1
00:00:00,600 --> 00:00:03,650
I'm Miko Pawlikowski and
this is Hockey Stick.

2
00:00:04,650 --> 00:00:09,119
Today we're talking about the role of
the CTO, the Chief Technology Officer.

3
00:00:09,749 --> 00:00:10,429
Who is that?

4
00:00:10,839 --> 00:00:11,829
What's their job like?

5
00:00:12,359 --> 00:00:16,839
And why is it so hard to survive
the first 100 days in the role?

6
00:00:17,850 --> 00:00:21,939
To answer these questions, I brought
in an expert on the subject, Alan

7
00:00:21,990 --> 00:00:27,639
Williamson, the author of the book
"Think Like a CTO" . Alan is the partner

8
00:00:27,639 --> 00:00:32,489
of the Portfolio Operations Group at
New Harbor Capital, and through that

9
00:00:32,549 --> 00:00:37,725
private equity experience he's seen
mentored and worked with countless CTOs.

10
00:00:38,175 --> 00:00:42,185
If you think of becoming a CTO,
Alan has plenty of advice for you.

11
00:00:43,114 --> 00:00:45,894
Welcome to this episode and please enjoy.

12
00:00:46,726 --> 00:00:49,176
Miko, it's an absolute pleasure
and honor to be here, sir.

13
00:00:49,176 --> 00:00:50,496
Thank you very much for inviting me.

14
00:00:50,706 --> 00:00:51,376
Absolutely.

15
00:00:51,956 --> 00:00:55,166
Alan, would you like to tell us a
little bit about yourself and how

16
00:00:55,166 --> 00:00:56,626
you ended up writing this book?

17
00:00:57,876 --> 00:00:58,416
Yes.

18
00:00:58,446 --> 00:01:03,446
so basically I was thrust into
a Chief Technology Officer role.

19
00:01:03,476 --> 00:01:07,866
I looked around and discovered there
was nothing to help me figure out

20
00:01:07,876 --> 00:01:09,396
what the hell this role was all about.

21
00:01:09,916 --> 00:01:14,606
so from that perspective, I've always
kept detailed notes and journals, et

22
00:01:14,606 --> 00:01:16,816
cetera, as I go throughout my life.

23
00:01:17,226 --> 00:01:19,876
and I thought, I'm going to
start documenting this journey.

24
00:01:20,211 --> 00:01:26,651
and after 10 years, as it were, I had
enough material and I had enough stories,

25
00:01:26,651 --> 00:01:30,791
war stories, successes, failures,
to be able to bring that together.

26
00:01:31,631 --> 00:01:35,911
And through my work with various private
equity, I got the opportunity to meet

27
00:01:36,381 --> 00:01:41,911
many other people ascending the same
ladder that I was ascending and both

28
00:01:41,921 --> 00:01:43,911
helping them, learning from them,

29
00:01:43,921 --> 00:01:49,111
So in many respects, I cover a
very large spectrum of subjects.

30
00:01:49,501 --> 00:01:53,581
And frankly, your typical CTO
will not need all of the chapters.

31
00:01:54,061 --> 00:01:57,121
They'll need maybe 70% of the
chapters, depending on the

32
00:01:57,121 --> 00:01:58,721
company that they're with.

33
00:01:59,141 --> 00:02:05,501
So the book is purely there
to help the upcoming CTO to

34
00:02:05,501 --> 00:02:07,201
figure out what is the role?

35
00:02:08,081 --> 00:02:10,411
What do you need to do
to succeed in the role?

36
00:02:11,291 --> 00:02:12,041
And the sort of.

37
00:02:12,731 --> 00:02:17,441
range of subjects that you're expected
to deal with when you are a CTO.

38
00:02:18,011 --> 00:02:22,171
Most people assume a CTO is just
basically a senior developer or a

39
00:02:22,171 --> 00:02:26,261
senior engineer, but in many respects
it's the complete opposite of that.

40
00:02:26,301 --> 00:02:29,671
And that's the sort of culture
shock that a lot of engineers

41
00:02:29,671 --> 00:02:33,161
face when they're looking up to
see, 'is this something I can do?'

42
00:02:33,881 --> 00:02:37,451
to be very honest with you, when
I first saw it, I thought, that

43
00:02:37,451 --> 00:02:38,761
sounds a little bit pretentious.

44
00:02:38,791 --> 00:02:42,451
But then, as I was reading that,
I realized that, pretty much

45
00:02:42,491 --> 00:02:46,451
everybody who works in tech,
they obviously know who CTO is.

46
00:02:46,611 --> 00:02:48,211
And then you ask them questions.

47
00:02:48,721 --> 00:02:52,271
"So what's the scope exactly,
and what should they be doing?"

48
00:02:52,341 --> 00:02:57,551
And it gets a little bit more fuzzy and a
little bit more vague than it should be.

49
00:02:57,551 --> 00:03:01,771
So it's great that you just went
ahead and wrote a book about that.

50
00:03:02,781 --> 00:03:09,121
the big shock that most people realize
is how little technology most of

51
00:03:09,121 --> 00:03:10,701
your day really is involved with.

52
00:03:11,531 --> 00:03:14,481
what you're usually doing, particularly
of companies of certain sizes.

53
00:03:14,941 --> 00:03:19,611
You're trying to help the company fully
utilize technology in such a way that the

54
00:03:19,611 --> 00:03:26,291
company does not get hindered or slowed
down through its use of technology.

55
00:03:27,061 --> 00:03:32,271
So if we take a step back, for
everybody who is already hooked, just

56
00:03:32,271 --> 00:03:35,421
go to manning.com and get the book.

57
00:03:35,841 --> 00:03:37,171
It's called "Think like a CTO".

58
00:03:37,181 --> 00:03:37,831
Once again.

59
00:03:38,526 --> 00:03:41,926
But for anybody else who needs
a little bit more convincing,

60
00:03:42,676 --> 00:03:44,796
What does CTO actually do?

61
00:03:46,231 --> 00:03:51,741
The chief technology officer, as per
the sort of phrase, sits at the c-level.

62
00:03:52,281 --> 00:03:56,781
And from that perspective, they're
there to help the CEO and the rest of

63
00:03:56,791 --> 00:04:02,491
the c-levels determine what technology
can do for the company and what

64
00:04:02,631 --> 00:04:04,971
opportunities can that technology open up?

65
00:04:05,501 --> 00:04:12,401
What friction can that technology remove
and how best utilize the core assets

66
00:04:12,401 --> 00:04:14,061
of the company, which is usually data.

67
00:04:14,976 --> 00:04:17,366
in order to provide more
value to the client.

68
00:04:18,216 --> 00:04:24,226
one of their biggest goals is to lay
out of roadmap or a vision statement,

69
00:04:24,226 --> 00:04:28,316
and I go into that in extreme detail
in one of my chapters, to figure

70
00:04:28,316 --> 00:04:32,186
out where are we going as a company
from a technology point of view.

71
00:04:33,226 --> 00:04:34,946
What platforms am I investing in?

72
00:04:35,596 --> 00:04:37,566
what features am I looking to develop?

73
00:04:37,946 --> 00:04:39,356
there's lots of different types of CTOs.

74
00:04:39,396 --> 00:04:44,796
But, you usually have a company that's
either producing technology, and

75
00:04:44,796 --> 00:04:50,221
therefore the CTO's got a much more
hands-on architectural engineering role.

76
00:04:50,671 --> 00:04:54,531
And then you have CTOs that are more
consuming technology and they're more

77
00:04:54,571 --> 00:04:59,461
aggregating technology from many different
vendors, many different sources, but

78
00:04:59,461 --> 00:05:02,761
still providing an overall roadmap
as to where we're going as a company.

79
00:05:04,096 --> 00:05:09,366
I believe in your book, You gave the
analogy of chess, as I was reading it

80
00:05:09,396 --> 00:05:16,056
for the first time, I completely misread
it because they all CEO, CFO, CTO, and

81
00:05:16,056 --> 00:05:20,086
I just assumed that the CTO would be the
queen, but that's not the case, is it?

82
00:05:20,466 --> 00:05:24,156
I believe that the CFO
is the queen, right?

83
00:05:24,156 --> 00:05:26,636
So in the game of chess, the queen
is the most powerful piece on the

84
00:05:26,636 --> 00:05:30,936
board, because that piece can move
anywhere, any direction, and any

85
00:05:30,936 --> 00:05:32,986
number of squares at the same time.

86
00:05:34,276 --> 00:05:36,796
A king can only move one piece at a time.

87
00:05:37,936 --> 00:05:42,506
in business parlance, the CEO
sets the big direction for the

88
00:05:42,506 --> 00:05:47,106
company, the big decisions of
where the company is going to go.

89
00:05:47,406 --> 00:05:51,101
And in many respects, They
shouldn't be able to move quickly.

90
00:05:51,501 --> 00:05:55,181
They shouldn't be able to go to the
other side of the board quickly,

91
00:05:55,201 --> 00:05:56,901
leaving everybody else behind.

92
00:05:57,331 --> 00:05:59,751
Because part of the CEO is to
bring everybody along with them.

93
00:06:00,426 --> 00:06:04,066
So therefore they've got to be far
more strategic, far more calculated

94
00:06:04,206 --> 00:06:05,606
in their decision making process.

95
00:06:06,926 --> 00:06:11,296
A queen, however, or the
CFO, they basically know the

96
00:06:11,296 --> 00:06:12,426
logistics of the company.

97
00:06:12,951 --> 00:06:14,581
They basically finance everything.

98
00:06:15,091 --> 00:06:18,611
They know how much money we've got,
how much money we need, where is it

99
00:06:18,611 --> 00:06:20,601
being spent, what needs to be done.

100
00:06:20,781 --> 00:06:26,291
So therefore, nobody makes a decision
without the CFO truly blessing it,

101
00:06:26,331 --> 00:06:29,271
because if we can't fund it, we
can have all the dreams we want,

102
00:06:29,281 --> 00:06:30,561
but they'll just remain dreams.

103
00:06:31,801 --> 00:06:35,451
So from that perspective, the two of them
work very closely together, particularly

104
00:06:35,451 --> 00:06:39,011
in a private equity environment
where it's a much smaller company.

105
00:06:39,011 --> 00:06:43,416
When I say a much smaller company, it's
maybe 200M revenue and less, which is

106
00:06:43,416 --> 00:06:45,111
where the majority of us all work in.

107
00:06:45,571 --> 00:06:50,451
so the CFO and the CEO work very
hand in hand with each other.

108
00:06:50,841 --> 00:06:54,651
One sort of holding back the other
one, egging on the other, et cetera,

109
00:06:54,651 --> 00:06:58,251
depending on the roles and depending
on what's happening now, the CTO,

110
00:06:58,901 --> 00:07:00,311
I always liken that to the rook.

111
00:07:01,391 --> 00:07:06,146
We are the ones providing the backbone
of the, the overall infrastructure.

112
00:07:06,566 --> 00:07:08,966
We are there to provide stability.

113
00:07:09,536 --> 00:07:13,256
We're there to effectively
keep out the bad people.

114
00:07:14,776 --> 00:07:19,476
Keep the business running and make
sure that we're dependable and we're

115
00:07:19,476 --> 00:07:22,326
not making any big sweeping changes.

116
00:07:23,259 --> 00:07:25,329
my mind went a different path.

117
00:07:25,409 --> 00:07:29,849
I was immediately picturing
the rook being the CFO.

118
00:07:30,344 --> 00:07:36,034
Guarding, the purse and making sure
that the money spent the right way and

119
00:07:36,034 --> 00:07:42,314
making sure that we prioritize income
to the extent that we need to survive

120
00:07:42,414 --> 00:07:49,714
And I guess this is why mentally already
Had that image when I was reading that

121
00:07:49,799 --> 00:07:54,819
It is a fun game to play when you're maybe
sitting at a fireplace, drinking some beer

122
00:07:54,819 --> 00:07:58,879
or some whiskey to debate it, the other
one that's been interesting is the Bishop.

123
00:07:59,859 --> 00:08:01,059
Is the Bishop the sales team?

