1
00:00:00,586 --> 00:00:02,649
I'm going to talk a little bit about

2
00:00:02,649 --> 00:00:06,823
strategy and its relationship with technology.

3
00:00:06,823 --> 00:00:09,718
We tend to think of business strategy

4
00:00:09,718 --> 00:00:11,681
as being a rather abstract body

5
00:00:11,681 --> 00:00:13,396
of essentially economic thought,

6
00:00:13,396 --> 00:00:15,026
perhaps rather timeless.

7
00:00:15,026 --> 00:00:16,966
I'm going to argue that, in fact,

8
00:00:16,966 --> 00:00:19,521
business strategy has always been premised

9
00:00:19,521 --> 00:00:21,940
on assumptions about technology,

10
00:00:21,940 --> 00:00:23,913
that those assumptions are changing,

11
00:00:23,913 --> 00:00:26,630
and, in fact, changing quite dramatically,

12
00:00:26,630 --> 00:00:29,702
and that therefore what that will drive us to

13
00:00:29,702 --> 00:00:32,605
is a different concept of what we mean

14
00:00:32,605 --> 00:00:35,043
by business strategy.

15
00:00:35,043 --> 00:00:36,524
Let me start, if I may,

16
00:00:36,524 --> 00:00:39,294
with a little bit of history.

17
00:00:39,294 --> 00:00:41,038
The idea of strategy in business

18
00:00:41,038 --> 00:00:44,338
owes its origins to two intellectual giants:

19
00:00:44,338 --> 00:00:46,472
Bruce Henderson, the founder of BCG,

20
00:00:46,472 --> 00:00:50,244
and Michael Porter, professor
at the Harvard Business School.

21
00:00:50,244 --> 00:00:53,076
Henderson's central idea was what you might call

22
00:00:53,076 --> 00:00:56,067
the Napoleonic idea of concentrating mass

23
00:00:56,067 --> 00:00:58,936
against weakness, of overwhelming the enemy.

24
00:00:58,936 --> 00:01:00,729
What Henderson recognized was that,

25
00:01:00,729 --> 00:01:02,125
in the business world,

26
00:01:02,125 --> 00:01:04,780
there are many phenomena which are characterized

27
00:01:04,780 --> 00:01:06,815
by what economists would call increasing returns --

28
00:01:06,815 --> 00:01:08,178
scale, experience.

29
00:01:08,178 --> 00:01:10,052
The more you do of something,

30
00:01:10,052 --> 00:01:12,449
disproportionately the better you get.

31
00:01:12,449 --> 00:01:15,412
And therefore he found a logic for investing

32
00:01:15,412 --> 00:01:18,154
in such kinds of overwhelming mass

33
00:01:18,154 --> 00:01:20,707
in order to achieve competitive advantage.

34
00:01:20,707 --> 00:01:22,775
And that was the first introduction

35
00:01:22,775 --> 00:01:25,439
of essentially a military concept of strategy

36
00:01:25,439 --> 00:01:28,421
into the business world.

37
00:01:28,421 --> 00:01:31,156
Porter agreed with that premise,

38
00:01:31,156 --> 00:01:32,707
but he qualified it.

39
00:01:32,707 --> 00:01:35,696
He pointed out, correctly, that that's all very well,

40
00:01:35,696 --> 00:01:39,571
but businesses actually have multiple steps to them.

41
00:01:39,571 --> 00:01:41,144
They have different components,

42
00:01:41,144 --> 00:01:43,676
and each of those components might be driven

43
00:01:43,676 --> 00:01:45,461
by a different kind of strategy.

44
00:01:45,461 --> 00:01:47,789
A company or a business
might actually be advantaged

45
00:01:47,789 --> 00:01:51,267
in some activities but disadvantaged in others.

