32:52.110,0:32:53.679
That's all I got.

0:32:53.680,0:32:54.340
Thank you very much.

0:33:00.370,0:33:03.229
Yeah, let's take some[br]questions if you guys

0:33:03.230,0:33:03.912
have any.

0:33:03.912,0:33:04.854
Right in the back there.

0:33:04.854,0:33:07.339
>> Building products that[br]users love?

0:33:07.339,0:33:10.341
You might have multiple[br]different types of users.

0:33:10.342,0:33:14.150
How do you build one product[br]that all users love?

0:33:14.150,0:33:16.150
Maybe there is a feature[br]that one really likes but

0:33:16.150,0:33:18.110
detracts value from one[br]that.

0:33:18.110,0:33:19.149
>> All right.[br]So what do you do

0:33:19.150,0:33:20.950
when you have a product with[br]lots of

0:33:20.950,0:33:22.250
different type of users,[br]right?

0:33:22.250,0:33:23.490
Some users will love one[br]thing and

0:33:23.490,0:33:25.250
another will, will another.

0:33:25.250,0:33:25.970
And I agree, there's

0:33:25.970,0:33:27.290
a interesting fine line for[br]that.

0:33:27.290,0:33:29.370
What I always, usually tell[br]people,

0:33:29.370,0:33:31.700
is focus on the people who[br]are the most passionate,

0:33:31.700,0:33:32.360
especially in the early[br]stages.

0:33:32.360,0:33:34.200
Right?[br]Whoever's, whatever niche

0:33:34.200,0:33:37.320
it's gonna be, that's who I[br]focus on completely.

0:33:37.320,0:33:39.280
Things that a lot of[br]different projects did.

0:33:39.280,0:33:40.139
I think Ben Silverman of

0:33:40.140,0:33:43.290
Pinterest started off with a[br]designer bloggers, right?

0:33:43.290,0:33:44.900
Curtail your thing for

0:33:44.900,0:33:46.610
them and eventually you'll[br]figure out sort of universal

0:33:46.610,0:33:49.149
values that will appeal to a[br]lot of other people.

0:33:49.150,0:33:50.920
So, just start one at a[br]time.

0:33:50.920,0:33:51.720
And.

0:33:51.720,0:33:53.900
The, a lot of the examples[br]that you see up there,

0:33:53.900,0:33:55.280
a lot of people make the[br]mistake is like, oh,

0:33:55.280,0:33:56.960
I'll just make my app funny.

0:33:56.960,0:34:00.130
But, humor is like really[br]difficult to do.

0:34:00.130,0:34:01.200
Right?[br]What you wanna shoot for

0:34:01.200,0:34:02.350
is something sort of witty.

0:34:02.350,0:34:03.719
And, quite honestly,

0:34:03.720,0:34:05.730
you have to get[br]functionality right.

0:34:05.730,0:34:07.220
So like the Japanese[br]quality.

0:34:07.220,0:34:10.210
If you don't have a on[br]there, right, don't try to

0:34:10.210,0:34:13.100
do anything witty, right,[br]cuz it will backfire on you.

0:34:13.100,0:34:15.806
So hands down, our number[br]one focus is make it as

0:34:15.806,0:34:18.107
easy to use as possible for[br]and

0:34:18.108,0:34:20.159
anything else on top was[br]polish.

0:34:20.159,0:34:20.596
Right here.

0:34:20.596,0:34:28.310
>> So so everybody says that[br]to focus on your product.

0:34:28.310,0:34:29.290
I'm also good at that.

0:34:29.300,0:34:32.300
I love to do a project and I[br]love to make it the best.

0:34:32.300,0:34:35.100
But we are to that certain[br]point that we are focused on

0:34:35.100,0:34:37.509
our product but we don't get[br]like constantly right?

0:34:37.510,0:34:38.270
Sorry.

0:34:38.270,0:34:41.639
So second thing so how much[br]we should focus on product?

0:34:41.639,0:34:43.639
But because we should do now[br]marketing.

0:34:43.639,0:34:45.489
We should get somebody[br]customers and

0:34:45.489,0:34:47.978
like and like start talking[br]to customers but

0:34:47.980,0:34:50.179
when you are too focused on[br]your product.