124
00:08:01,059 --> 00:08:03,899
Because they've got the line of sight as
to what's coming and what's coming on.

125
00:08:04,199 --> 00:08:08,609
it's a fun game to play, but I think
that, between the King, the Queen and the

126
00:08:08,609 --> 00:08:11,879
Rook, it covers the three big C-levels.

127
00:08:13,331 --> 00:08:15,041
What about the different sizes?

128
00:08:15,121 --> 00:08:18,631
Another thing that I was immediately
scanning your book for was like,

129
00:08:18,661 --> 00:08:22,171
'Oh, okay, obviously it's going
to be very different when you have

130
00:08:22,181 --> 00:08:26,261
three people on the team than do
when you have 300K people on that'.

131
00:08:26,961 --> 00:08:30,851
How does that affect the CTO position?

132
00:08:32,231 --> 00:08:36,301
That is a great point to make
out, and it's something I go

133
00:08:36,301 --> 00:08:38,001
into with a lot of detail.

134
00:08:38,351 --> 00:08:42,386
in your typical start up, scenario
where there is maybe only three of you.

135
00:08:43,226 --> 00:08:45,266
The CTO isn't really the CTO.

136
00:08:45,266 --> 00:08:47,926
You're the chief engineer,
to be honest with you.

137
00:08:47,956 --> 00:08:51,076
And probably you're getting that title
because we can't afford to pay you.

138
00:08:51,716 --> 00:08:53,576
so it's going to look really
good in your business card.

139
00:08:53,966 --> 00:08:57,806
but you're not going to stand
side by side with, the CTO of AWS.

140
00:08:57,886 --> 00:08:59,646
You're just in a different league.

141
00:09:00,156 --> 00:09:03,416
So from that perspective, you're
not really interested in the

142
00:09:03,416 --> 00:09:04,996
long term future of the company.

143
00:09:05,896 --> 00:09:07,366
You're more interested in just survival.

144
00:09:08,041 --> 00:09:09,811
Can we build a proof of concept?

145
00:09:09,901 --> 00:09:11,101
Can we build a product?

146
00:09:11,111 --> 00:09:15,351
Can we build something that proves
that this company has indeed got legs?

147
00:09:16,001 --> 00:09:19,001
And from that perspective,
you're far more hands-on.

148
00:09:19,411 --> 00:09:23,311
You are literally coding every
line of code more than likely.

149
00:09:23,651 --> 00:09:27,451
You know exactly how to deploy
the website, how to do the

150
00:09:27,451 --> 00:09:29,601
database, how to do the email.

151
00:09:29,711 --> 00:09:31,851
You are pretty much IT, CTO.

152
00:09:31,881 --> 00:09:35,931
Everything that's technology-related
comes under your belt.

153
00:09:35,941 --> 00:09:37,531
So you're pretty much
a jack of all trades.

154
00:09:38,451 --> 00:09:41,531
And that's a very good place
to start, because you do get a

155
00:09:41,541 --> 00:09:42,771
phenomenal amount of exposure.

156
00:09:43,251 --> 00:09:48,886
Now, As part of that sort of scrappy
nature of bootstrapping a company to

157
00:09:48,886 --> 00:09:51,546
start up, you will be taking shortcuts.

158
00:09:52,296 --> 00:09:54,826
You will not be doing everything
that you should be doing.

159
00:09:54,846 --> 00:09:57,116
Because why do you need to?

160
00:09:57,596 --> 00:10:00,156
You're the only person between
you and maybe two or three other

161
00:10:00,156 --> 00:10:01,566
people that know everything.

162
00:10:01,826 --> 00:10:03,346
So you don't need to do documentation.

163
00:10:04,056 --> 00:10:06,906
You don't need necessarily do
backups because, 'hey, it's either

164
00:10:06,906 --> 00:10:08,226
on the server or it's on my laptop'.

165
00:10:08,236 --> 00:10:09,486
One of the two of us will have it.

166
00:10:10,066 --> 00:10:13,306
security, probably not as
high on your priority list.

167
00:10:13,486 --> 00:10:15,846
Yes, you're going to keep your
client data secure, but there's

168
00:10:15,846 --> 00:10:17,056
only four of you in the company.

169
00:10:17,056 --> 00:10:18,536
yeah, I don't need to keep audit logs.

170
00:10:19,216 --> 00:10:21,426
there are certain things that
you'll take shortcuts on.

171
00:10:22,356 --> 00:10:26,826
I'll use my own personal credit card to
sign up the GitHub or the AWS account.

172
00:10:27,546 --> 00:10:31,926
The, domain that I registered with
GoDaddy or Network Solutions is

173
00:10:32,116 --> 00:10:33,996
probably on the CEO's credit card.

174
00:10:35,221 --> 00:10:38,091
So it's a lot more scrappy
on that perspective.

175
00:10:38,591 --> 00:10:40,131
However, you are the go to person.

176
00:10:40,431 --> 00:10:43,491
You probably are the only person
that can restart something or push

177
00:10:43,491 --> 00:10:45,361
something out or get something done.

178
00:10:45,411 --> 00:10:50,031
from a CTO perspective at that point,
you are, in the literal sense, the

179
00:10:50,041 --> 00:10:52,981
chief technology person in that company.

180
00:10:53,119 --> 00:10:58,360
And as the company grows, it's going
to become more structured, right?

181
00:10:59,524 --> 00:11:05,054
Yes, And this could be a hard transition
for startup, CTOs here is getting rid

182
00:11:05,054 --> 00:11:09,934
of the sort of the hero complex, which
is everything's got to go through me.

183
00:11:10,904 --> 00:11:15,534
what you generally find is your
sort of litmus test to see if you

184
00:11:15,534 --> 00:11:19,064
do indeed have a hero complex,
go on vacation for two weeks.

185
00:11:20,674 --> 00:11:22,474
How many releases were
done in your absence?

186
00:11:23,084 --> 00:11:25,094
How many server restarts
were done in your absence?

187
00:11:25,654 --> 00:11:27,484
How many new features were
built in your absence?

188
00:11:27,484 --> 00:11:31,494
If you come back and zero was done,
then pretty much the company stood

189
00:11:31,494 --> 00:11:32,754
still while you were on vacation.

190
00:11:34,494 --> 00:11:38,514
Or, did you have to log in and do
everything that you needed to do?

191
00:11:39,184 --> 00:11:42,754
A company is never going to scale
if everything is going to be

192
00:11:42,754 --> 00:11:44,054
bottlenecked through one person.

193
00:11:45,134 --> 00:11:53,379
So therefore, the next evolution is trying
to bring in other people and again, I talk

194
00:11:53,379 --> 00:11:54,849
about this in the recruitment chapter.

195
00:11:55,929 --> 00:12:00,959
It is very easy to always be
hiring people not as smart as you.

196
00:12:02,079 --> 00:12:05,199
Every book tells you always
hire somebody smarter than you.

197
00:12:05,389 --> 00:12:09,619
every management meeting, every
management lecture, always

198
00:12:09,619 --> 00:12:10,859
hire smarter people than you.

199
00:12:10,919 --> 00:12:12,519
You should never be the
smartest person in the room.

200
00:12:13,619 --> 00:12:16,309
That is wonderful car
bumper sticker advice.

201
00:12:17,719 --> 00:12:20,879
But it's really hard to implement
because the human nature,

202
00:12:20,879 --> 00:12:23,709
particularly technologists,
because we're very ego-driven.

203
00:12:24,199 --> 00:12:25,749
we love our voices to be heard.

204
00:12:25,749 --> 00:12:28,339
We love to be showing off
how great our code is.

205
00:12:28,339 --> 00:12:34,309
It takes a very strong personality to
be able to build a team around them

206
00:12:34,739 --> 00:12:36,359
where they aren't the smartest person.

207
00:12:37,459 --> 00:12:41,679
But that's going to serve the
company and you phenomenally well

208
00:12:41,749 --> 00:12:42,789
as you lay down the groundwork

209
00:12:43,999 --> 00:12:46,399
and that's true of any
level of any company.

210
00:12:47,079 --> 00:12:51,619
it does take, discipline to figure
out where your weaknesses are.

211
00:12:52,109 --> 00:12:58,059
Where can I shore that up by bringing
in outside talent and where can I

212
00:12:58,069 --> 00:13:03,549
remove myself from the daily grind
to grind of running this company?

213
00:13:04,149 --> 00:13:08,789
Because as a CTO, you should not be
involved in every single release.

214
00:13:09,639 --> 00:13:11,499
You should not be involved
with every commit.

215
00:13:12,299 --> 00:13:13,599
Every pull request.

216
00:13:14,219 --> 00:13:19,379
You have to take yourself out of the
day-to-day running of the company.

217
00:13:19,939 --> 00:13:24,679
And the sort of evolution is at some
point you'll get to a stage where you'll

218
00:13:24,679 --> 00:13:26,649
probably bring in a VP of Engineering.

219
00:13:27,899 --> 00:13:31,609
And that is that person that is
involved more in the day-to-day grind

220
00:13:31,984 --> 00:13:36,654
of what's going on to allow you to
focus on the greater strategy, but

221
00:13:36,654 --> 00:13:40,824
also to start communicating more with
outside of the engineering group.

222
00:13:42,044 --> 00:13:48,964
And one of the sort of, again, red
flags is in your day-to-day work,

223
00:13:49,674 --> 00:13:52,794
how often are you speaking with your
engineering team versus how often

224
00:13:52,794 --> 00:13:53,834
you're speaking with the business?

225
00:13:53,834 --> 00:13:57,904
If it's 90% engineering team and 10%
of the business, you're operating

226
00:13:57,904 --> 00:13:59,024
more as a VP of engineering.

227
00:13:59,024 --> 00:14:00,564
You're not operating as a CTO.

228
00:14:01,664 --> 00:14:08,864
It should be 60-70% talking to
the business and 40% talking

229
00:14:08,864 --> 00:14:09,934
with your engineering group.

230
00:14:10,904 --> 00:14:12,974
So it depends on the
evolution of the company.

231
00:14:12,974 --> 00:14:18,134
But as we grow, as we evolve, our
wants, our desires, our skillsets have

232
00:14:18,154 --> 00:14:22,054
to evolve to adapt to the environment
that we are now being placed in.

233
00:14:22,754 --> 00:14:23,614
And sometimes.

234
00:14:24,239 --> 00:14:28,039
You get to evolve with the company if
the startup's really going well, etc.

235
00:14:28,899 --> 00:14:34,249
But usually you have to leave a
company to jump in at another level in

236
00:14:34,249 --> 00:14:35,939
order for you to have career growth.

237
00:14:35,939 --> 00:14:38,629
Problem being, for example,
is if you're in a five man

238
00:14:39,209 --> 00:14:41,859
company, there's only one CTO.

239
00:14:43,249 --> 00:14:45,539
And the four of the people
underneath you isn't going to get

240
00:14:45,539 --> 00:14:48,909
to that role unless you advocate.

241
00:14:49,399 --> 00:14:52,739
So therefore they will probably
have to leave that company in

242
00:14:52,739 --> 00:14:53,969
order for their own career growth.

243
00:14:54,069 --> 00:14:56,419
that's some of the problems
with the smaller company.

244
00:14:57,154 --> 00:15:00,918
Did you find, through your experience
with private equity, you mentioned,

245
00:15:01,136 --> 00:15:06,145
did you find that founders who were
there from the beginning, did they make

246
00:15:06,263 --> 00:15:09,891
good CTOs often, sometimes, rarely?

247
00:15:10,200 --> 00:15:14,018
You mentioned the jumping out of the
company, but for those who stay, we

248
00:15:14,018 --> 00:15:15,927
obviously have some famous examples.

249
00:15:15,936 --> 00:15:19,209
everybody's probably thinking about
Zuckerberg or people like that, but

250
00:15:19,709 --> 00:15:22,345
for the average company, let's say.

251
00:15:23,013 --> 00:15:24,850
what would you say is the answer to that?

252
00:15:26,595 --> 00:15:30,395
9 times out of 10 it's usually a poor
decision that they've made with a CTO.