46
00:01:51,267 --> 00:01:53,368
He formed the concept of the value chain,

47
00:01:53,368 --> 00:01:56,438
essentially the sequence of steps with which

48
00:01:56,438 --> 00:01:59,590
a, shall we say, raw material, becomes a component,

49
00:01:59,590 --> 00:02:00,953
becomes assembled into a finished product,

50
00:02:00,953 --> 00:02:03,594
and then is distributed, for example,

51
00:02:03,594 --> 00:02:06,010
and he argued that advantage accrued

52
00:02:06,010 --> 00:02:07,551
to each of those components,

53
00:02:07,551 --> 00:02:09,265
and that the advantage of the whole

54
00:02:09,265 --> 00:02:11,396
was in some sense the sum or the average

55
00:02:11,396 --> 00:02:13,744
of that of its parts.

56
00:02:13,744 --> 00:02:16,321
And this idea of the value chain was predicated

57
00:02:16,321 --> 00:02:18,889
on the recognition that

58
00:02:18,889 --> 00:02:22,535
what holds a business together is transaction costs,

59
00:02:22,535 --> 00:02:24,830
that in essence you need to coordinate,

60
00:02:24,830 --> 00:02:27,405
organizations are more efficient at coordination

61
00:02:27,405 --> 00:02:29,067
than markets, very often,

62
00:02:29,067 --> 00:02:32,123
and therefore the nature and role and boundaries

63
00:02:32,123 --> 00:02:35,962
of the cooperation are defined by transaction costs.

64
00:02:35,962 --> 00:02:38,526
It was on those two ideas,

65
00:02:38,526 --> 00:02:41,912
Henderson's idea of increasing returns

66
00:02:41,912 --> 00:02:43,460
to scale and experience,

67
00:02:43,460 --> 00:02:45,883
and Porter's idea of the value chain,

68
00:02:45,883 --> 00:02:47,796
encompassing heterogenous elements,

69
00:02:47,796 --> 00:02:51,131
that the whole edifice of business strategy

70
00:02:51,131 --> 00:02:53,908
was subsequently erected.

71
00:02:53,908 --> 00:02:56,337
Now what I'm going to argue is

72
00:02:56,337 --> 00:03:02,051
that those premises are, in fact, being invalidated.

73
00:03:02,051 --> 00:03:04,605
First of all, let's think about transaction costs.

74
00:03:04,605 --> 00:03:07,002
There are really two components
to transaction costs.

75
00:03:07,002 --> 00:03:09,872
One is about processing information,
and the other is about communication.

76
00:03:09,872 --> 00:03:13,153
These are the economics of
processing and communicating

77
00:03:13,153 --> 00:03:15,627
as they have evolved over a long period of time.

78
00:03:15,627 --> 00:03:18,637
As we all know from so many contexts,

79
00:03:18,637 --> 00:03:20,893
they have been radically transformed

80
00:03:20,893 --> 00:03:23,205
since the days when Porter and Henderson

81
00:03:23,205 --> 00:03:25,299
first formulated their theories.

82
00:03:25,299 --> 00:03:27,365
In particular, since the mid-'90s,

83
00:03:27,365 --> 00:03:29,233
communications costs have actually been falling

84
00:03:29,233 --> 00:03:31,224
even faster than transaction costs,

85
00:03:31,224 --> 00:03:33,783
which is why communication, the Internet,

86
00:03:33,783 --> 00:03:38,398
has exploded in such a dramatic fashion.

87
00:03:38,398 --> 00:03:40,568
Now, those falling transaction costs

88
00:03:40,568 --> 00:03:42,544
have profound consequences,

89
00:03:42,544 --> 00:03:44,337
because if transaction costs are the glue

90
00:03:44,337 --> 00:03:46,759
that hold value chains together, and they are falling,

91
00:03:46,759 --> 00:03:48,401
there is less to economize on.

92
00:03:48,401 --> 00:03:51,333
There is less need for vertically
integrated organization,

93
00:03:51,333 --> 00:03:54,211
and value chains at least can break up.

94
00:03:54,211 --> 00:03:56,710
They needn't necessarily, but they can.

95
00:03:56,710 --> 00:03:58,850
In particular, it then becomes possible for

96
00:03:58,850 --> 00:04:00,616
a competitor in one business

97
00:04:00,616 --> 00:04:03,941
to use their position in one step of the value chain

98
00:04:03,941 --> 00:04:05,773
in order to penetrate or attack

99
00:04:05,773 --> 00:04:08,937
or disintermediate the competitor in another.