0:34:50.179,0:34:52.940
Like users online have them.

0:34:52.940,0:34:55.460
Right, so what exactly do[br]you guys mean when you

0:34:55.460,0:34:57.856
are saying like focus only[br]on your product and

0:34:57.856,0:34:59.130
give the best product?

0:34:59.130,0:35:01.548
>> Okay.

0:35:01.548,0:35:04.955
So the question sort of is[br]how do we balance this

0:35:04.955,0:35:07.889
sort of thing where we wanna[br]be obsessed with

0:35:07.889,0:35:09.800
working on product.

0:35:09.800,0:35:11.630
Yet.[br]But all the other skills,

0:35:11.630,0:35:14.800
and sort of tasks that are[br]needed by a company,

0:35:14.800,0:35:15.509
like marketing and branding[br]and

0:35:15.510,0:35:17.730
all that stuff, and how, how[br]we sort of balance that.

0:35:17.730,0:35:20.100
And the thing is, like, it[br]starts off as you juggling,

0:35:20.100,0:35:22.540
like, tons of things[br]constantly in the air.

0:35:22.540,0:35:24.440
The thing is, if you're[br]working on products, like,

0:35:24.440,0:35:26.500
you should also always have[br]this flip side as

0:35:26.500,0:35:28.500
when you're talking to[br]users.

0:35:28.500,0:35:28.560
Right?[br]And for

0:35:28.560,0:35:31.900
us inside of Wufoo, the way[br]we got people to talk to

0:35:31.100,0:35:32.440
users is they just did[br]customer support.

0:35:32.440,0:35:34.680
And they got to see[br]firsthand, right away.

0:35:34.680,0:35:35.930
Whether that feature sucked[br]or not,

0:35:35.930,0:35:37.500
and also impacted everyone[br]else in the company,

0:35:37.500,0:35:39.420
because everyone had a[br]customer support shift.

0:35:39.420,0:35:42.170
So you have this sort of[br]social incentive to sort of

0:35:42.170,0:35:43.360
make everything work.

0:35:43.360,0:35:46.150
And so, like I said, there[br]should be no point where

0:35:46.150,0:35:47.230
you're only focused on[br]product.

0:35:47.230,0:35:49.340
You should always have time[br]where you work on product,

0:35:49.340,0:35:51.420
and then you see sort of[br]what users say, say to you.

0:35:51.420,0:35:53.270
And you should always have[br]this virtual, like,

0:35:53.270,0:35:54.420
feedback loop on there.

0:35:54.420,0:35:56.630
So be careful when you don't[br]have that.

0:35:56.630,0:35:58.340
Usually what ends up[br]happening, if you're lucky.

0:35:59.380,0:36:00.700
In terms of marketing and

0:36:00.700,0:36:03.700
sales, like, usually my[br]feeling is like, you have to

0:36:03.700,0:36:05.900
spend money on marketing and[br]advertising, all this stuff.

0:36:05.900,0:36:06.680
It's usually a tax you pay[br]because you

0:36:06.680,0:36:08.279
haven't made your product[br]remarkable.

0:36:08.280,0:36:09.510
Right?

0:36:09.510,0:36:11.630
Word of mouth growth is the[br]easiest kind of growth, and

0:36:11.630,0:36:14.260
it's how a lot of the great[br]companies sort of grow.

0:36:14.260,0:36:16.220
So figure out how to wait,[br]how to like,

0:36:16.220,0:36:18.120
have a story that people[br]want to tell.

0:36:18.120,0:36:18.830
About your product.

0:36:18.830,0:36:20.400
Where they're the most[br]interesting person at

0:36:20.400,0:36:21.660
the dinner table, right?

0:36:21.660,0:36:23.819
And then that person is your[br]sales person, right?

0:36:23.820,0:36:25.462
That person is your sales[br]force for you.

0:36:25.462,0:36:28.106
Right here.

0:36:28.106,0:36:32.112
>> like, where do you find[br]crystal clear customer or

0:36:32.112,0:36:36.140
user need and the demand is[br]there is the right solution.