253
00:15:31,245 --> 00:15:36,725
it's not uncommon for, companies doing
phenomenally well, but the person

254
00:15:36,725 --> 00:15:40,385
that's acting as CTO has been the
person that's been there from the

255
00:15:40,385 --> 00:15:49,005
start and has been somebody that maybe
the CEO or the founder knew he lifted

256
00:15:49,025 --> 00:15:51,825
up the street or he knew a friend of
a friend, because he couldn't really

257
00:15:51,825 --> 00:15:55,280
afford an engineer or CTO at that level.

258
00:15:55,290 --> 00:16:00,130
So basically pair of them have learned
together and that's usually no problems.

259
00:16:00,270 --> 00:16:02,020
They have successfully built a business.

260
00:16:03,110 --> 00:16:07,790
However, experience has shown is
that they're usually very narrow.

261
00:16:08,530 --> 00:16:14,680
They haven't accepted new ideas or new
thoughts coming in from the outside.

262
00:16:15,000 --> 00:16:17,850
They have rarely hired
somebody smarter than them.

263
00:16:18,445 --> 00:16:22,985
Because they want to protect the
domain and they've usually reached

264
00:16:22,995 --> 00:16:30,215
the ceiling when it comes to like
scalability or, security or management,

265
00:16:30,255 --> 00:16:32,665
documentation, all of that usual stuff.

266
00:16:33,295 --> 00:16:38,335
And we often see that manifest itself
in terms of, there is no documentation

267
00:16:38,365 --> 00:16:40,325
because they know everything.

268
00:16:41,870 --> 00:16:45,620
Or they've put everything onto the cloud
and they think they're cloud-enabled but

269
00:16:45,640 --> 00:16:48,990
in actual fact when you look at their AWS
environment or their Azure environment,

270
00:16:49,110 --> 00:16:51,000
they're using it like a data center.

271
00:16:51,105 --> 00:16:53,035
you've just evolved but you
haven't thought about it.

272
00:16:53,325 --> 00:16:57,895
And again that shows where they
haven't kept up and they haven't

273
00:16:57,895 --> 00:17:02,205
spent time teaching themselves
about the new technologies.

274
00:17:02,525 --> 00:17:07,135
Because as we know, our world completely
reinvents itself every five years.

275
00:17:08,100 --> 00:17:11,110
What we were doing now isn't going
to be what we're doing in five

276
00:17:11,110 --> 00:17:13,360
years time and likewise, vice versa.

277
00:17:14,020 --> 00:17:19,680
you have to carve out time to teach
yourself and to keep yourself up to date.

278
00:17:20,090 --> 00:17:21,720
Otherwise, you'll get left behind.

279
00:17:22,720 --> 00:17:24,630
And that's a problem.

280
00:17:24,770 --> 00:17:26,980
of course it manifests itself
as the imposter syndrome.

281
00:17:27,000 --> 00:17:28,320
We'll talk about that later, I'm sure.

282
00:17:28,830 --> 00:17:32,845
But ultimately, You have to
keep up to date with the latest

283
00:17:32,845 --> 00:17:34,305
and greatest technologies.

284
00:17:34,605 --> 00:17:38,445
That's not to say you have to implement
the latest and greatest technologies, but

285
00:17:38,445 --> 00:17:42,705
you've got to be at least aware of what's
going on, particularly within the domain

286
00:17:43,055 --> 00:17:44,715
that your company is operating within.

287
00:17:44,915 --> 00:17:49,875
What do you think of people who
seem to be successful despite

288
00:17:49,915 --> 00:17:53,275
going completely against basically
everything you said right now?

289
00:17:53,745 --> 00:17:58,635
I'm thinking people like Elon Musk,
who are running multiple companies.

290
00:17:58,695 --> 00:18:04,395
They happen to be CEO, CTO, whatever else
they happen to feel like on that day.

291
00:18:05,085 --> 00:18:10,835
They claim to be, at the very least, like
a chief engineer on the rockets team.

292
00:18:11,725 --> 00:18:13,525
Is that even possible?

293
00:18:14,385 --> 00:18:15,635
What's your take on all of that?

294
00:18:16,095 --> 00:18:19,245
I personally don't think Elon is
as successful as he really is.

295
00:18:20,175 --> 00:18:24,905
and I'll give you another example, Larry
Ellison, 5th richest man in the world.

296
00:18:25,595 --> 00:18:27,385
His title is CTO of Oracle.

297
00:18:27,385 --> 00:18:29,755
Is he really a CTO of Oracle?

298
00:18:30,305 --> 00:18:32,930
Or is there somebody underneath him
that's really playing that role?

299
00:18:32,930 --> 00:18:36,590
So I think there's a lot of facades going
on it's good for the marketing buzz.

300
00:18:36,590 --> 00:18:37,610
It's good for the story.

301
00:18:37,610 --> 00:18:38,710
It's good for that.

302
00:18:39,210 --> 00:18:43,860
But I don't really think that they're
operating at that level, but you're

303
00:18:43,860 --> 00:18:45,210
always going to get your outliers.

304
00:18:45,310 --> 00:18:48,630
You are always gonna get your
98-year-old man that just doesn't seem

305
00:18:48,630 --> 00:18:50,280
to die, but smokes every single day.

306
00:18:50,410 --> 00:18:53,620
contrary to medical science, there's
always gonna be those outliers.

307
00:18:54,100 --> 00:18:57,610
The vast majority of us, we
gotta work with our realities

308
00:18:59,001 --> 00:19:05,501
Alan, for the newly implanted CTOs,
for the CTOs who are just getting

309
00:19:05,501 --> 00:19:11,756
started, Why is it so difficult for
new CTOs to survive the first 100 days?

310
00:19:13,023 --> 00:19:16,173
because you have to fight
absolutely every single instinct

311
00:19:16,473 --> 00:19:18,933
in your body to change stuff.

312
00:19:20,433 --> 00:19:21,833
We love to fix things.

313
00:19:21,903 --> 00:19:22,723
That's what we do.

314
00:19:23,023 --> 00:19:24,513
We see a problem, we fix a problem.

315
00:19:25,343 --> 00:19:30,323
And it is very tempting to go in
and start quickly fixing things.

316
00:19:31,543 --> 00:19:35,833
You see a process that should be
changed, but the first thing you've

317
00:19:35,833 --> 00:19:41,093
got to do is to understand why the
process the way it is, what is the

318
00:19:41,093 --> 00:19:43,103
true value of your human capital?

319
00:19:44,033 --> 00:19:47,163
And what I mean by that is
how strong is your team?

320
00:19:47,263 --> 00:19:51,413
you've got to get to know your team
over a period of time before you

321
00:19:51,413 --> 00:19:55,773
truly know who are the go to people
for whatever thing you need to go to.

322
00:19:55,873 --> 00:19:57,183
Who are the people that are reliable?

323
00:19:57,183 --> 00:20:01,543
Who are the people that switch off
completely the moment they leave work and

324
00:20:01,543 --> 00:20:04,643
therefore they have to remember everything
again when they come back in the next day?

325
00:20:05,223 --> 00:20:09,733
who are the people that know how
everything works in this company?

326
00:20:10,153 --> 00:20:15,578
And you'll always find those one or two
people that, Yeah, that's why we did that.

327
00:20:16,018 --> 00:20:20,098
They will always know the reason why
some hokey process is the way it is,

328
00:20:20,138 --> 00:20:23,498
because it's okay, there was a client
at a given time that needed a given

329
00:20:23,498 --> 00:20:24,528
feature, da, and we never did it.

330
00:20:24,748 --> 00:20:25,948
Okay, so this is where we've got.

331
00:20:25,968 --> 00:20:26,568
Okay, fine.

332
00:20:27,318 --> 00:20:33,778
So the first 100 days, is you just
listening, absolutely listening

333
00:20:33,778 --> 00:20:34,938
and watching and learning.

334
00:20:35,488 --> 00:20:41,238
And one of the analogies I use in the
book is, you're basically like Mel Gibson

335
00:20:41,248 --> 00:20:44,668
in the Maverick movie when he first sits
down at the poker table and he loses

336
00:20:44,678 --> 00:20:46,238
all his money for the first two hours.

337
00:20:48,298 --> 00:20:53,683
He's using that money to buy knowledge
about all the other players on the table.

338
00:20:54,303 --> 00:20:58,313
So when he does start to decide
to start playing, he knows how

339
00:20:58,313 --> 00:20:59,583
to play poker against them all.

340
00:21:00,413 --> 00:21:04,903
So your first 100 days in a company
is not only listening and learning

341
00:21:04,903 --> 00:21:06,353
about your own engineering team.

342
00:21:06,353 --> 00:21:09,963
More importantly is understanding
the wants, the desires and the

343
00:21:09,963 --> 00:21:11,783
frustrations of the business as a whole.

344
00:21:11,783 --> 00:21:13,533
Where are we getting this right?

345
00:21:13,613 --> 00:21:14,983
Where are we getting this wrong?

346
00:21:15,063 --> 00:21:16,393
Where could we do better?

347
00:21:17,913 --> 00:21:22,453
And you're not making anything
materialistically huge

348
00:21:22,803 --> 00:21:24,353
changes in those 100 days.

349
00:21:24,353 --> 00:21:27,643
You may do the odd little things,
tighten up certain meeting structures,

350
00:21:27,673 --> 00:21:32,153
get a little bit more cadence
with respect to daily stand ups or

351
00:21:33,003 --> 00:21:34,993
bulletins, et cetera, whatever it is.

352
00:21:36,093 --> 00:21:40,583
But you're not changing anything
yet because you need to sit down And

353
00:21:40,583 --> 00:21:43,663
then after that 100 days, usually
doesn't take that long, you've

354
00:21:43,663 --> 00:21:47,103
probably got something within the
first sort of 60 days, is figure out

355
00:21:47,323 --> 00:21:48,573
what is going to be your roadmap?

356
00:21:49,093 --> 00:21:50,683
What's your vision for this company?

357
00:21:50,753 --> 00:21:54,093
Where are you going to take this
group going forward in order to

358
00:21:54,093 --> 00:21:55,343
be successful for the company?

359
00:21:55,631 --> 00:21:59,321
Does every company need a CTO?

360
00:22:00,463 --> 00:22:00,873
No.

361
00:22:01,783 --> 00:22:03,063
It's a great question.

362
00:22:03,463 --> 00:22:08,413
And it's very hard to give a
set of checklists as to when you

363
00:22:08,413 --> 00:22:11,113
need a CTO and when you don't.

364
00:22:11,933 --> 00:22:16,323
By and large, if you're producing
technology, then yes, you do.

365
00:22:17,393 --> 00:22:18,173
That's an easy one.

366
00:22:18,933 --> 00:22:22,263
if you're not producing technology,
but you've got a significant amount

367
00:22:22,273 --> 00:22:27,473
of data and you're producing a
lot of aggregated services, etc.

368
00:22:28,298 --> 00:22:30,468
Then yes, you probably do.

369
00:22:30,568 --> 00:22:34,878
At that point in the equation,
the blurry line between CTO

370
00:22:34,878 --> 00:22:37,858
and CIO now starts to come up.

371
00:22:37,868 --> 00:22:39,058
Chief Information Officer.

372
00:22:39,058 --> 00:22:42,528
But by and large, they're both
operating at the same sort of level

373
00:22:42,528 --> 00:22:44,738
and the same goals and what have you.

374
00:22:45,098 --> 00:22:50,778
So in many respects, a CTO can be a CIO
for a non technology producing company.

375
00:22:52,248 --> 00:22:57,138
If, however, you're just A small company,
you're using a handful, maybe one or

376
00:22:57,138 --> 00:23:02,308
two different services such as like
an EHR platform or Salesforce or, if

377
00:23:02,308 --> 00:23:06,548
everything is in one given environment,
you probably don't need a CTO.

378
00:23:07,308 --> 00:23:13,023
every company will need an IT manager iT
is usually not under the remit of the CTO.