100
00:04:08,937 --> 00:04:11,598
That is not just an abstract proposition.

101
00:04:11,598 --> 00:04:13,277
There are many very specific stories

102
00:04:13,277 --> 00:04:15,049
of how that actually happened.

103
00:04:15,049 --> 00:04:18,542
A poster child example was
the encyclopedia business.

104
00:04:18,542 --> 00:04:19,991
The encyclopedia business

105
00:04:19,991 --> 00:04:22,093
in the days of leatherbound books

106
00:04:22,093 --> 00:04:23,797
was basically a distribution business.

107
00:04:23,797 --> 00:04:26,251
Most of the cost was the
commission to the salesmen.

108
00:04:26,251 --> 00:04:28,877
The CD-ROM and then the Internet came along,

109
00:04:28,877 --> 00:04:32,293
new technologies made the distribution of knowledge

110
00:04:32,293 --> 00:04:34,572
many orders of magnitude cheaper,

111
00:04:34,572 --> 00:04:37,286
and the encyclopedia industry collapsed.

112
00:04:37,286 --> 00:04:40,118
It's now, of course, a very familiar story.

113
00:04:40,118 --> 00:04:42,160
This, in fact, more generally was the story

114
00:04:42,160 --> 00:04:44,669
of the first generation of the Internet economy.

115
00:04:44,669 --> 00:04:46,869
It was about falling transaction costs

116
00:04:46,869 --> 00:04:48,344
breaking up value chains

117
00:04:48,344 --> 00:04:50,973
and therefore allowing disintermediation,

118
00:04:50,973 --> 00:04:53,310
or what we call deconstruction.

119
00:04:53,310 --> 00:04:55,627
One of the questions I was occasionally asked was,

120
00:04:55,627 --> 00:04:58,143
well, what's going to replace the encyclopedia

121
00:04:58,143 --> 00:05:00,794
when Britannica no longer has a business model?

122
00:05:00,794 --> 00:05:02,960
And it was a while before
the answer became manifest.

123
00:05:02,960 --> 00:05:05,747
Now, of course, we know
what it is: it's the Wikipedia.

124
00:05:05,747 --> 00:05:08,970
Now what's special about the
Wikipedia is not its distribution.

125
00:05:08,970 --> 00:05:11,374
What's special about the Wikipedia
is the way it's produced.

126
00:05:11,374 --> 00:05:13,614
The Wikipedia, of course, is an encyclopedia

127
00:05:13,614 --> 00:05:16,141
created by its users.

128
00:05:16,141 --> 00:05:18,021
And this, in fact, defines what you might call

129
00:05:18,021 --> 00:05:20,509
the second decade of the Internet economy,

130
00:05:20,509 --> 00:05:23,869
the decade in which the Internet as a noun

131
00:05:23,869 --> 00:05:25,776
became the Internet as a verb.

132
00:05:25,776 --> 00:05:27,645
It became a set of conversations,

133
00:05:27,645 --> 00:05:31,721
the era in which user-generated
content and social networks

134
00:05:31,721 --> 00:05:34,412
became the dominant phenomenon.

135
00:05:34,412 --> 00:05:36,310
Now what that really meant

136
00:05:36,310 --> 00:05:39,684
in terms of the Porter-Henderson framework

137
00:05:39,684 --> 00:05:43,075
was the collapse of certain
kinds of economies of scale.

138
00:05:43,075 --> 00:05:45,413
It turned out that tens of thousands

139
00:05:45,413 --> 00:05:48,496
of autonomous individuals writing an encyclopedia

140
00:05:48,496 --> 00:05:50,042
could do just as good a job,

141
00:05:50,042 --> 00:05:51,733
and certainly a much cheaper job,

142
00:05:51,733 --> 00:05:54,712
than professionals in a hierarchical organization.