0:36:36.140,0:36:38.100
How do you communicate with[br]engineering and

0:36:38.100,0:36:41.710
designing team to make sure[br]that because sometimes

0:36:41.710,0:36:44.470
people in the team come up[br]with ideas, and but

0:36:44.470,0:36:45.620
still at the end of the day,

0:36:45.620,0:36:48.339
how can you make a decision[br]with where to go?

0:36:48.340,0:36:50.300
>> Oh, so how do you make a[br]decision on product?

0:36:51.580,0:36:53.000
And communicate that with[br]your sort of

0:36:53.000,0:36:54.350
engineering team when[br]there's like lots of

0:36:54.350,0:36:55.410
different directions to go?

0:36:57.290,0:36:58.950
My feeling is that.

0:37:01.150,0:37:03.180
So for us we just looked at[br]support.

0:37:03.180,0:37:04.180
It was really easy cuz

0:37:04.180,0:37:06.480
you often just saw what are[br]things that

0:37:06.480,0:37:08.410
people are having the most[br]amount of problems with?

0:37:08.410,0:37:09.980
Or people asking all the[br]time.

0:37:09.980,0:37:12.360
You cannot help but get[br]feature requests from

0:37:12.360,0:37:14.280
people no matter like[br]whatever opening or

0:37:14.280,0:37:15.940
orifice you have in your[br]product or app.

0:37:15.940,0:37:18.430
Like, people will like jam[br]feature requests in there.

0:37:18.430,0:37:19.210
So you're easily going to

0:37:19.210,0:37:20.450
know sort of what they sort[br]of what.

0:37:20.450,0:37:22.270
Your job as a product person[br]and

0:37:22.270,0:37:25.500
engineer is to not just do[br]what they say, because that

0:37:25.500,0:37:27.680
way, you'll just be a slave,[br]is to figure out,

0:37:27.680,0:37:30.930
sort of deeply, what are the[br]reasons why underlying those

0:37:30.930,0:37:33.410
things and sort of solve[br]that deep underlying reason.

0:37:33.410,0:37:34.606
The thing is that everyone[br]wants to

0:37:34.606,0:37:35.990
have a different way of.

0:37:35.990,0:37:38.390
To sort of go, then[br]ultimately it comes down to,

0:37:38.390,0:37:39.850
like, someone's gonna figure[br]something out.

0:37:39.850,0:37:43.180
But I also make the smallest[br]version of each little idea.

0:37:43.180,0:37:46.800
No longer than a week or two[br]weeks to build it out there.

0:37:46.800,0:37:46.980
And you can try them out and

0:37:46.980,0:37:48.890
see sort of what works and[br]don't work.

0:37:48.890,0:37:50.600
I think it's dangerous to

0:37:50.600,0:37:51.799
have multiple different[br]product directions that

0:37:51.800,0:37:53.879
requires lots of time to[br]sort of figure out.

0:37:55.430,0:37:56.490
Sam.

0:37:56.500,0:37:57.760
>> Related to that can you[br]tell the story of

0:37:57.760,0:38:00.910
how the king for a day thing[br]>> Yeah.

0:38:00.910,0:38:01.770
Okay.[br]So

0:38:03.920,0:38:06.760
so I don't like hackathons.

0:38:06.760,0:38:09.660
I think they sort of suck in[br]terms of ones done inside

0:38:09.660,0:38:10.210
of companies.

0:38:10.210,0:38:11.270
Because.

0:38:11.270,0:38:13.937
You spend like 48 hours[br]working on something really

0:38:13.937,0:38:16.125
hard that you're sort of[br]passionate about and

0:38:16.125,0:38:20.280
99% of them never make it to[br]production right, and

0:38:20.280,0:38:22.410
it's sort, sort of real like[br]super sad.

0:38:22.410,0:38:24.190
So for us we like flipped it[br]on it's head and

0:38:24.190,0:38:25.720
we came up with an idea that[br]we called king for

0:38:25.720,0:38:28.600
a day and it actually worked[br]over the weekend.

0:38:28.600,0:38:31.920
But how it worked is someone[br]randomly in the company got

0:38:31.920,0:38:34.660
drawn and they got to be the[br]king, and the king got to

0:38:34.660,0:38:36.399
tell everyone else what to[br]do on the product.