379
00:23:13,643 --> 00:23:17,803
So that's basically the person that
ensures that our laptops, our machines,

380
00:23:17,803 --> 00:23:21,183
our back office systems are in check.

381
00:23:21,413 --> 00:23:24,263
It's usually not the CTO
that takes over that role.

382
00:23:25,051 --> 00:23:29,160
I've had this version of this
conversation with a few of my friends

383
00:23:29,479 --> 00:23:37,379
about AI, how now every company is
AI-first, and, I think before AI data

384
00:23:37,379 --> 00:23:40,879
and like data-first or data-driven
was another one of those catchwords

385
00:23:40,879 --> 00:23:42,606
that, were fairly popular as well.

386
00:23:43,204 --> 00:23:43,864
Absolutely.

387
00:23:43,954 --> 00:23:46,224
And mobile-first was
another one before that.

388
00:23:46,314 --> 00:23:46,694
yes.

389
00:23:46,724 --> 00:23:47,524
mobile first.

390
00:23:48,274 --> 00:23:50,784
yeah, we have gone through
our seasons, haven't we?

391
00:23:50,784 --> 00:23:51,744
it was big data.

392
00:23:51,744 --> 00:23:52,874
It was web 2.0.

393
00:23:52,894 --> 00:23:53,694
It was crypto.

394
00:23:54,234 --> 00:23:55,204
now it's AI.

395
00:23:55,244 --> 00:23:58,534
and I think the whole AI world,
it's absolutely fascinating.

396
00:23:58,534 --> 00:24:02,154
We probably talk about that in a
completely separate podcast, but that

397
00:24:02,154 --> 00:24:06,954
one, Particularly for us in the private
equity world, we see AI being splattered

398
00:24:07,044 --> 00:24:11,244
everywhere over a company's deck in
order for them to say, Look, we're so

399
00:24:11,244 --> 00:24:13,164
much more valuable because we use AI.

400
00:24:14,394 --> 00:24:19,334
Yeah, but you're not really using
AI, you're consuming AI, which means

401
00:24:19,894 --> 00:24:21,294
there's no real competitive advantage.

402
00:24:22,254 --> 00:24:22,574
Okay.

403
00:24:22,844 --> 00:24:25,564
Kind of like me saying,
Hey, I'm, I'm Office 365.

404
00:24:26,424 --> 00:24:27,094
So what?

405
00:24:27,584 --> 00:24:30,773
My competitor can be Office 365 too.

406
00:24:30,774 --> 00:24:31,974
What's the big advantage?

407
00:24:32,024 --> 00:24:32,724
There is none.

408
00:24:33,304 --> 00:24:37,234
I think with respect to that, one thing
that we're looking at is, it's more in

409
00:24:37,234 --> 00:24:38,804
the machine learning side of the fence.

410
00:24:39,844 --> 00:24:46,204
what are you utilizing your data for that
is unique to you that only you've got?

411
00:24:47,494 --> 00:24:52,214
As opposed to just slapping on those two
little letters and hoping that people

412
00:24:52,214 --> 00:24:56,654
will get starstruck by whatever that AI
is going to produce for your company.

413
00:24:57,846 --> 00:25:04,706
So I wanted to transition now and,
unveil a few of, the bigger thoughts

414
00:25:04,776 --> 00:25:06,676
that, people will find in your book.

415
00:25:07,066 --> 00:25:12,316
But before we go there, can we
also talk a little bit about

416
00:25:12,316 --> 00:25:13,846
the negative side of this?

417
00:25:14,076 --> 00:25:19,696
What would you say is the least
pleasant part of being a CTO?

418
00:25:21,051 --> 00:25:23,501
You're the first person to
ever ask me that question.

419
00:25:23,551 --> 00:25:27,901
I think from that perspective,
it's the culture shock that you

420
00:25:27,901 --> 00:25:29,471
don't get to do coding anymore.

421
00:25:29,471 --> 00:25:33,001
Or you shouldn't be doing coding anymore,
particularly if you're a CTO from a

422
00:25:33,001 --> 00:25:34,761
developer's engineering background.

423
00:25:34,761 --> 00:25:40,776
Because ultimately, you will have far
more responsibility Then just simply

424
00:25:40,786 --> 00:25:46,276
producing code and checking it in and
you never want to be the bottleneck for

425
00:25:46,276 --> 00:25:52,156
a project or a feature to be released,
because you were pulled into an executive

426
00:25:52,156 --> 00:25:55,696
meeting, you were pulled into whatever
fire was raising at that point, you

427
00:25:55,696 --> 00:26:00,106
couldn't get to that code or you realize
you probably weren't as good as you

428
00:26:00,126 --> 00:26:04,856
thought you were in the code and what
you did slap dash in is now proving to

429
00:26:04,856 --> 00:26:07,276
be more problems than it was solving.

430
00:26:08,046 --> 00:26:09,316
And you're still holding people up.

431
00:26:09,336 --> 00:26:11,586
But because you've been
pulled elsewhere, it's there.

432
00:26:12,056 --> 00:26:17,176
So that's a very hard thing for
CTOs to let go of the keyboard.

433
00:26:18,961 --> 00:26:23,061
Now, I personally, I carve out small
little side projects for myself

434
00:26:23,061 --> 00:26:24,721
in order to keep myself relevant.

435
00:26:25,231 --> 00:26:29,681
But I make sure that those side projects
are never critical path projects.

436
00:26:29,751 --> 00:26:33,481
I'm more than happy to bootstrap
an idea or kickstart some

437
00:26:33,491 --> 00:26:35,151
thoughts to prove the concept.

438
00:26:35,911 --> 00:26:39,506
And then hand it over and say,
'Okay, productionalize this.

439
00:26:40,206 --> 00:26:42,706
make it to make sure that it'll
never fall over' type stuff.

440
00:26:43,376 --> 00:26:46,506
That's my personal way of
keeping my hands on the keyboard.

441
00:26:47,116 --> 00:26:51,566
But, there would be weeks that would go
by I would rarely open up visual code.

442
00:26:51,596 --> 00:26:55,046
and a little bit dies inside
of me when that happens.

443
00:26:55,046 --> 00:26:57,966
But that's the reality of
the role that we playing.

444
00:26:58,701 --> 00:27:02,861
I think to an extent, it's also
shared with, other slightly higher

445
00:27:02,861 --> 00:27:08,981
level positions that not necessarily
CTO, some staff engineers I know,

446
00:27:09,551 --> 00:27:11,361
they struggle with the same.

447
00:27:11,461 --> 00:27:17,041
That's, to scale the impact they have
to decrease the percentage of time

448
00:27:17,041 --> 00:27:21,881
to actually doing hands-on work and
focus more on things like architecture,

449
00:27:22,481 --> 00:27:27,491
focus on resolving, potentially
political issues around, internal

450
00:27:27,491 --> 00:27:29,711
office, It never lands well with them.

451
00:27:29,801 --> 00:27:35,061
I think there's just something about the
type of people who end up doing this roles

452
00:27:35,071 --> 00:27:38,861
that it's always like a cross to bear

453
00:27:39,646 --> 00:27:40,066
It is.

454
00:27:40,086 --> 00:27:40,746
It absolutely is.

455
00:27:40,826 --> 00:27:44,586
if you've got an engineer's
heart, then feed that passion.

456
00:27:45,136 --> 00:27:48,386
It may not be 80% of your day anymore.

457
00:27:48,396 --> 00:27:52,576
It may only be like 10 or 20%
of the day, but keep feeding it.

458
00:27:53,356 --> 00:27:55,196
Don't take them out of the server.

459
00:27:55,196 --> 00:27:56,736
Okay, you're now just project management.

460
00:27:56,836 --> 00:27:59,226
You're now going to make sure this
team of engineers is doing everything

461
00:27:59,226 --> 00:28:00,456
right and all you do is code review.

462
00:28:01,606 --> 00:28:04,316
They are just going to hit a
wall and one day just leave.

463
00:28:05,156 --> 00:28:11,336
feed the passion and carve out time
to allow them to basically play and

464
00:28:11,336 --> 00:28:17,606
to innovate and to truly feed that so
they don't get bogged down with 'Ugh,

465
00:28:18,246 --> 00:28:22,261
I spent the first six hours of every
day doing emails and Jira updates and I

466
00:28:22,671 --> 00:28:24,261
Is this what my life has come to now?'

467
00:28:25,721 --> 00:28:26,981
Some people enjoy that.

468
00:28:27,691 --> 00:28:28,311
Great.

469
00:28:29,441 --> 00:28:32,061
But if you've come from an
engineer's heart, you probably don't.

470
00:28:33,131 --> 00:28:36,761
For anybody who is not going
to stop listening now and say,

471
00:28:36,781 --> 00:28:41,921
okay, now I can't do that wants
to learn more about the CTOs.

472
00:28:42,031 --> 00:28:47,581
your book is basically, a long list
of chapters covering on different

473
00:28:47,641 --> 00:28:53,551
focus areas for CTO, like planning,
division, naming narrative.

474
00:28:54,026 --> 00:28:58,036
Interviewing onboarding, building
the team, growing people that you've

475
00:28:58,036 --> 00:29:03,696
already hired, obviously the tech
decisions themselves, the development,

476
00:29:03,736 --> 00:29:05,416
documentation, all that stuff.

477
00:29:05,936 --> 00:29:12,746
So let's go through some of this
and try to focus on one piece

478
00:29:12,746 --> 00:29:16,686
of advice that you consider the
highest return on investment there.

479
00:29:16,736 --> 00:29:21,196
the first thing that you said about the
CTO, the grand vision, what would you say

480
00:29:21,216 --> 00:29:26,696
is like the one thing that might be the
most important, for a new starting CTO?

481
00:29:27,696 --> 00:29:31,906
I'll give you an example when
we look to buy a company, we'll,

482
00:29:32,126 --> 00:29:33,436
part of that is due diligence.

483
00:29:33,506 --> 00:29:36,346
And that's really just getting to
know the team and getting to know

484
00:29:36,476 --> 00:29:37,956
what the company is, et cetera.

485
00:29:38,436 --> 00:29:41,996
And it's also verifying that they're
doing what they say they're doing.

486
00:29:42,386 --> 00:29:42,816
Okay.

487
00:29:42,826 --> 00:29:48,206
And I will run up the technical
diligence on that side of the fence.

488
00:29:48,896 --> 00:29:51,326
And I love talking to people.

489
00:29:51,326 --> 00:29:52,366
I just absolutely love.

490
00:29:52,991 --> 00:29:57,331
Especially if you've found an engaging
CTO or an engaging VP of engineering.

491
00:29:57,331 --> 00:30:02,111
What I usually do is to start
off the meeting is get them to

492
00:30:02,131 --> 00:30:03,471
stand in front of a whiteboard.

493
00:30:04,281 --> 00:30:08,201
I say, draw me the five year plan
that you've got for this company.

494
00:30:08,601 --> 00:30:12,441
If they can draw it like that,
then they're a great CTO.

495
00:30:13,391 --> 00:30:17,676
If they can't or they struggle, And
they can't get past six months or they

496
00:30:17,676 --> 00:30:19,946
can't get past a handful of features.

497
00:30:20,116 --> 00:30:21,206
You're not a CTO.

498
00:30:22,406 --> 00:30:26,526
You're more of a VP of engineering you're
not showing a path of where the company

499
00:30:26,546 --> 00:30:33,166
can go because in many respects, even
with this AI stuff, your average C level

500
00:30:33,506 --> 00:30:39,436
isn't going to have a clue what the hell
they can do with AI, but you as a CTO,

501
00:30:40,316 --> 00:30:41,826
you're supposed to be able to lay out.

502
00:30:41,826 --> 00:30:43,306
Here's what we could do with AI.

503
00:30:44,476 --> 00:30:47,426
I'm not saying we're going to do this
with AI, because that's an executive

504
00:30:47,426 --> 00:30:51,116
decision we make as a company, but
I'm going to show you what's possible.

505
00:30:51,716 --> 00:30:53,526
Or I'm going to show you what's
possible with the mobile.