143
00:05:54,712 --> 00:05:57,169
So basically what was happening was that one layer

144
00:05:57,169 --> 00:06:00,063
of this value chain was becoming fragmented,

145
00:06:00,063 --> 00:06:01,907
as individuals could take over

146
00:06:01,907 --> 00:06:05,194
where organizations were no longer needed.

147
00:06:05,194 --> 00:06:07,826
But there's another question
that obviously this graph poses,

148
00:06:07,826 --> 00:06:10,109
which is, okay, we've
gone through two decades --

149
00:06:10,109 --> 00:06:12,787
does anything distinguish the third?

150
00:06:12,787 --> 00:06:14,902
And what I'm going to argue is that indeed

151
00:06:14,902 --> 00:06:16,407
something does distinguish the third,

152
00:06:16,407 --> 00:06:18,853
and it maps exactly on to the kind of

153
00:06:18,853 --> 00:06:21,321
Porter-Henderson logic that
we've been talking about.

154
00:06:21,321 --> 00:06:23,908
And that is, about data.

155
00:06:23,908 --> 00:06:25,577
If we go back to around 2000,

156
00:06:25,577 --> 00:06:27,936
a lot of people were talking
about the information revolution,

157
00:06:27,936 --> 00:06:30,237
and it was indeed true that the world's stock of data

158
00:06:30,237 --> 00:06:32,511
was growing, indeed growing quite fast.

159
00:06:32,511 --> 00:06:35,369
but it was still at that point overwhelmingly analog.

160
00:06:35,369 --> 00:06:37,187
We go forward to 2007,

161
00:06:37,187 --> 00:06:40,372
not only had the world's stock of data exploded,

162
00:06:40,372 --> 00:06:42,892
but there'd been this massive substitution

163
00:06:42,892 --> 00:06:44,907
of digital for analog.

164
00:06:44,907 --> 00:06:46,776
And more important even than that,

165
00:06:46,776 --> 00:06:48,516
if you look more carefully at this graph,

166
00:06:48,516 --> 00:06:50,536
what you will observe is that about a half

167
00:06:50,536 --> 00:06:52,150
of that digital data

168
00:06:52,150 --> 00:06:54,536
is information that has an I.P. address.

169
00:06:54,536 --> 00:06:57,387
It's on a server or it's on a P.C.

170
00:06:57,387 --> 00:06:59,194
But having an I.P. address means that it

171
00:06:59,194 --> 00:07:01,364
can be connected to any other data

172
00:07:01,364 --> 00:07:03,227
that has an I.P. address.

173
00:07:03,227 --> 00:07:04,873
It means it becomes possible

174
00:07:04,873 --> 00:07:07,960
to put together half of the world's knowledge

175
00:07:07,960 --> 00:07:09,612
in order to see patterns,

176
00:07:09,612 --> 00:07:11,969
an entirely new thing.

177
00:07:11,969 --> 00:07:13,937
If we run the numbers forward to today,

178
00:07:13,937 --> 00:07:15,363
it probably looks something like this.

179
00:07:15,363 --> 00:07:16,531
We're not really sure.

180
00:07:16,531 --> 00:07:18,559
If we run the numbers forward to 2020,

181
00:07:18,559 --> 00:07:21,459
we of course have an exact number, courtesy of IDC.

182
00:07:21,459 --> 00:07:25,799
It's curious that the future is so much
more predictable than the present.

183
00:07:25,799 --> 00:07:30,194
And what it implies is a hundredfold multiplication

184
00:07:30,194 --> 00:07:33,167
in the stock of information that is connected

185
00:07:33,167 --> 00:07:35,395
via an I.P. address.

186
00:07:35,395 --> 00:07:38,856
Now, if the number of connections that we can make

187
00:07:38,856 --> 00:07:41,996
is proportional to the number of pairs of data points,

188
00:07:41,996 --> 00:07:44,459
a hundredfold multiplication in the quantity of data

189
00:07:44,459 --> 00:07:46,562
is a ten-thousandfold multiplication

190
00:07:46,562 --> 00:07:48,127
in the number of patterns

191
00:07:48,127 --> 00:07:50,204
that we can see in that data,

192
00:07:50,204 --> 00:07:52,864
this just in the last 10 or 11 years.