0:38:36.400,0:38:39.850
So everything that was[br]bothering them about Wufoo.

0:38:39.850,0:38:40.890
About the customer support[br]stuff, or

0:38:40.890,0:38:42.529
some feature they really[br]want to have built.

0:38:42.530,0:38:44.440
They've got the engineering[br]resources, the market

0:38:44.440,0:38:46.650
resources, the advertising[br]resources of everyone inside

0:38:46.650,0:38:48.670
of the company, to make it[br]sort of happen.

0:38:48.670,0:38:49.770
And of course, we'd work[br]with them to

0:38:49.770,0:38:53.200
figure out like what can be[br]actually done in 48 hours.

0:38:53.200,0:38:56.730
But we would do this one to[br]two times a year.

0:38:56.730,0:38:57.820
And it was like a huge hit[br]and

0:38:57.820,0:39:00.310
it was a boost to morale,[br]cuz what people most love.

0:39:00.310,0:39:02.810
It is like working on things[br]where it's like, oh,

0:39:02.810,0:39:04.200
I made a difference to the[br]app.

0:39:05.300,0:39:05.700
Right?

0:39:05.700,0:39:09.290
And so, for us, that's one[br]way that we would like

0:39:09.300,0:39:11.370
sort of divide time for like[br]product direction.

0:39:11.370,0:39:13.900
It's like some times the[br]people that work for

0:39:13.100,0:39:14.730
you, they have a strong[br]opinion about where it,

0:39:14.730,0:39:15.400
where it should go.

0:39:15.400,0:39:17.580
And it's a good way to sort[br]of democratize it

0:39:17.580,0:39:18.939
a little bit, by rotating it[br]around.

0:39:20.760,0:39:21.520
Yes.

0:39:21.520,0:39:23.759
>> You said you guys all[br]work from home.

0:39:23.760,0:39:25.450
Usually seems like a[br]nightmare.

0:39:25.450,0:39:27.509
In that office, how do you[br]make that work?

0:39:27.510,0:39:28.510
>> Okay, so we all work from[br]home.

0:39:28.510,0:39:29.540
So, I will tell you this.

0:39:29.540,0:39:31.610
We all still work within the[br]Tampa Bay area.

0:39:31.610,0:39:32.760
We would allow anybody to

0:39:32.760,0:39:34.510
work from any where but[br]usually.

0:39:34.510,0:39:35.904
As we tried to recruit[br]them,they sort of

0:39:35.904,0:39:37.820
meet our team, and they just[br]decide,

0:39:37.820,0:39:39.500
okay we just want to come[br]and move here anyway.

0:39:41.970,0:39:46.562
Remote working, is[br]especially tricky,

0:39:46.563,0:39:48.350
a lot of people like to[br]romanticize it,

0:39:48.350,0:39:50.430
especially people, who are[br]like employees.

0:39:50.430,0:39:54.109
But the thing is, An office[br]gives you a lot of, sort of,

0:39:54.110,0:39:56.870
benefits, right, and[br]efficiencies that you now

0:39:56.870,0:39:58.880
have to compensate for when[br]you remote working.

0:39:58.880,0:40:00.770
But remote working also has[br]these other sort of

0:40:00.770,0:40:02.590
efficiencies in place, for[br]example,

0:40:02.590,0:40:04.870
I don't have to worry about[br]my employees losing two

0:40:04.870,0:40:07.569
hours of their day to[br]commuting, for instance.

0:40:07.570,0:40:08.980
So the biggest thing that we[br]had to do for

0:40:08.980,0:40:11.860
remote working is to respect[br]people's time.

0:40:11.860,0:40:14.500
And so the way we had it set[br]up is we actually had

0:40:14.500,0:40:16.470
a four and a half day work[br]week at Wufoo.

0:40:16.470,0:40:18.839
Half day on Friday was for[br]all the meetings and stuff.

0:40:18.840,0:40:20.830
We said like no business[br]deal,

0:40:20.830,0:40:22.990
meetings, no talking with[br]other outside parties.

0:40:22.990,0:40:25.569
They all have to be done on[br]Friday, on that half day.

0:40:25.570,0:40:27.320
It couldn't be done in the[br]middle of the week.