506
00:30:53,696 --> 00:30:56,356
Or I'm going to show you what's possible
if I'm in a manufacturing company.

507
00:30:56,606 --> 00:31:00,576
Here are the latest and greatest IoT
devices, here's what we could do.

508
00:31:00,766 --> 00:31:03,146
So I'm going to show you a possibilities.

509
00:31:03,146 --> 00:31:05,166
And you're also going
to let yourself dream.

510
00:31:05,926 --> 00:31:08,236
If we were to go down a certain path,
then we could do this, then we could do

511
00:31:08,236 --> 00:31:09,556
this, we could do this, we could do this.

512
00:31:09,556 --> 00:31:14,126
And that sort of seeds the
c-level as to what is possible.

513
00:31:14,126 --> 00:31:19,026
Now once the c-level has decided
'we're going in this direction', now

514
00:31:19,336 --> 00:31:24,256
you have to effectively put a lot more
meat on those bones and truly come up

515
00:31:24,256 --> 00:31:26,586
with a vision plan to walk that path.

516
00:31:27,391 --> 00:31:30,741
And to show that here's where we're
going to be at each step of the way

517
00:31:30,941 --> 00:31:33,941
in year one, we'll be here, year
two, we'll be here, year three.

518
00:31:34,121 --> 00:31:37,681
Hey, it could all change depending
on what we learn in year two and

519
00:31:37,691 --> 00:31:39,051
depending on what the industry has done.

520
00:31:39,051 --> 00:31:41,281
But hey, assuming everything is
going the right way we're going

521
00:31:41,281 --> 00:31:42,561
to continue along this path.

522
00:31:42,561 --> 00:31:49,791
Now that is a far more engaging and
exciting person that is being proactive

523
00:31:50,391 --> 00:31:54,741
as opposed to sitting there waiting
to be reactive as to say, 'okay, we'll

524
00:31:54,741 --> 00:31:55,971
wait and to see what our clients say'.

525
00:31:56,586 --> 00:32:00,496
We'll wait and see what features they
want, or we'll wait and see what comes up.

526
00:32:00,496 --> 00:32:03,366
That's not a CTO as far as I'm concerned.

527
00:32:03,696 --> 00:32:06,726
So that's what I'm looking for there,
is somebody that truly knows what it is.

528
00:32:06,866 --> 00:32:10,846
Now part of that vision planning
could be moving off of legacy systems.

529
00:32:11,566 --> 00:32:13,601
Because, it's aging out.

530
00:32:14,071 --> 00:32:18,021
We can't get the resources, either
hardware, software or even human

531
00:32:18,021 --> 00:32:19,671
capital to support it anymore.

532
00:32:20,071 --> 00:32:22,131
It's causing us more
problems than it needs to be.

533
00:32:22,341 --> 00:32:25,721
So that needs to be vision
planned out as well.

534
00:32:25,721 --> 00:32:27,341
So all of that.

535
00:32:27,341 --> 00:32:32,061
I'd like to see a whiteboard completely
filled with arrows and plans and

536
00:32:32,061 --> 00:32:34,911
dreams and but it comes naturally.

537
00:32:35,191 --> 00:32:38,581
And you'll know when somebody
sells you that story.

538
00:32:39,441 --> 00:32:44,101
And that's the other part of a
good CTO is, can you sell me the

539
00:32:44,101 --> 00:32:46,801
vision without using buzzwords?

540
00:32:46,801 --> 00:32:50,891
Can you sell me the vision without
making me feel intimidated because

541
00:32:50,891 --> 00:32:52,371
you're using things like blockchain?

542
00:32:52,411 --> 00:32:54,061
I have no clue what
the hell blockchain is.

543
00:32:54,061 --> 00:32:54,811
What does it mean?

544
00:32:55,361 --> 00:32:56,441
I don't give a crap.

545
00:32:56,861 --> 00:32:58,251
What does it do for me as a business?

546
00:32:58,511 --> 00:32:59,931
What does it do for me as a customer?

547
00:33:00,511 --> 00:33:03,381
Cloud, I don't give a crap
about cloud, what does it mean?

548
00:33:03,681 --> 00:33:08,311
you need to be able to articulate that
in such a way that the CEO and everybody

549
00:33:08,311 --> 00:33:14,326
else on the chessboard can contribute
their thoughts and their ideas in such a

550
00:33:14,326 --> 00:33:20,086
way that you're not coming up and saying,
'this is the way it's going to be'.

551
00:33:20,696 --> 00:33:23,546
I'm the technologist, you don't
know technology, trust me,

552
00:33:23,546 --> 00:33:24,436
this is the way it's going.

553
00:33:25,646 --> 00:33:29,186
I've met many people like
that and they fail quickly.

554
00:33:30,366 --> 00:33:34,246
You need to be able to bring
along every other person.

555
00:33:34,686 --> 00:33:37,296
Bringing along your engineering
group, relatively easy.

556
00:33:37,296 --> 00:33:39,826
They're in the trenches, they
know the problems, they know

557
00:33:39,956 --> 00:33:42,236
also where their industry is
going, they know the buzzwords.

558
00:33:42,976 --> 00:33:43,986
Preaching to the choir.

559
00:33:43,986 --> 00:33:47,856
You may get into a tabs versus
spaces argument with some of your

560
00:33:47,856 --> 00:33:51,646
engineers as to which framework,
which language, that's fine, those

561
00:33:51,646 --> 00:33:53,016
are good conversations to have.

562
00:33:53,726 --> 00:34:00,431
But from a business perspective, can
you sell what it is you're doing in such

563
00:34:00,461 --> 00:34:05,091
a way that they can understand it and
also get excited about it because When

564
00:34:05,091 --> 00:34:11,021
you're selling your vision to the CEO
He's not the end person or she's not the

565
00:34:11,021 --> 00:34:14,601
end person, because they've got to then
go and sell it to the board if you're

566
00:34:14,601 --> 00:34:16,761
not Representing on the board yourself.

567
00:34:17,001 --> 00:34:18,781
They're going to represent
it to the clients.

568
00:34:18,881 --> 00:34:23,681
They're going to represent it to the
investors So you're really telling them

569
00:34:23,681 --> 00:34:29,681
the soundbites That they can then reuse
using their words and their understanding.

570
00:34:29,681 --> 00:34:34,271
So whenever they tell the vision of where
the company is going, it sounds authentic.

571
00:34:34,441 --> 00:34:40,471
It sounds real and we've all met people
where you're thinking 'you've no clue

572
00:34:40,481 --> 00:34:41,791
what that buzzword means, do you.

573
00:34:42,421 --> 00:34:46,941
You're sounding confident, but you've no
clue what you're talking about, do you?'

574
00:34:48,061 --> 00:34:52,561
And that's a bad thing, you always want
to make sure that your CEO, your board,

575
00:34:52,561 --> 00:34:58,261
your investors understand completely
what it is you're trying to sell.

576
00:34:58,801 --> 00:35:03,237
So to an extent you're actually
in a translator role, right?

577
00:35:03,319 --> 00:35:09,710
You have to explain these, very,
Boring to a lot of people, details of

578
00:35:09,719 --> 00:35:14,065
what is going to enable the company
to achieve the things that it wants

579
00:35:14,065 --> 00:35:17,619
to on the technical level without
actually using the technical terms.

580
00:35:18,531 --> 00:35:21,911
you're absolutely right and it's
something I say to all of my engineers

581
00:35:21,921 --> 00:35:26,221
at all levels is the biggest part
of your job isn't QA'ing, isn't

582
00:35:26,221 --> 00:35:29,541
producing code, isn't documenting.

583
00:35:30,191 --> 00:35:31,611
You're a salesperson.

584
00:35:32,121 --> 00:35:35,921
That's the biggest thing that you're
doing because at every level we're always

585
00:35:35,921 --> 00:35:40,996
selling what we're doing, we're always
trying to persuade somebody that this

586
00:35:40,996 --> 00:35:42,696
function is better than this function.

587
00:35:42,866 --> 00:35:44,626
This feature is better than this feature.

588
00:35:44,626 --> 00:35:45,876
This way is better than this way.

589
00:35:45,936 --> 00:35:51,396
We're always trying to influence,
persuade, sell our thoughts and

590
00:35:51,396 --> 00:35:53,296
our ideas to those around us.

591
00:35:54,186 --> 00:35:59,411
The only difference at a CTO level is
to your point, you're not allowed to use

592
00:35:59,721 --> 00:36:03,071
all of the buzzwords and the ingredients
that you're usually allowed to use.

593
00:36:03,121 --> 00:36:07,966
Now, the only time you are allowed to
use buzzwords And it's again, my little

594
00:36:07,966 --> 00:36:12,856
litmus test is, as soon as either one
of the big tabloids start using the

595
00:36:12,856 --> 00:36:17,356
buzzword, CNN, New York Times, Wall
Street Journal, pick whatever one it

596
00:36:17,366 --> 00:36:21,656
is, as soon as they start using the
buzzword, you're not allowed to use that

597
00:36:21,656 --> 00:36:27,486
buzzword, but you have to caution that
buzzword to make sure that, okay, let's

598
00:36:27,486 --> 00:36:30,111
have a quick session on what, Crypto is.

599
00:36:30,541 --> 00:36:32,801
Let's have a quick session
on what blockchain is.

600
00:36:33,091 --> 00:36:38,661
So you become a teacher, you become an
educator in order to make sure that when

601
00:36:38,661 --> 00:36:44,811
you're giving that five minute overview of
what that new term that has been published

602
00:36:44,811 --> 00:36:50,821
in the Wall Street Journal really is, that
the c-level can go, 'ha, I got it now.

603
00:36:51,591 --> 00:36:52,861
I actually get it'.

604
00:36:53,781 --> 00:36:55,991
Because again, we're also running hard.

605
00:36:57,476 --> 00:37:01,506
We usually don't take the time to explore
some of these words and we hope that by

606
00:37:01,516 --> 00:37:05,006
the time we've read it for the sixth,
seventh, eighth time, we'll figure out

607
00:37:05,006 --> 00:37:07,946
the context that it keeps being used
in that we'll be able to bluff our

608
00:37:07,946 --> 00:37:11,116
way until such times we've got a true
understanding of what it really means.

609
00:37:11,866 --> 00:37:16,536
But a CTO should always be putting
out thought pieces to the company

610
00:37:16,826 --> 00:37:20,886
to say, 'Hey, you've probably seen
this' or one that I always love to do.

611
00:37:21,246 --> 00:37:25,726
If you realize of a big public hack
or a security breach has happened.

612
00:37:27,411 --> 00:37:28,341
Use that.

613
00:37:29,071 --> 00:37:30,591
Send it out to your company.

614
00:37:31,571 --> 00:37:32,441
Break it down.

615
00:37:32,531 --> 00:37:34,031
It may have nothing to do with you.

616
00:37:34,251 --> 00:37:37,881
But educate all of your employees
as to how did that happen?

617
00:37:38,011 --> 00:37:39,391
What steps were missed?

618
00:37:39,891 --> 00:37:43,501
What was the consequences
of that particular stuff?

619
00:37:43,981 --> 00:37:45,336
Become a thought leader.

620
00:37:46,106 --> 00:37:51,216
in technology in such a way that people
feel you're far more approachable and

621
00:37:51,226 --> 00:37:56,486
therefore you're not the guy that's going
to make you feel this height because

622
00:37:56,486 --> 00:37:57,996
you come and ask them a dumb question.

623
00:37:58,936 --> 00:38:03,526
That reminds me of, a few engineering
blogs that I enjoy reading.

624
00:38:04,006 --> 00:38:09,656
Do you think that's a good use
of, CTO's time to do this thought

625
00:38:09,746 --> 00:38:12,566
leadership really that you're
talking about in the form of a blog.

626
00:38:13,326 --> 00:38:14,846
Yes, for two reasons.

627
00:38:15,096 --> 00:38:19,076
One, it gives you time to
sit down and think about what

628
00:38:19,076 --> 00:38:20,886
it is you're writing about.