193
00:07:52,864 --> 00:07:55,688
This, I would submit, is a sea change,

194
00:07:55,688 --> 00:07:58,065
a profound change in the economics

195
00:07:58,065 --> 00:07:59,369
of the world that we live in.

196
00:07:59,893 --> 00:08:01,550
The first human genome,

197
00:08:01,550 --> 00:08:02,685
that of James Watson,

198
00:08:02,685 --> 00:08:06,774
was mapped as the culmination of the
Human Genome Project in the year 2000,

199
00:08:06,774 --> 00:08:08,966
and it took about 200 million dollars

200
00:08:08,966 --> 00:08:10,870
and about 10 years of work to map

201
00:08:10,870 --> 00:08:13,270
just one person's genomic makeup.

202
00:08:13,270 --> 00:08:15,982
Since then, the costs of mapping
the genome have come down.

203
00:08:15,982 --> 00:08:17,438
In fact, they've come down in recent years

204
00:08:17,438 --> 00:08:19,374
very dramatically indeed,

205
00:08:19,374 --> 00:08:21,774
to the point where the cost 
is now below 1,000 dollars,

206
00:08:21,774 --> 00:08:25,014
and it's confidently predicted that by the year 2015

207
00:08:25,014 --> 00:08:26,534
it will be below 100 dollars --

208
00:08:26,534 --> 00:08:29,438
a five or six order of magnitude drop

209
00:08:29,438 --> 00:08:31,582
in the cost of genomic mapping

210
00:08:31,582 --> 00:08:33,974
in just a 15-year period,

211
00:08:33,974 --> 00:08:36,630
an extraordinary phenomenon.

212
00:08:36,630 --> 00:08:40,604
Now, in the days when mapping a genome

213
00:08:40,604 --> 00:08:43,752
cost millions, or even tens of thousands,

214
00:08:43,752 --> 00:08:45,776
it was basically a research enterprise.

215
00:08:45,776 --> 00:08:48,288
Scientists would gather some representative people,

216
00:08:48,288 --> 00:08:49,609
and they would see patterns, and they would try

217
00:08:49,609 --> 00:08:52,208
and make generalizations about
human nature and disease

218
00:08:52,208 --> 00:08:54,152
from the abstract patterns they find

219
00:08:54,152 --> 00:08:57,535
from these particular selected individuals.

220
00:08:57,535 --> 00:09:00,368
But when the genome can 
be mapped for 100 bucks,

221
00:09:00,368 --> 00:09:02,224
99 dollars while you wait,

222
00:09:02,224 --> 00:09:04,783
then what happens is, it becomes retail.

223
00:09:04,783 --> 00:09:06,824
It becomes above all clinical.

224
00:09:06,824 --> 00:09:08,079
You go the doctor with a cold,

225
00:09:08,079 --> 00:09:09,832
and if he or she hasn't done it already,

226
00:09:09,832 --> 00:09:12,064
the first thing they do is map your genome,

227
00:09:12,064 --> 00:09:13,463
at which point what they're now doing

228
00:09:13,463 --> 00:09:18,361
is not starting from some abstract 
knowledge of genomic medicine

229
00:09:18,361 --> 00:09:20,225
and trying to work out how it applies to you,

230
00:09:20,225 --> 00:09:22,780
but they're starting from your particular genome.

231
00:09:22,780 --> 00:09:24,388
Now think of the power of that.

232
00:09:24,388 --> 00:09:26,100
Think of where that takes us

233
00:09:26,100 --> 00:09:28,892
when we can combine genomic data

234
00:09:28,892 --> 00:09:30,644
with clinical data

235
00:09:30,644 --> 00:09:32,556
with data about drug interactions

236
00:09:32,556 --> 00:09:34,676
with the kind of ambient data that devices

237
00:09:34,676 --> 00:09:36,604
like our phone and medical sensors

238
00:09:36,604 --> 00:09:38,508
will increasingly be collecting.

239
00:09:38,508 --> 00:09:40,892
Think what happens when we collect all of that data

240
00:09:40,892 --> 00:09:42,180
and we can put it together

241
00:09:42,180 --> 00:09:44,900
in order to find patterns we wouldn't see before.