0:40:27.320,0:40:29.420
And then also one day of[br]everyone was

0:40:29.420,0:40:30.780
already dedicated customer[br]support.

0:40:30.780,0:40:32.910
So everyone in our company[br]effectively only had

0:40:32.910,0:40:35.170
three days each week to[br]actually build or

0:40:35.170,0:40:36.570
work on whatever they were[br]doing.

0:40:36.570,0:40:38.370
But I actually firmly[br]believe that if you

0:40:38.370,0:40:41.270
have three solid days,[br]right, eight to ten hours,

0:40:41.270,0:40:43.230
where you are only working[br]on what you need to build,

0:40:43.230,0:40:45.350
you can get a ton of shit[br]done.

0:40:45.350,0:40:47.600
And so what we said was,

0:40:47.600,0:40:48.770
you have to respect[br]everyone's time during that

0:40:48.770,0:40:51.130
three day period, if they're[br]in that three day period.

0:40:51.130,0:40:53.270
And what we came up with is[br]a 15 minute rule and the way

0:40:53.270,0:40:56.580
it worked is you can have a[br]discussion like a chat or

0:40:56.580,0:40:59.380
a phone call or whatever[br]with someone but it could go

0:40:59.380,0:41:01.640
no longer than 15 minutes so[br]if you have some complicated

0:41:01.640,0:41:04.490
issue that you couldn't[br]figure out we'd say,

0:41:04.500,0:41:07.320
at 15 minutes you are to[br]immediately table that item.

0:41:07.320,0:41:10.500
Right and have us discuss it[br]on Friday and

0:41:10.500,0:41:12.610
you'd move on to the next[br]item on your list right,

0:41:12.610,0:41:13.690
the enhanced productivity.

0:41:15.500,0:41:18.430
More often than not I, I[br]would say 90% of the time

0:41:18.430,0:41:20.450
the item never got brought[br]up on Friday because usually

0:41:20.450,0:41:22.810
what would happen is people[br]would sleep on it and then

0:41:22.810,0:41:25.730
they'd just magically say[br]like, hey I found a solution

0:41:25.730,0:41:27.740
or like hey that's not a big[br]problem whatsoever.

0:41:29.300,0:41:30.760
Because most problems inside[br]of companies,

0:41:30.760,0:41:34.290
they don't need to be solved[br]in real time, or right away.

0:41:34.290,0:41:36.990
The only things are like[br]when the site is down, or

0:41:36.990,0:41:38.310
payments aren't working.

0:41:38.310,0:41:39.490
Right?[br]Everything outside of

0:41:39.490,0:41:43.479
that is basically kind of[br]luxury, so

0:41:43.480,0:41:45.520
focus on your primaries as[br]much as possible.

0:41:45.520,0:41:46.419
And, as a result,

0:41:46.419,0:41:48.444
our ten person team did far[br]more than many,

0:41:48.445,0:41:51.150
many other companies, as a[br]result.

0:41:51.150,0:41:55.400
But, it takes extra work to[br]make remote working happen.

0:41:55.400,0:41:57.700
We are an extremely[br]disciplined sort of team.

0:41:57.700,0:41:58.370
And I would have to say

0:41:59.790,0:42:02.850
there's almost not many YC[br]companies that actually have

0:42:02.850,0:42:04.299
been able to replicate sort[br]of what we do.

0:42:04.300,0:42:06.250
I think there's only two[br]other companies in YC that

0:42:06.250,0:42:09.600
sort of have the same sort[br]of discipline working style.

0:42:09.600,0:42:11.560
It takes more work in a very[br]different fashion.

0:42:11.560,0:42:12.100
Right?

0:42:12.100,0:42:14.779
An office allows you to be a[br]little bit lazier, right,

0:42:14.780,0:42:16.620
in terms of all these things[br]around productivity.

0:42:20.880,0:42:21.880
Okay.[br]Over here.

0:42:21.880,0:42:23.330
>> Right.

0:42:23.330,0:42:26.440
Just to go off that question[br]as the leader of the team

0:42:26.440,0:42:29.891
how did you manage to[br]instill a company culture

0:42:29.891,0:42:30.948
but and also the count,

0:42:30.948,0:42:34.180
accountability the employees[br]especially cuz.