629
00:38:21,896 --> 00:38:25,756
At university, for example, when you
were at a lecture, those that wrote

630
00:38:25,766 --> 00:38:30,276
down the notes had a higher likelihood
of remembering what was going on.

631
00:38:31,166 --> 00:38:35,716
So a blog helps you with that part of
the equation in a modern day environment.

632
00:38:36,346 --> 00:38:39,646
What it also allows you to do, which
is where I get a lot of comfort from it

633
00:38:39,896 --> 00:38:45,576
as I run a lot of internal blogs, is it
allows me to sit back and say, 'okay,

634
00:38:46,286 --> 00:38:50,076
if I'm to read this without all the
buzzwords, am I communicating properly?'

635
00:38:50,461 --> 00:38:53,381
Can I remove words in such a
way that I can get this right?

636
00:38:53,381 --> 00:38:59,101
So it allows me to truly hone in
on what it is I'm trying to say and

637
00:38:59,101 --> 00:39:00,621
what it is I'm trying to communicate.

638
00:39:01,011 --> 00:39:03,971
I'm going to make the assumption
that the reader is going to skim it.

639
00:39:05,111 --> 00:39:06,861
They're not going to
really read it in depth.

640
00:39:07,141 --> 00:39:10,221
So I've got to be able to write
it in such a way that, that I

641
00:39:10,221 --> 00:39:12,331
get what I need to get across.

642
00:39:12,661 --> 00:39:14,161
And it's a great.

643
00:39:14,746 --> 00:39:21,906
tool to allow you to tighten up and
to really make your email succinct.

644
00:39:22,626 --> 00:39:26,562
That's actually one of my
own favorite litmus tests.

645
00:39:26,626 --> 00:39:30,244
if you can't explain it to someone
else, that means you don't understand

646
00:39:30,244 --> 00:39:34,590
it, uh, well enough, but it can
also be a big time commitment.

647
00:39:34,590 --> 00:39:40,121
writing a proper in-depth, blog post
that can easily take a day or two.

648
00:39:40,323 --> 00:39:41,193
Oh, completely.

649
00:39:41,523 --> 00:39:46,203
But, you never know when that
blog post will pay off for you.

650
00:39:46,623 --> 00:39:48,443
It may be turned into a white paper.

651
00:39:48,463 --> 00:39:50,063
It may be turned into a green paper.

652
00:39:50,493 --> 00:39:55,483
It may be that the CEO suddenly
says, Hey Alan, I need you to

653
00:39:55,483 --> 00:39:57,183
talk to a big potential customer.

654
00:39:57,183 --> 00:39:59,373
They have no clue about
this particular technology.

655
00:39:59,403 --> 00:40:00,863
Can you help walk them through it?

656
00:40:01,463 --> 00:40:05,258
you never know when on a dime
you're going to be asked to turn

657
00:40:05,528 --> 00:40:07,248
and to present and to do something.

658
00:40:07,248 --> 00:40:10,278
So I always like to be prepared
for that sort of stuff.

659
00:40:10,418 --> 00:40:14,688
And it's something I go into with respect
to, the future planning of a company is

660
00:40:14,698 --> 00:40:17,838
that, particularly when a company gets
to the point where it's maybe being

661
00:40:17,838 --> 00:40:21,858
sold, you'll never know when suddenly
you'll get a tap on the door and say,

662
00:40:21,858 --> 00:40:25,458
Hey, got some people I'd love you to
talk to, could you walk them through

663
00:40:26,208 --> 00:40:27,508
what we do and where we're going?

664
00:40:27,958 --> 00:40:31,078
And that's basically code word for,
'hey, we've got a bunch of new buyers.

665
00:40:31,858 --> 00:40:33,638
Can you put our best foot forward here?'

666
00:40:33,979 --> 00:40:36,588
So a hint for everybody listening to this.

667
00:40:36,588 --> 00:40:40,570
I guess if Alan is looking into
your company That's the thing to do.

668
00:40:41,152 --> 00:40:42,442
Go and write some blog posts.

669
00:40:42,442 --> 00:40:43,497
That's a green flag

670
00:40:43,802 --> 00:40:45,002
It absolutely is.

671
00:40:45,162 --> 00:40:47,412
And an excellent book that
I would recommend people

672
00:40:47,412 --> 00:40:48,602
getting is "Smart Brevity".

673
00:40:49,382 --> 00:40:51,102
it's a 70-page book.

674
00:40:51,102 --> 00:40:55,513
It's well worth the investment,
It helps explain how to tighten

675
00:40:55,513 --> 00:40:58,753
up your communication in such
a way that you use a lot less

676
00:40:58,753 --> 00:41:01,013
words, but communicate a lot more.

677
00:41:02,313 --> 00:41:05,863
I hate to be promoting somebody else's
book while I'm promoting my own, but

678
00:41:05,913 --> 00:41:08,593
hey, there's enough for everybody.

679
00:41:09,233 --> 00:41:12,093
Yeah, there's this quote that I'm
forgetting now I think it was Blaise

680
00:41:12,233 --> 00:41:18,092
Pascal saying that if I had more time, I
would have written a shorter letter and

681
00:41:18,093 --> 00:41:21,813
that's definitely a good motto to have,
I hope I didn't misattribute that now

682
00:41:23,468 --> 00:41:24,128
Let's move on.

683
00:41:24,508 --> 00:41:29,238
so we've got a vision, the CTO,
a CEO, we all agreed on the

684
00:41:29,248 --> 00:41:30,998
grand vision, going forward.

685
00:41:31,518 --> 00:41:35,148
And now, we need people to
actually make that vision possible.

686
00:41:35,848 --> 00:41:36,408
What do we do?

687
00:41:36,878 --> 00:41:39,258
How do we go about
interviewing and onboarding?

688
00:41:40,130 --> 00:41:43,000
We first of all have to
decide who do we need.

689
00:41:43,000 --> 00:41:47,070
What skills do we need and
what skills will we need today?

690
00:41:47,090 --> 00:41:48,870
And what skills will we need tomorrow?

691
00:41:48,980 --> 00:41:54,430
And that's very important because you
want to be able to bring in people that

692
00:41:54,430 --> 00:41:59,690
can evolve to where you need them to
go in the future because your product

693
00:41:59,730 --> 00:42:01,750
will grow, your product will change.

694
00:42:01,750 --> 00:42:02,480
It will adapt.

695
00:42:02,490 --> 00:42:04,500
So you need to have people that's
going to come along with you.

696
00:42:05,320 --> 00:42:11,865
from that perspective, you also have
to figure out my timing in all of this.

697
00:42:12,415 --> 00:42:16,495
do I hire engineers anything
from coders to data scientists to

698
00:42:16,495 --> 00:42:19,235
anybody that's under the CTO banner.

699
00:42:19,235 --> 00:42:21,675
Don't keep assuming I'm always
talking about programmers.

700
00:42:22,205 --> 00:42:26,685
so when you're hiring your team, do
you need to have full time employees?

701
00:42:27,215 --> 00:42:33,320
Or can you bring in contractors to
get you initially past a given point

702
00:42:33,870 --> 00:42:39,000
or is this something whereby I could
partner with a third party development

703
00:42:39,000 --> 00:42:45,950
company and let them do that part and
instead of us investing in say 10 or

704
00:42:45,950 --> 00:42:51,370
20 people, let them do that because we
may need 10 or 20 people for the next

705
00:42:51,370 --> 00:42:54,580
six months But after that once the
development is done, we'll probably

706
00:42:54,580 --> 00:42:56,690
be down to five or 10 whatever it is.

707
00:42:56,980 --> 00:42:59,690
So do I want to have to downsize my group?

708
00:42:59,690 --> 00:43:00,800
after six months.

709
00:43:00,920 --> 00:43:01,430
I don't know.

710
00:43:01,970 --> 00:43:06,290
it's about figuring out your
overall plan and working closely

711
00:43:06,290 --> 00:43:08,180
with the CFO and the CEO.

712
00:43:08,630 --> 00:43:11,690
The three of you have to figure
this out together once you've given

713
00:43:11,690 --> 00:43:15,870
certain options, and those options
will be, a factor of time and cost.

714
00:43:17,235 --> 00:43:20,445
And if you want to go faster,
it's probably going to cost more.

715
00:43:20,525 --> 00:43:23,025
All of the usual, stuff that
comes in and around there.

716
00:43:23,305 --> 00:43:26,445
And that's where having a
conversation with both of those

717
00:43:26,555 --> 00:43:28,485
roles will help you solidify.

718
00:43:28,495 --> 00:43:31,975
Because they will know
what they can tolerate.

719
00:43:32,575 --> 00:43:36,285
Both from a, I can wait six months
for this feature or I can only wait

720
00:43:36,285 --> 00:43:40,405
three months for this feature or I've
only got a million dollars versus

721
00:43:40,465 --> 00:43:43,505
a hundred thousand dollars to spend
on this at this precise moment.

722
00:43:44,425 --> 00:43:47,285
So that's a very important
thing to do there.

723
00:43:47,445 --> 00:43:51,785
So once you've decided that, then it's
effectively an interviewing process, and

724
00:43:51,785 --> 00:43:55,055
the interviewing process should always
be the same irrespective of whether

725
00:43:55,055 --> 00:44:00,590
that person is going to be on the team
full time or as part of another company.

726
00:44:00,970 --> 00:44:04,560
You're interviewing the company, but I
always like to see the resumes of the sort

727
00:44:04,610 --> 00:44:08,520
of people that are going to be working
with me, even though they're going to be

728
00:44:08,520 --> 00:44:10,050
working with me for a short space of time.

729
00:44:10,840 --> 00:44:11,607
they're still humans.

730
00:44:11,657 --> 00:44:16,187
You cannot simply trust the fact
that this outsource company has said,

731
00:44:16,507 --> 00:44:19,477
'yeah, we'll give you ten React.

732
00:44:19,497 --> 00:44:21,047
js developers in the world best'.

733
00:44:21,672 --> 00:44:22,742
Yeah, but are they?

734
00:44:23,322 --> 00:44:23,832
Really?

735
00:44:24,092 --> 00:44:24,502
Okay.

736
00:44:24,752 --> 00:44:25,832
Can I talk to one of them?

737
00:44:25,892 --> 00:44:26,172
Can I?

738
00:44:26,432 --> 00:44:27,072
Let's meet them.

739
00:44:27,072 --> 00:44:30,852
Let's have a conversation just to
make sure that they fit culturally.

740
00:44:30,852 --> 00:44:33,792
They've got the sort of ethics
that I want to lay down.

741
00:44:33,802 --> 00:44:36,982
They've got the code quality,
the standards, the way I want

742
00:44:36,982 --> 00:44:38,482
to work and produce this.

743
00:44:38,692 --> 00:44:41,952
Even though they're a third party
company, you're still a boss.

744
00:44:42,192 --> 00:44:46,012
You're still the one that's
paying their salaries effectively.

745
00:44:46,115 --> 00:44:51,975
I also learned that, and surprisingly
took me a good few years into my technical

746
00:44:51,975 --> 00:44:56,975
jobs to understand that a collection
of individuals, as impressive as they

747
00:44:56,975 --> 00:45:02,075
might be, each one of them separately,
doesn't necessarily make a team.

748
00:45:03,345 --> 00:45:06,525
How do you build a team out of your hires?

749
00:45:06,525 --> 00:45:06,725
What

750
00:45:06,840 --> 00:45:08,395
advice do you great point.

751
00:45:08,785 --> 00:45:10,275
don't hire all rock stars.

752
00:45:11,920 --> 00:45:13,800
off the bat, don't hire all rock stars.

753
00:45:14,230 --> 00:45:16,800
they will always be competing,
and overwriting each other,

754
00:45:16,840 --> 00:45:18,370
and undermining each other.

755
00:45:18,990 --> 00:45:21,120
there's a reason there's only
one rock star in every band.

756
00:45:21,120 --> 00:45:23,330
The Rolling Stones has got
Mick Jagger, live with it.

757
00:45:23,330 --> 00:45:25,330
Queen, Freddy Mercury, live with it.