242
00:09:44,900 --> 00:09:47,764
This, I would suggest, perhaps it will take a while,

243
00:09:47,764 --> 00:09:50,236
but this will drive a revolution in medicine.

244
00:09:50,236 --> 00:09:52,540
Fabulous, lots of people talk about this.

245
00:09:52,540 --> 00:09:54,964
But there's one thing that
doesn't get much attention.

246
00:09:54,964 --> 00:09:58,340
How is that model of colossal sharing

247
00:09:58,340 --> 00:10:00,940
across all of those kinds of databases

248
00:10:00,940 --> 00:10:03,340
compatible with the business models

249
00:10:03,340 --> 00:10:06,108
of institutions and organizations and corporations

250
00:10:06,108 --> 00:10:08,236
that are involved in this business today?

251
00:10:08,236 --> 00:10:10,996
If your business is based on proprietary data,

252
00:10:10,996 --> 00:10:14,084
if your competitive advantage
is defined by your data,

253
00:10:14,084 --> 00:10:17,791
how on Earth is that company or is that society

254
00:10:17,791 --> 00:10:19,980
in fact going to achieve the value

255
00:10:19,980 --> 00:10:22,724
that's implicit in the technology? They can't.

256
00:10:22,724 --> 00:10:24,740
So essentially what's happening here,

257
00:10:24,740 --> 00:10:27,492
and genomics is merely one example of this,

258
00:10:27,492 --> 00:10:29,668
is that technology is driving

259
00:10:29,668 --> 00:10:32,252
the natural scaling of the activity

260
00:10:32,252 --> 00:10:35,332
beyond the institutional boundaries within which

261
00:10:35,332 --> 00:10:37,460
we have been used to thinking about it,

262
00:10:37,460 --> 00:10:39,684
and in particular beyond the institutional boundaries

263
00:10:39,684 --> 00:10:41,924
in terms of which business strategy

264
00:10:41,924 --> 00:10:45,348
as a discipline is formulated.

265
00:10:45,348 --> 00:10:48,356
The basic story here is that what used to be

266
00:10:48,356 --> 00:10:52,557
vertically integrated, oligopolistic competition

267
00:10:52,557 --> 00:10:55,317
among essentially similar kinds of competitors

268
00:10:55,317 --> 00:10:57,653
is evolving, by one means or another,

269
00:10:57,653 --> 00:11:01,229
from a vertical structure to a horizontal one.

270
00:11:01,229 --> 00:11:02,821
Why is that happening?

271
00:11:02,821 --> 00:11:05,141
It's happening because
transaction costs are plummeting

272
00:11:05,141 --> 00:11:07,269
and because scale is polarizing.

273
00:11:07,269 --> 00:11:08,981
The plummeting of transaction costs

274
00:11:08,981 --> 00:11:11,557
weakens the glue that holds value chains together,

275
00:11:11,557 --> 00:11:13,333
and allows them to separate.

276
00:11:13,333 --> 00:11:15,237
The polarization of scale economies

277
00:11:15,237 --> 00:11:18,301
towards the very small -- small is beautiful --

278
00:11:18,301 --> 00:11:20,893
allows for scalable communities

279
00:11:20,893 --> 00:11:24,133
to substitute for conventional corporate production.

280
00:11:24,133 --> 00:11:26,245
The scaling in the opposite direction,

281
00:11:26,245 --> 00:11:28,037
towards things like big data,

282
00:11:28,037 --> 00:11:29,609
drive the structure of business

283
00:11:29,609 --> 00:11:32,502
towards the creation of new kinds of institutions

284
00:11:32,502 --> 00:11:34,341
that can achieve that scale.

285
00:11:34,341 --> 00:11:37,093
But either way, the typically vertical structure

286
00:11:37,093 --> 00:11:40,165
gets driven to becoming more horizontal.

287
00:11:40,165 --> 00:11:42,657
The logic isn't just about big data.