0:42:34.180,0:42:35.700
Working space.

0:42:35.700,0:42:38.700
>> Okay, how do we, how do[br]we set up accountability for

0:42:38.700,0:42:39.390
employees?

0:42:39.390,0:42:40.230
As, as a manager.

0:42:40.230,0:42:43.530
Alright, so, at Wufoo we[br]were profitable nine months

0:42:43.530,0:42:46.920
after launch, so, we had[br]profit sharing, right?

0:42:46.920,0:42:49.510
And so, it makes pretty[br]simple and clear.

0:42:49.510,0:42:53.110
It, it would be a multiple[br]of whatever bonus pull that

0:42:53.110,0:42:54.420
we sort of had, and

0:42:54.420,0:42:56.720
the performance measures[br]would be based on sort of,

0:42:56.720,0:42:58.350
how they did in customer[br]support.

0:42:58.350,0:42:59.880
All right, on their duties[br]there, and sort of

0:42:59.880,0:43:02.810
what they said they were[br]wanting to accomplish or do.

0:43:02.810,0:43:04.790
I don't like process, and I[br]don't like lots of

0:43:04.790,0:43:07.110
tools to help get people to[br]be productive.

0:43:07.110,0:43:08.547
So the only thing that we[br]had for

0:43:08.547,0:43:10.400
helping people manage, like,

0:43:10.400,0:43:12.270
sort of their projects, is[br]to-do lists.

0:43:12.270,0:43:13.810
And that is, like, simple[br]text files that we

0:43:13.810,0:43:15.490
shared in our Dropbox[br]account.

0:43:15.490,0:43:17.379
Each person had their name[br]on it, and you got to

0:43:17.380,0:43:19.710
see every time somebody[br]updated their to-do list.

0:43:19.710,0:43:21.600
What we said is every single[br]night just

0:43:21.600,0:43:22.650
set everything that you did[br]that day.

0:43:22.650,0:43:23.560
Right?[br]And then on Friday,

0:43:23.560,0:43:24.690
we would just go over.

0:43:24.690,0:43:25.770
Okay, this is what you

0:43:25.770,0:43:26.970
said last week that you're[br]gonna do.

0:43:26.970,0:43:28.240
This is what you actually[br]got done, or

0:43:28.240,0:43:29.350
the sort of the problems at[br]hand.

0:43:29.350,0:43:31.370
And it's super simple,[br]right?

0:43:31.370,0:43:33.650
It creates this like nice[br]written trail for

0:43:33.650,0:43:35.550
how to sort of handle stuff,[br]right?

0:43:35.550,0:43:37.720
And I don't have to worry[br]about managing them, right?

0:43:37.720,0:43:39.299
They sorta set the tone for

0:43:39.300,0:43:41.610
how they want to be sort of[br]assessed.

0:43:41.610,0:43:42.550
And it makes it really[br]simple.

0:43:42.550,0:43:45.370
And for people who are[br]excellent at what they do.

0:43:45.370,0:43:47.490
Right?[br]It works very, very well.

0:43:47.500,0:43:48.900
And then when you actually[br]have problems,

0:43:48.900,0:43:50.570
it's very easy to fire[br]people.

0:43:50.570,0:43:51.760
I was fortunate that I

0:43:51.760,0:43:54.110
never had to fire anyone at[br]Wufoo, right?

0:43:54.110,0:43:55.490
But we were able to correct[br]a lot of

0:43:55.490,0:43:57.529
people's behavior very, very[br]quickly.

0:43:57.530,0:43:58.750
Cuz we just kinda look at[br]this, and

0:43:58.750,0:44:00.160
it's like look, this is your[br]pattern of behavior.

0:44:00.160,0:44:02.509
You finish a fraction of the[br]items on your list.

0:44:02.510,0:44:03.900
You do most of

0:44:03.900,0:44:05.810
the items at the last second[br]right before Friday.

0:44:05.810,0:44:07.240
That's a problem, you've[br]gotta manage your time

0:44:07.240,0:44:10.259
better and this is evidence[br]that you've provided to us.

