1
00:00:00,961 --> 00:00:02,944
I'm really excited to share with you

2
00:00:02,944 --> 00:00:05,534
some findings that really surprise me

3
00:00:05,534 --> 00:00:08,085
about what makes companies 
succeed the most,

4
00:00:08,085 --> 00:00:11,987
what factors actually matter the most
for startup success.

5
00:00:13,063 --> 00:00:15,212
I believe that the startup organization

6
00:00:15,212 --> 00:00:18,970
is one of the greatest forms
to make the world a better place.

7
00:00:19,630 --> 00:00:22,756
If you take a group of people 
with the right equity incentives

8
00:00:22,756 --> 00:00:24,635
and organize them in a startup,

9
00:00:24,635 --> 00:00:28,609
you can unlock human potential
in a way never before possible.

10
00:00:28,609 --> 00:00:31,139
You get them to achieve 
unbelievable things.

11
00:00:31,679 --> 00:00:33,799
But if the startup 
organization is so great,

12
00:00:33,799 --> 00:00:35,281
why do so many fail?

13
00:00:35,281 --> 00:00:36,878
That's what I wanted to find out.

14
00:00:36,878 --> 00:00:39,363
I wanted to find out what
actually matters most

15
00:00:39,363 --> 00:00:40,573
for startup success.

16
00:00:40,573 --> 00:00:42,782
And I wanted to try 
to be systematic about it,

17
00:00:42,782 --> 00:00:45,935
avoid some of my instincts
and maybe misperceptions I have

18
00:00:45,935 --> 00:00:48,201
from so many companies
I've seen over the years.

19
00:00:48,201 --> 00:00:49,428
I wanted to know this

20
00:00:49,428 --> 00:00:52,476
because I've been starting businesses
since I was 12 years old

21
00:00:52,476 --> 00:00:55,131
when I sold candy at the bus stop
in junior high school,

22
00:00:55,131 --> 00:00:57,442
to high school, when I made
solar energy devices,

23
00:00:57,442 --> 00:00:59,229
to college, when I made loudspeakers.

24
00:00:59,229 --> 00:01:02,255
And when I graduated from college,
I started software companies.

25
00:01:02,255 --> 00:01:04,064
And 20 years ago,
I started Idealab,

26
00:01:04,064 --> 00:01:07,305
and in the last 20 years,
we started more than 100 companies,

27
00:01:07,305 --> 00:01:09,629
many successes, and many big failures.

28
00:01:09,629 --> 00:01:11,458
We learned a lot from those failures.

29
00:01:11,988 --> 00:01:14,735
So I tried to look across what factors

30
00:01:14,735 --> 00:01:17,898
accounted the most for company
success and failure.

31
00:01:17,898 --> 00:01:19,458
So I looked at these five.

32
00:01:19,458 --> 00:01:20,418
First, the idea.

33
00:01:20,418 --> 00:01:22,574
I used to think that 
the idea was everything.

34
00:01:22,574 --> 00:01:24,881
I named my company Idealab
for how much I worship

35
00:01:24,881 --> 00:01:27,474
the "aha!" moment when you first
come up with the idea.

36
00:01:27,474 --> 00:01:28,385
But then over time,

37
00:01:28,385 --> 00:01:31,970
I came to think that maybe the team,
the execution, adaptability,

38
00:01:31,970 --> 00:01:34,121
that mattered even more than the idea.

39
00:01:34,121 --> 00:01:38,483
I never thought I'd be quoting
boxer Mike Tyson on the TED stage,

40
00:01:38,483 --> 00:01:40,342
but he once said,

41
00:01:40,342 --> 00:01:44,945
"Everybody has a plan, until they get
punched in the face." (Laughter)

42
00:01:44,945 --> 00:01:47,864
And I think that's so true
about business as well.

43
00:01:47,864 --> 00:01:50,140
So much about a team's execution


44
00:01:50,140 --> 00:01:53,900
is its ability to adapt to getting punched
in the face by the customer.

45
00:01:53,900 --> 00:01:55,588
The customer is the true reality.

46
00:01:55,588 --> 00:01:57,039
And that's why I came to think

47
00:01:57,039 --> 00:01:59,556
that the team maybe
was the most important thing.

48
00:02:00,036 --> 00:02:02,148
Then I started looking
at the business model.

49
00:02:02,148 --> 00:02:05,398
Does the company have a very clear path
generating customer revenues?

50
00:02:05,398 --> 00:02:07,502
That started rising to the top
in my thinking

51
00:02:07,502 --> 00:02:09,819
about maybe what mattered
most for success.

52
00:02:09,819 --> 00:02:11,216
Then I looked at the funding.

53
00:02:11,216 --> 00:02:13,916
Sometimes companies received
intense amounts of funding.

54
00:02:13,916 --> 00:02:15,754
Maybe that's the most important thing?

55
00:02:15,754 --> 00:02:17,300
And then of course,
the timing.