758
00:45:25,870 --> 00:45:29,410
Yes, there's other people in the
band, but they're not all rock stars.

759
00:45:30,170 --> 00:45:34,200
fundamentally, You need a mixture
of senior, mid level and junior.

760
00:45:35,060 --> 00:45:36,630
Everybody needs to be able to aspire.

761
00:45:37,170 --> 00:45:39,880
Everybody needs to be able
to learn from another.

762
00:45:40,410 --> 00:45:42,300
Not every developer is equal.

763
00:45:42,540 --> 00:45:44,080
Not every role is equal.

764
00:45:44,110 --> 00:45:46,100
You'll have different levels
doing different things.

765
00:45:46,200 --> 00:45:51,460
The human tends towards certain things
because they like that particular part of

766
00:45:51,460 --> 00:45:55,610
it, but they don't like that part of it,
but on paper they're both the same title.

767
00:45:56,180 --> 00:45:59,110
so you want a good blend of people.

768
00:45:59,570 --> 00:46:05,910
And more importantly, skills
can be taught, but an asshole

769
00:46:05,910 --> 00:46:06,840
will always be an asshole.

770
00:46:06,840 --> 00:46:08,460
So culture is huge.

771
00:46:08,460 --> 00:46:11,810
You want to define what your
culture is going to be and I've

772
00:46:11,810 --> 00:46:13,300
seen many different variations.

773
00:46:13,350 --> 00:46:16,910
From one end of the spectrum you've
got the sort of the real gritty team

774
00:46:16,910 --> 00:46:22,040
that will curse and swear and I mean
it's it truly is like a builder's

775
00:46:22,040 --> 00:46:25,540
yard but they all love and appreciate
each other It's their love language.

776
00:46:26,090 --> 00:46:28,920
Then you've got the other end
of the equation where it's all

777
00:46:28,920 --> 00:46:31,890
horrendously polite, we use
our title, we use our stuff.

778
00:46:31,890 --> 00:46:35,970
I first started out my career
in the Ministry of Defence.

779
00:46:36,450 --> 00:46:41,620
I was a civilian working in a military
environment and I was struggling big

780
00:46:41,620 --> 00:46:47,438
time because I kept forgetting the
bands on everybody's shirts as to

781
00:46:47,438 --> 00:46:50,910
how I was supposed to address them.

782
00:46:51,470 --> 00:46:57,280
So I was insulting managers,
inadvertently, by simply not being

783
00:46:57,280 --> 00:46:58,570
his part of that environment.

784
00:46:58,620 --> 00:47:01,130
so I did not succeed in that environment.

785
00:47:02,600 --> 00:47:07,180
and what was interesting was I then went
to Harman Kardon, that was my eye opener.

786
00:47:07,915 --> 00:47:12,145
Going from shirt and tie military
situation to effectively sandals,

787
00:47:12,155 --> 00:47:16,365
open shoes, and hippie culture,
and sound studio, it was like,

788
00:47:16,455 --> 00:47:18,655
'oh my god, these are my people!

789
00:47:19,195 --> 00:47:24,745
I love this!' So you gotta find
the right people that will fit

790
00:47:24,765 --> 00:47:26,585
into your personality of your team.

791
00:47:27,185 --> 00:47:33,015
Because it can just take one person
to truly spoil the rest of the team.

792
00:47:34,285 --> 00:47:36,735
And that's hard, that truly is hard.

793
00:47:36,735 --> 00:47:41,865
So you've got to be very upfront in your
interview to make sure that you test them.

794
00:47:41,875 --> 00:47:45,615
So if, for example, you are a
very sweary environment, then

795
00:47:45,615 --> 00:47:46,875
let them know in the interview.

796
00:47:46,875 --> 00:47:49,075
Would this make you feel uncomfortable?

797
00:47:49,335 --> 00:47:52,875
Or likewise, this is a
very strict environment.

798
00:47:52,915 --> 00:47:54,835
You've got to be here at 9 o'clock.

799
00:47:54,835 --> 00:47:58,485
If you're not here at 9:15, then we're
going to assume something's wrong.

800
00:47:59,285 --> 00:48:01,125
Are you okay in that environment?

801
00:48:01,625 --> 00:48:04,845
Whatever it is, whatever the
blended environment is, make

802
00:48:04,845 --> 00:48:07,015
sure culturally they fit.

803
00:48:07,935 --> 00:48:11,755
Because there's so many different
versions of it, and that's what

804
00:48:11,755 --> 00:48:13,065
makes this world so wonderful.

805
00:48:13,065 --> 00:48:17,535
But finding the right person will,
by and large, outstrip any technical

806
00:48:17,535 --> 00:48:22,495
skills that they'll bring to the
role because we can teach them how

807
00:48:22,495 --> 00:48:26,755
to be a better React developer, or
how to be a better QA person, or how

808
00:48:26,755 --> 00:48:30,665
to do the new version of SQL Server
because they're on two versions below.

809
00:48:31,375 --> 00:48:34,285
If they're culturally the right
fit, I want that person on my team.

810
00:48:34,285 --> 00:48:36,905
So that brings us to an
interesting situation.

811
00:48:37,015 --> 00:48:40,155
like you said, you want everybody
to have a chance to grow.

812
00:48:40,205 --> 00:48:43,455
You want to actively be growing the
team for everybody to get better.

813
00:48:44,265 --> 00:48:49,705
But you also want to hire people who
are better than you or smarter than

814
00:48:49,705 --> 00:48:52,327
you, who are more skilled than you.

815
00:48:52,327 --> 00:49:00,580
How do you, as a CTO, give those
people a chance to grow if they might

816
00:49:00,620 --> 00:49:02,410
be, already beyond what you know.

817
00:49:03,031 --> 00:49:06,831
from that perspective, I like
to elevate them above me.

818
00:49:07,181 --> 00:49:11,651
for example, if they're starting to
shine and I know, unless I leave,

819
00:49:12,291 --> 00:49:14,891
they've probably reached their ceiling
at this point, but, however, I want

820
00:49:14,891 --> 00:49:16,511
to make sure that they continue on.

821
00:49:16,511 --> 00:49:20,641
Instead of me presenting at the
board meeting, I'd probably invite

822
00:49:20,641 --> 00:49:22,831
them along and say, hey, I'm going
to bring you in for 10 minutes.

823
00:49:22,831 --> 00:49:26,896
I want you to present this area
and that's given them visibility.

824
00:49:26,896 --> 00:49:29,346
It's given them experience,
it's given them that

825
00:49:29,346 --> 00:49:32,226
responsibility to go at that area.

826
00:49:32,226 --> 00:49:34,866
I'm also a big believer of
celebrating when somebody leaves.

827
00:49:35,776 --> 00:49:40,546
I've had people in my team that I
hated to lose, but I'm so proud of

828
00:49:40,546 --> 00:49:41,956
the things that they bounced into.

829
00:49:42,536 --> 00:49:46,006
Big titles that they went into,
because we were part of their journey.

830
00:49:46,606 --> 00:49:49,306
We were part of their growth,
part of their learning, part

831
00:49:49,306 --> 00:49:50,886
of their desire to get there.

832
00:49:51,526 --> 00:49:57,556
I never would send out an email
to say, 'George has left, we never

833
00:49:57,556 --> 00:50:00,706
liked him, he was a pain in the ass
and he always kept breaking stuff'.

834
00:50:00,706 --> 00:50:04,916
I want to celebrate the fact that George
has been with us for the past two years.

835
00:50:04,976 --> 00:50:07,866
He's done a phenomenal amount of
stuff and we're really excited

836
00:50:08,046 --> 00:50:09,526
at what he's going towards next.

837
00:50:09,556 --> 00:50:13,856
And he'll always have a home for him back
here if he ever wants to come and see us.

838
00:50:14,506 --> 00:50:18,926
I want to make sure that people know
that, and this is something that Reid

839
00:50:18,926 --> 00:50:22,846
Hoffman of LinkedIn said, it's one of
the questions he asks in the interview,

840
00:50:23,546 --> 00:50:24,716
"when are you going to be leaving us?"

841
00:50:24,816 --> 00:50:28,186
because you're not joining
a company for life anymore.

842
00:50:29,396 --> 00:50:33,766
So let's pretend that you are going
to be leaving within two years.

843
00:50:34,396 --> 00:50:34,966
Great.

844
00:50:35,956 --> 00:50:40,416
what point in this career are you going
to say I've outgrown this company?

845
00:50:40,936 --> 00:50:42,416
Tell me about your future plans.

846
00:50:42,416 --> 00:50:46,136
Cause I want to get you there,
but I'm going to utilize you and

847
00:50:46,136 --> 00:50:50,216
we're going to work together for
the few years that I've got of you

848
00:50:50,286 --> 00:50:51,196
and then you're going to move on.

849
00:50:52,191 --> 00:50:53,321
And that's a celebration.

850
00:50:53,321 --> 00:50:57,381
Never see it as a deign against
loyalty or somebody turned on you.

851
00:50:57,391 --> 00:50:57,591
no.

852
00:50:58,031 --> 00:51:02,391
And that's part of the emotional
growth that a CTO has to go through.

853
00:51:03,231 --> 00:51:07,531
Just because your lead engineer has left
you, have they left you for good reasons?

854
00:51:08,641 --> 00:51:13,171
And if it's good reasons, then pat
them on the back, wish them the best

855
00:51:13,891 --> 00:51:16,711
and look forward to listening in on
their journey as they continue forward.

856
00:51:18,001 --> 00:51:21,431
I think the world would be a much better
place if everybody thought that way.

857
00:51:23,451 --> 00:51:24,151
It is hard.

858
00:51:24,151 --> 00:51:27,611
Again, these damn emotions
that these humans have really

859
00:51:27,611 --> 00:51:28,721
do get in the way at times.

860
00:51:29,706 --> 00:51:30,496
They really do.

861
00:51:31,626 --> 00:51:34,606
What about the non-human element?

862
00:51:34,796 --> 00:51:38,496
So far we've focused on that,
and I think that's probably

863
00:51:38,546 --> 00:51:40,296
the hardest bit, like you said.

864
00:51:40,396 --> 00:51:46,376
but there's gonna be some other things
that we should be focusing on as well.

865
00:51:46,916 --> 00:51:50,226
with the technical being
a hint in the name.

866
00:51:50,926 --> 00:51:52,686
What about the tech decisions?

867
00:51:52,746 --> 00:51:56,896
what would you say would be some
of the most crucial elements and

868
00:51:56,916 --> 00:52:00,636
the biggest decisions that every
CTO is going to have to face?

869
00:52:01,366 --> 00:52:02,496
so it's always hard.

870
00:52:02,686 --> 00:52:06,416
particularly if we go to some more of the
engineering side, deciding, for example,

871
00:52:07,226 --> 00:52:10,566
which language you're going to use, or
which framework you're going to use, or

872
00:52:10,616 --> 00:52:14,136
which cloud provider you're going to use,
or which database you're going to use.

873
00:52:16,206 --> 00:52:19,836
We, by our nature, love new shiny things.

874
00:52:20,701 --> 00:52:21,381
We love

875
00:52:21,541 --> 00:52:24,431
to test and play and use
the latest and greatest.

876
00:52:24,431 --> 00:52:29,341
That's the hardest part of this
role, is saying, ' I can't be

877
00:52:29,341 --> 00:52:30,621
on the cutting edge anymore'.

878
00:52:30,621 --> 00:52:33,711
I can be on the leading edge,
but the cutting and the bleeding

879
00:52:33,711 --> 00:52:36,261
edge, I can't do that anymore.

880
00:52:36,406 --> 00:52:39,576
a new language that may have
popped up that seems to answer

881
00:52:39,576 --> 00:52:40,936
everything that you've wanted.

882
00:52:42,236 --> 00:52:45,636
If a CTO was to say, now we're moving to
this language and everything's going to

883
00:52:45,636 --> 00:52:48,126
be ported to this, huge warning signal.

884
00:52:48,656 --> 00:52:50,686
Because that's a vanity project.

885
00:52:50,686 --> 00:52:52,776
You're not thinking in the
best interest of the company.