0:44:10.260,0:44:12.420
All we have to do is sort of[br]describe it back to you and

0:44:12.420,0:44:14.650
because everyone in the[br]company sort of sees it,

0:44:14.650,0:44:16.000
right, there's social[br]pressure that's put

0:44:16.000,0:44:17.870
into place that helps make[br]it all sort of happen.

0:44:19.510,0:44:20.810
Right here.

0:44:20.810,0:44:22.930
>> How did you hire people[br]that, you know,

0:44:22.930,0:44:25.629
you felt would be able to[br]work with in this kind of

0:44:25.630,0:44:28.216
environment that's, that's[br]>> So,

0:44:28.216,0:44:31.720
how do you hire people that[br]can work, remotely?

0:44:31.720,0:44:33.990
And then sort of work in[br]this sort of fashion.

0:44:33.990,0:44:35.879
So, pretty easily,

0:44:35.880,0:44:38.170
you have them work on a side[br]project for you.

0:44:38.170,0:44:39.330
So you contract them out,[br]and

0:44:39.330,0:44:40.920
they have to work remotely.

0:44:40.920,0:44:42.610
As such, usually, the[br]projects I like to

0:44:42.610,0:44:45.160
have them work on is about a[br]month long, right?

0:44:45.160,0:44:46.839
I could do things much[br]faster for a week.

0:44:46.840,0:44:48.100
But usually get a good sense[br]of,

0:44:48.100,0:44:49.870
like, how well people sort[br]of manage themselves, and

0:44:49.870,0:44:51.930
work on things from a[br]project like that.

0:44:51.930,0:44:53.750
So that was always the first[br]assessment.

0:44:53.750,0:44:55.560
Like we never did anything[br]just by interviews.

0:44:56.840,0:44:59.740
The other thing we had to,[br]sort of, screen them for

0:44:59.740,0:45:01.350
is their ability to do[br]customer support.

0:45:01.350,0:45:02.190
Because not any,

0:45:02.190,0:45:05.980
every engineer sort of has[br]the empathy skills to

0:45:05.980,0:45:08.130
handle that stress.

0:45:08.130,0:45:12.100
So sometimes I would have[br]people write breakup letters

0:45:12.100,0:45:14.440
to me, right, in an[br]interview, just like hey,

0:45:14.440,0:45:17.310
pretend you have to break up[br]with me you have 15 minutes

0:45:17.310,0:45:19.320
to write in on there and you[br]can only write it by hand,

0:45:19.320,0:45:21.600
what are you gonna say?

0:45:21.600,0:45:23.740
And so you get a good sense[br]of sort of their writing

0:45:23.740,0:45:25.669
skills, because like 90% of[br]what you do in customer

0:45:25.670,0:45:27.940
support is tell them bad[br]news, like, oh, we don't

0:45:27.940,0:45:30.710
support that feature, sorry[br]that's not gonna work, or.

0:45:30.710,0:45:32.700
It's not gonna be available.

0:45:32.700,0:45:33.985
And so people have to have[br]sort of tacked at that.

0:45:33.985,0:45:36.300
>> How 'bout one more[br]question?

0:45:36.300,0:45:38.650
>> One question, right here[br]the glasses.

0:45:38.650,0:45:40.430
>> So, since Wiki has all[br]these like tricks and

0:45:40.430,0:45:42.529
experiments that have really[br]helped the company.

0:45:42.530,0:45:44.950
Do you INAUDIBLE] of ones[br]that didn't work out?

0:45:44.950,0:45:46.120
>> Have all these tricks and

0:45:46.120,0:45:47.240
experiments to help the[br]company,

0:45:47.240,0:45:49.359
are there any ones that[br]didn't work out?

0:45:50.470,0:45:51.520
All right, I'll talk about[br]one.

0:45:51.520,0:45:54.770
So one of the things that we[br]did early on to try to

0:45:54.770,0:45:56.940
motivate ourselves was try[br]to get,

0:45:58.530,0:46:01.300
like we understood this idea[br]of like crunch mode and

0:46:01.300,0:46:02.820
that it's really bad for[br]people.

0:46:02.820,0:46:04.370
Like if you're doing the[br]subscription business you

0:46:04.370,0:46:06.230
need people to last for the[br]long term, and

0:46:06.230,0:46:08.400
video games a lot of times[br]they like crunch people.