56
00:02:17,300 --> 00:02:20,110
Is the idea way too early and
the world's not ready for it?

57
00:02:20,110 --> 00:02:23,557
Is it early, as in, you're in advance
and you have to educate the world?

58
00:02:23,557 --> 00:02:24,366
Is it just right?

59
00:02:24,366 --> 00:02:27,229
Or is it too late, and there's
already too many competitors?

60
00:02:27,229 --> 00:02:29,818
So I tried to look very carefully
at these five factors

61
00:02:29,818 --> 00:02:30,869
across many companies.

62
00:02:30,869 --> 00:02:33,064
And I looked across all 100
Idealab companies,

63
00:02:33,064 --> 00:02:34,518
and 100 non-Idealab companies

64
00:02:34,518 --> 00:02:37,154
to try and come up with 
something scientific about it.

65
00:02:37,574 --> 00:02:39,899
So first, on these Idealab companies,

66
00:02:39,899 --> 00:02:41,772
the top five companies --

67
00:02:41,772 --> 00:02:45,188
Citysearch, CarsDirect, GoTo,
NetZero, Tickets.com --

68
00:02:45,188 --> 00:02:47,262
those all became billion-dollar successes.

69
00:02:47,262 --> 00:02:49,175
And the five companies on the bottom --

70
00:02:49,175 --> 00:02:51,985
Z.com, Insider Pages, MyLife,
Desktop Factory, Peoplelink --

71
00:02:51,985 --> 00:02:54,376
we all had high hopes for,
but didn't succeed.

72
00:02:54,888 --> 00:02:57,833
So I tried to rank across all 
of those attributes

73
00:02:57,833 --> 00:03:01,075
how I felt those companies scored
on each of those dimensions.

74
00:03:01,075 --> 00:03:04,540
And then for non-Idealab companies,
I looked at wild successes,

75
00:03:04,540 --> 00:03:08,122
like Airbnb and Instagram and Uber
and Youtube and LinkedIn.

76
00:03:08,133 --> 00:03:09,385
And some failures:

77
00:03:09,385 --> 00:03:11,185
Webvan, Kozmo, Pets.com

78
00:03:11,185 --> 00:03:12,465
Flooz and Friendster.

79
00:03:12,465 --> 00:03:14,469
The bottom companies had intense funding,

80
00:03:14,469 --> 00:03:16,548
they even had business models
in some cases,

81
00:03:16,548 --> 00:03:17,695
but they didn't succeed.

82
00:03:17,695 --> 00:03:20,556
I tried to look at what factors
actually accounted the most

83
00:03:20,556 --> 00:03:23,109
for success and failure across
all of these companies,

84
00:03:23,109 --> 00:03:24,938
and the results really surprised me.

85
00:03:25,338 --> 00:03:27,008
The number one thing was timing.

86
00:03:27,638 --> 00:03:30,140
Timing accounted for 42 percent

87
00:03:30,140 --> 00:03:32,640
of the difference
between success and failure.

88
00:03:32,640 --> 00:03:34,742
Team and execution came in second,

89
00:03:34,742 --> 00:03:35,564
and the idea,

90
00:03:35,564 --> 00:03:38,481
the differentiability of the idea,
the uniqueness of the idea,

91
00:03:38,481 --> 00:03:39,884
that actually came in third.

92
00:03:39,884 --> 00:03:41,849
Now, this isn't absolutely definitive,

93
00:03:41,849 --> 00:03:44,066
it's not to say that 
the idea isn't important,

94
00:03:44,066 --> 00:03:47,706
but it very much surprised me that
the idea wasn't the most important thing.

95
00:03:47,706 --> 00:03:50,306
Sometimes it mattered more when 
it was actually timed.

96
00:03:50,306 --> 00:03:53,681
The last two, business model and funding,
made sense to me actually.

97
00:03:53,681 --> 00:03:56,048
I think business model
makes sense to be that low

98
00:03:56,048 --> 00:03:58,428
because you can start out
without a business model

99
00:03:58,428 --> 00:04:01,895
and add one later if your customers
are demanding what you're creating.

100
00:04:01,895 --> 00:04:03,301
And funding, I think as well,

101
00:04:03,301 --> 00:04:06,076
if you're underfunded at first
but you're gaining traction,

102
00:04:06,076 --> 00:04:07,365
especially in today's age,

103
00:04:07,365 --> 00:04:09,426
it's very, very easy to get
intense funding.

104
00:04:09,426 --> 00:04:12,546
So now let me give you some specific
examples about each of these.

105
00:04:12,546 --> 00:04:15,459
So take a wild success like Airbnb
that everybody knows about.

106
00:04:15,459 --> 00:04:18,524
Well, that company was famously
passed on by many smart investors

107
00:04:18,524 --> 00:04:19,666
because people thought,

108
00:04:19,666 --> 00:04:22,719
"No one's going to rent out a space
in their home to a stranger."

109
00:04:22,719 --> 00:04:24,444
Of course, people proved that wrong.