288
00:11:42,657 --> 00:11:45,814
If we were to look, for example,
at the telecommunications industry,

289
00:11:45,814 --> 00:11:47,958
you can tell the same story about fiber optics.

290
00:11:47,958 --> 00:11:50,309
If we look at the pharmaceutical industry,

291
00:11:50,309 --> 00:11:52,125
or, for that matter, university research,

292
00:11:52,125 --> 00:11:53,845
you can say exactly the same story

293
00:11:53,845 --> 00:11:55,861
about so-called "big science."

294
00:11:55,861 --> 00:11:57,181
And in the opposite direction,

295
00:11:57,181 --> 00:11:59,501
if we look, say, at the energy sector,

296
00:11:59,501 --> 00:12:02,005
where all the talk is about how households

297
00:12:02,005 --> 00:12:05,853
will be efficient producers of green energy

298
00:12:05,853 --> 00:12:08,327
and efficient conservers of energy,

299
00:12:08,327 --> 00:12:10,253
that is, in fact, the reverse phenomenon.

300
00:12:10,253 --> 00:12:12,277
That is the fragmentation of scale

301
00:12:12,277 --> 00:12:14,261
because the very small can substitute

302
00:12:14,261 --> 00:12:17,117
for the traditional corporate scale.

303
00:12:17,117 --> 00:12:18,877
Either way, what we are driven to

304
00:12:18,877 --> 00:12:22,417
is this horizontalization of the structure of industries,

305
00:12:22,417 --> 00:12:24,777
and that implies fundamental changes

306
00:12:24,777 --> 00:12:26,962
in how we think about strategy.

307
00:12:26,962 --> 00:12:29,194
It means, for example, that we need to think

308
00:12:29,194 --> 00:12:31,633
about strategy as the curation

309
00:12:31,633 --> 00:12:34,017
of these kinds of horizontal structure,

310
00:12:34,017 --> 00:12:35,810
where things like business definition

311
00:12:35,810 --> 00:12:37,465
and even industry definition

312
00:12:37,465 --> 00:12:39,897
are actually the outcomes of strategy,

313
00:12:39,897 --> 00:12:43,232
not something that the strategy presupposes.

314
00:12:43,232 --> 00:12:46,496
It means, for example, we need to work out

315
00:12:46,496 --> 00:12:48,776
how to accommodate collaboration

316
00:12:48,776 --> 00:12:50,888
and competition simultaneously.

317
00:12:50,888 --> 00:12:51,867
Think about the genome.

318
00:12:51,867 --> 00:12:53,691
We need to accommodate the very large

319
00:12:53,691 --> 00:12:55,810
and the very small simultaneously.

320
00:12:55,810 --> 00:12:57,723
And we need industry structures

321
00:12:57,723 --> 00:13:00,603
that will accommodate very,
very different motivations,

322
00:13:00,603 --> 00:13:03,283
from the amateur motivations
of people in communities

323
00:13:03,283 --> 00:13:04,947
to maybe the social motivations

324
00:13:04,947 --> 00:13:07,283
of infrastructure built by governments,

325
00:13:07,283 --> 00:13:10,108
or, for that matter, cooperative institutions

326
00:13:10,108 --> 00:13:12,539
built by companies that are otherwise competing,

327
00:13:12,539 --> 00:13:15,579
because that is the only way
that they can get to scale.

328
00:13:15,579 --> 00:13:17,571
These kinds of transformations

329
00:13:17,571 --> 00:13:21,527
render the traditional premises 
of business strategy obsolete.

330
00:13:21,527 --> 00:13:23,912
They drive us into a completely new world.

331
00:13:23,912 --> 00:13:25,456
They require us, whether we are

332
00:13:25,456 --> 00:13:27,992
in the public sector or the private sector,

333
00:13:27,992 --> 00:13:30,400
to think very fundamentally differently

334
00:13:30,400 --> 00:13:32,297
about the structure of business,

335
00:13:32,297 --> 00:13:35,736
and, at last, it makes strategy interesting again.

336
00:13:35,736 --> 00:13:38,471
Thank you.

337
00:13:38,471 --> 00:13:40,755
(Applause)