0:46:08.400,0:46:09.660
All, all the time.

0:46:09.660,0:46:10.859
For like a specific deadline[br]or

0:46:10.860,0:46:12.610
have multiple sprints every[br]two weeks and

0:46:12.610,0:46:13.930
you have to shoot up to this[br]deadline and

0:46:13.930,0:46:15.290
it's like exhausting.

0:46:15.290,0:46:17.900
And so, because, when it's[br]happening it's like you

0:46:17.900,0:46:18.560
might get an increase in[br]productivity but

0:46:18.560,0:46:20.299
the recovery period that you[br]need for people is

0:46:20.300,0:46:23.770
always greater, right, than[br]the productivity you gain.

0:46:23.770,0:46:24.630
And at a company where you

0:46:24.630,0:46:25.810
need everyone doing customer[br]support and

0:46:25.810,0:46:27.420
being on their game and[br]constantly put, pushed out

0:46:27.420,0:46:29.910
features you don't have time[br]for recovery.

0:46:29.910,0:46:33.490
So, we were thinking about,[br]okay, we want to build like,

0:46:33.490,0:46:36.740
a company vacation into how[br]Wufoo sort of works to

0:46:36.740,0:46:38.299
reward our users every[br]single year.

0:46:38.300,0:46:38.830
And we said okay,

0:46:38.830,0:46:40.330
if the vacation is sort of[br]built in there for

0:46:40.330,0:46:43.290
the recovery, we could have[br]one crunch period, right,

0:46:43.300,0:46:44.300
before that vacation set up.

0:46:44.300,0:46:46.470
And we'll just only do[br]customer support that will

0:46:46.470,0:46:48.430
just sort of scale with[br]people.

0:46:48.430,0:46:49.259
So.

0:46:49.260,0:46:55.700
The way we did the very[br]first crunch mode

0:46:55.700,0:46:57.380
is that it was just between[br]the three founders.

0:46:57.380,0:47:00.900
And we had each of us draw a[br]ten-item to-do list.

0:47:00.900,0:47:02.140
That would be fairly[br]aggressive.

0:47:02.140,0:47:02.890
And the first person to

0:47:02.890,0:47:05.640
get through seven of their[br]items would win.

0:47:05.640,0:47:07.290
And the the last person to

0:47:07.300,0:47:08.630
get through seven of their[br]items.

0:47:08.630,0:47:10.950
Would become what we called[br]trippage.

0:47:10.950,0:47:13.180
And trippage meant that you[br]carried other

0:47:13.180,0:47:15.629
person's luggage and got[br]people drinks when you're on

0:47:15.630,0:47:16.500
the company vacation.

0:47:17.560,0:47:21.540
So we did that and during[br]that period,

0:47:21.540,0:47:23.140
everyone was like pretty[br]excited about it and

0:47:23.140,0:47:24.850
motivated, and only the[br]winner got to

0:47:24.850,0:47:27.640
choose the next company[br]vacation the following year.

0:47:27.640,0:47:31.240
And then all the sudden Ryan[br]had basically poorly

0:47:31.240,0:47:33.490
estimated the items on his[br]list.

0:47:33.500,0:47:36.400
And he realized very quickly[br]I'm going to fucking lose.

0:47:36.400,0:47:38.950
And so he was just like I[br]give up.

0:47:38.950,0:47:40.250
And he just sort of stopped.

0:47:40.250,0:47:42.410
So crunch mode turned out to[br]be blah mode for

0:47:42.410,0:47:44.339
him because he knew he was[br]gonna lose and

0:47:44.340,0:47:46.300
became pretty demoralized.

0:47:46.300,0:47:47.380
So as a result of doing[br]that, we

0:47:47.380,0:47:50.130
decided not to do it in that[br]similar fashion anymore.

0:47:50.130,0:47:52.610
So, good idea that we like[br]to talk about.

0:47:52.610,0:47:53.980
But is one that we, we never[br]did again.

0:47:55.120,0:47:55.750
All right, guys.

0:47:55.750,0:47:56.930
Thanks a lot.

0:47:56.930,0:47:58.928
You can email me at kevin@