110
00:04:24,444 --> 00:04:26,298
But one of the reasons it succeeded,

111
00:04:26,298 --> 00:04:29,283
aside from a good business model,
a good idea, great execution,

112
00:04:29,283 --> 00:04:30,020
is the timing.

113
00:04:30,020 --> 00:04:33,299
That company came out
right during the height of the recession

114
00:04:33,299 --> 00:04:35,117
when people really needed extra money,

115
00:04:35,117 --> 00:04:36,931
and that maybe helped people overcome

116
00:04:36,931 --> 00:04:39,777
their objection to renting out
their own home to a stranger.

117
00:04:39,777 --> 00:04:40,896
Same thing with Uber.

118
00:04:40,896 --> 00:04:41,949
Uber came out,

119
00:04:41,949 --> 00:04:44,110
incredible company,
incredible business model,

120
00:04:44,110 --> 00:04:45,174
great execution, too.

121
00:04:45,174 --> 00:04:46,656
But the timing was so perfect

122
00:04:46,656 --> 00:04:48,805
for their need to get drivers
into the system.

123
00:04:48,805 --> 00:04:51,927
Drivers were looking for extra money;
it was very, very important.

124
00:04:51,927 --> 00:04:55,705
Some of our early successes, Citysearch,
came out when people needed web pages.

125
00:04:55,705 --> 00:04:58,182
GoTo.com, which we announced
actually at TED in 1998,

126
00:04:58,182 --> 00:05:01,513
was when companies were looking for
cost-effective ways to get traffic.

127
00:05:01,513 --> 00:05:03,085
We thought the idea was so great,

128
00:05:03,085 --> 00:05:05,846
but actually, the timing was probably
maybe more important.

129
00:05:05,846 --> 00:05:07,284
And then some of our failures.

130
00:05:07,284 --> 00:05:10,761
We started a company called Z.com,
it was an online entertainment company.

131
00:05:10,761 --> 00:05:12,204
We were so excited about it --

132
00:05:12,204 --> 00:05:14,746
we raised enough money,
we had a great business model,

133
00:05:14,746 --> 00:05:17,994
we even signed incredibly great
Hollywood talent to join the company.

134
00:05:17,994 --> 00:05:20,382
But broadband penetration
was too low in 1999-2000.

135
00:05:20,382 --> 00:05:22,538
It was too hard to watch
video content online,

136
00:05:22,538 --> 00:05:25,347
you had to put codecs in your browser
and do all this stuff,

137
00:05:25,347 --> 00:05:27,978
and the company eventually
went out of business in 2003.

138
00:05:27,978 --> 00:05:29,074
Just two years later,

139
00:05:29,074 --> 00:05:31,959
when the codec problem
was solved by Adobe Flash

140
00:05:31,959 --> 00:05:35,633
and when broadband penetration
crossed 50 percent in America,

141
00:05:35,633 --> 00:05:37,590
YouTube was perfectly timed.

142
00:05:37,590 --> 00:05:39,319
Great idea, but unbelievable timing.

143
00:05:39,319 --> 00:05:42,759
In fact, YouTube didn't even have
a business model when it first started.

144
00:05:42,759 --> 00:05:45,042
It wasn't even certain that 
that would work out.

145
00:05:45,042 --> 00:05:47,177
But that was beautifully, 
beautifully timed.

146
00:05:47,177 --> 00:05:48,728
So what I would say, in summary,

147
00:05:48,728 --> 00:05:51,513
is execution definitely matters a lot.

148
00:05:51,513 --> 00:05:52,757
The idea matters a lot.

149
00:05:52,757 --> 00:05:54,516
But timing might matter even more.

150
00:05:54,516 --> 00:05:56,562
And the best way to really assess timing

151
00:05:56,562 --> 00:05:59,179
is to really look at whether 
consumers are really ready

152
00:05:59,179 --> 00:06:00,715
for what you have to offer them.

153
00:06:00,715 --> 00:06:02,677
And to be really, really honest about it,

154
00:06:02,677 --> 00:06:04,938
not be in denial about
any results that you see,

155
00:06:04,938 --> 00:06:08,149
because if you have something you love,
you want to push it forward,

156
00:06:08,149 --> 00:06:11,219
but you have to be very, very honest
about that factor on timing.

157
00:06:11,219 --> 00:06:12,089
As I said earlier,

158
00:06:12,089 --> 00:06:15,478
I think startups can change the world
and make the world a better place.

159
00:06:15,478 --> 00:06:16,876
I hope some of these insights

160
00:06:16,876 --> 00:06:19,528
can maybe help you
have a slightly higher success ratio,

161
00:06:19,528 --> 00:06:21,748
and thus make something great
come to the world

162
00:06:21,748 --> 00:06:23,594
that wouldn't have happened otherwise.

163
00:06:23,594 --> 00:06:25,931
Thank you very much,
you've been a great audience.

164
00:06:25,931 --> 00:06:27,425
(Applause)