886
00:52:53,246 --> 00:52:57,316
Because that language, or tool,
whatever it is, hasn't proven itself.

887
00:52:57,936 --> 00:53:00,866
We don't know, can we
recruit engineers for it?

888
00:53:00,866 --> 00:53:02,136
Do we have a support system?

889
00:53:02,156 --> 00:53:03,356
Is there an ecosystem?

890
00:53:03,636 --> 00:53:07,226
Is there a business system
around this to figure out?

891
00:53:08,086 --> 00:53:14,086
Yes, it's not sexy that I have to choose
SQL Server, but I know that I can throw

892
00:53:14,086 --> 00:53:18,606
a stone and I'm going to get a SQL
Server expert, a company, a support

893
00:53:18,606 --> 00:53:21,186
person, a backup, no matter where.

894
00:53:21,186 --> 00:53:24,936
MySQL Postgres have now earned
those rights as well, but there

895
00:53:24,936 --> 00:53:30,796
was a time that choosing MySQL
Was a 'whoo that's a big risk.

896
00:53:30,966 --> 00:53:34,356
Okay, who's really supporting who's
really doing' and until Oracle bought

897
00:53:34,376 --> 00:53:40,046
them, they then got their sort of
big boy validation at that point.

898
00:53:40,746 --> 00:53:46,096
So it's very tempting to chase the latest
and greatest, or the greatest buzzwords.

899
00:53:46,186 --> 00:53:50,906
And where we usually see this the
most is in the JavaScript frameworks.

900
00:53:51,716 --> 00:53:55,806
Holy crap, that's like friggin
Fashion, they go in and out of

901
00:53:55,806 --> 00:54:00,226
fashion, left, right and centre,
and it's really hard, to say it.

902
00:54:00,356 --> 00:54:04,086
And I bumped against a wonderful
Instagram reel the other day, and it

903
00:54:04,086 --> 00:54:07,916
was one of these sort of, Say the one
thing out loud that nobody wants to say.

904
00:54:08,376 --> 00:54:13,526
and it was one of the, one of the
software engineers said, "JQuery is okay.

905
00:54:14,316 --> 00:54:15,576
It'll always get the job done".

906
00:54:16,606 --> 00:54:17,456
Okay, I get you.

907
00:54:17,696 --> 00:54:21,326
Anyway, so from a technology
decision, you're not making technology

908
00:54:21,326 --> 00:54:23,086
decisions based on what you like.

909
00:54:23,096 --> 00:54:28,056
You're building it based on, can
the business build upon this?

910
00:54:28,456 --> 00:54:30,346
Can the business make money off of this?

911
00:54:30,666 --> 00:54:32,906
Can the business pay all
our salaries off of this?

912
00:54:33,026 --> 00:54:36,351
And in five years' time,
how's this going to look.

913
00:54:37,001 --> 00:54:39,591
In 10 years' time, how's
this going to look?

914
00:54:40,141 --> 00:54:41,181
Because here's the thing.

915
00:54:41,181 --> 00:54:44,051
While software technically
doesn't age, it does.

916
00:54:44,051 --> 00:54:46,321
But the years go by very quickly.

917
00:54:46,976 --> 00:54:49,986
So you have to keep pace with that.

918
00:54:50,516 --> 00:54:57,476
And I generally don't like to support
any big open source projects unless

919
00:54:57,476 --> 00:55:02,306
they've got a big commercial backer
behind them, because I want to know

920
00:55:02,686 --> 00:55:08,596
that somebody else is investing in the
health of this framework, project, etc.

921
00:55:08,936 --> 00:55:12,726
And it's not just us, because I'm
a smaller company that I'm not

922
00:55:12,726 --> 00:55:13,776
going to be able to influence it.

923
00:55:14,036 --> 00:55:17,336
And while it's cool to be able
to say, 'yeah, but it's open

924
00:55:17,336 --> 00:55:18,726
source, we could support it'.

925
00:55:19,476 --> 00:55:21,536
No, we've got our own day jobs.

926
00:55:22,256 --> 00:55:24,496
We don't have time to be able
to support that framework.

927
00:55:25,521 --> 00:55:29,501
And nor do we have the knowledge
of the framework at that level to

928
00:55:29,501 --> 00:55:30,791
be able to build and support it.

929
00:55:30,791 --> 00:55:33,481
So yeah, that's a placebo.

930
00:55:33,651 --> 00:55:34,961
Do people say it's open source?

931
00:55:35,981 --> 00:55:39,241
For me, the open source banner,
and I'm a huge open source fan,

932
00:55:39,561 --> 00:55:43,961
is that we collectively contribute
to the success of that project.

933
00:55:44,831 --> 00:55:49,661
But it still needs a big sponsor before
I'm going to bet the livelihoods of

934
00:55:49,681 --> 00:55:51,731
everybody working in my company on it.

935
00:55:52,571 --> 00:55:56,935
I guess another facet  to the open
source equation is that it helps you

936
00:55:56,935 --> 00:56:03,571
avoid lock-in, which is something that's
very appealing a lot of the time, right?

937
00:56:04,087 --> 00:56:04,557
for sure.

938
00:56:04,907 --> 00:56:09,187
And To be able to freeze frame, there's
been plenty of examples in the past

939
00:56:09,187 --> 00:56:13,597
of where a particular open source
has now moved over to GPL or moved

940
00:56:13,597 --> 00:56:17,887
over to commercial or just stopped
it and said, okay, I'll snapshot the

941
00:56:17,887 --> 00:56:22,057
previous version that we were using,
which was a license-friendly and we

942
00:56:22,057 --> 00:56:25,927
will run with this ourselves until
such times we find an alternative.

943
00:56:26,397 --> 00:56:30,167
So it gives you time as opposed
to a commercial company saying.

944
00:56:30,942 --> 00:56:32,942
We're out of business next week.

945
00:56:33,422 --> 00:56:33,992
What are we doing now?

946
00:56:34,262 --> 00:56:34,832
Scramble.

947
00:56:35,402 --> 00:56:39,442
because it is a core part of the
architecture that it's there.

948
00:56:39,942 --> 00:56:45,802
but again, as an architect, as a CTO,
I always like to build in, very much a

949
00:56:45,802 --> 00:56:51,552
microservices API facade type patterns,
whereby if a core component does

950
00:56:51,572 --> 00:56:57,542
disappear, I'm not writing to that core
components APIs per se, I'm writing to

951
00:56:57,542 --> 00:57:01,522
facades API, so I can then go and, Insert
another one, and where this is happening

952
00:57:01,522 --> 00:57:06,412
a lot, Miko, is AI companies are failing
as quickly as they're being created.

953
00:57:07,082 --> 00:57:10,812
So whether that's an AI company that's
doing OCR for something or audio

954
00:57:10,872 --> 00:57:15,752
transcription, whatever it is, when you're
building and utilizing that service,

955
00:57:15,752 --> 00:57:18,302
by all means do it, but put a layer in.

956
00:57:18,302 --> 00:57:21,272
that allows you to switch out
that vendor should that their

957
00:57:21,272 --> 00:57:24,972
pricing go way up because they're
not made able to make money.

958
00:57:24,972 --> 00:57:28,162
So they have to double their prices
or they've gone out of business

959
00:57:28,162 --> 00:57:29,192
because they can't compete.

960
00:57:29,572 --> 00:57:31,512
Then I have to make switch
in another provider.

961
00:57:32,002 --> 00:57:37,212
So a CTO is always keeping their
eye on the business consequences

962
00:57:37,212 --> 00:57:41,072
of that going away and ensuring
the architects, if they're not the

963
00:57:41,072 --> 00:57:45,102
architect is giving us an insurance
policy that I don't have to go to the

964
00:57:45,102 --> 00:57:49,097
board and say, Sorry, company's down'.

965
00:57:49,197 --> 00:57:49,637
Why?

966
00:57:50,257 --> 00:57:52,537
this company that we thought we were
going to be hanging around has just

967
00:57:52,547 --> 00:57:53,687
gone out of business and we're humped.

968
00:57:54,841 --> 00:57:58,055
I think the AI is something that
a lot of people have realized now.

969
00:57:58,101 --> 00:58:03,228
It's moving so quickly that literally
you might start using a tool today

970
00:58:03,255 --> 00:58:07,055
and a couple of weeks later, it's,
'Oh, wait, the website bounces'.

971
00:58:07,101 --> 00:58:07,710
What's happening.

972
00:58:08,357 --> 00:58:09,607
It's happening a lot, isn't it?

973
00:58:09,657 --> 00:58:11,457
It's frightening.

974
00:58:11,507 --> 00:58:13,787
I think a lot of these,
they're cool ideas.

975
00:58:13,787 --> 00:58:16,297
They're more features as
opposed to businesses.

976
00:58:16,830 --> 00:58:20,800
But like you said, slapping AI on
your pitch deck, definitely helps at

977
00:58:20,820 --> 00:58:23,370
the moment, at least for some people,

978
00:58:24,017 --> 00:58:25,837
It opens the door to the
conversation at least.

979
00:58:26,243 --> 00:58:26,789
exactly.

980
00:58:27,489 --> 00:58:32,816
from your experience, what are the
best patterns for arriving at the right

981
00:58:32,825 --> 00:58:39,525
decisions, because I know some people are
more, I don't want to say totalitarian

982
00:58:39,543 --> 00:58:41,625
in how they, Bring this decisions.

983
00:58:41,670 --> 00:58:46,043
but it's easy enough to picture a
CTO who just does their analysis.

984
00:58:46,043 --> 00:58:48,661
And then all of a sudden he says,
'Oh, we're doing this' and expects

985
00:58:48,661 --> 00:58:50,252
everybody could go through with that.

986
00:58:51,007 --> 00:58:53,934
I've also seen people who are
much more humble about that.

987
00:58:53,952 --> 00:58:58,079
And they do things like drafting a
thing to gather people's opinions,

988
00:58:58,089 --> 00:59:02,798
or maybe write a white paper or some
kind of RFC request for comments

989
00:59:02,798 --> 00:59:03,916
to gather people's opinions.

990
00:59:03,916 --> 00:59:08,284
It's obviously not possible to get
everybody on the same page all the

991
00:59:08,293 --> 00:59:14,129
time, but what's your take on what
you've seen being successful the most

992
00:59:14,148 --> 00:59:18,748
In terms of arriving at decisions
that are not always straight cut.

993
00:59:18,802 --> 00:59:23,884
There's always some subjective, part to
choosing this database over that database

994
00:59:23,884 --> 00:59:25,884
or this provider versus that provider.

995
00:59:26,475 --> 00:59:27,355
You're absolutely right.

996
00:59:27,355 --> 00:59:32,695
And I think at some point, it always does
come down to a Coke versus Pepsi argument.

997
00:59:32,705 --> 00:59:35,655
It doesn't really matter which one
you're choosing, you're still getting

998
00:59:35,875 --> 00:59:38,145
that Coca Cola flavor, if you will.

999
00:59:39,165 --> 00:59:42,945
My personal way of doing
stuff is, I love wikis.

1000
00:59:43,385 --> 00:59:44,655
I love to show my workings.

1001
00:59:45,215 --> 00:59:47,855
And I love when I'm putting
out a new idea, a thought,

1002
00:59:48,020 --> 00:59:49,410
architecture, whatever it is.

1003
00:59:50,240 --> 00:59:52,990
I will actually put it out so
everybody can see it and review it.

1004
00:59:53,930 --> 00:59:56,160
and I invite criticism.

1005
00:59:56,260 --> 00:59:57,920
I invite holes in it.

1006
00:59:58,200 --> 01:00:03,630
Now, I have a very strong, and I would
recommend this for any CTOs, make

1007
01:00:03,630 --> 01:00:05,420
sure you have a strong right hand.

1008
01:00:05,420 --> 01:00:08,950
And that right hand is somebody
that is going to pull you aside and

1009
01:00:08,950 --> 01:00:10,270
say: ' what the hell are you thinking?'

1010
01:00:10,270 --> 01:00:11,380
What are you doing?

1011
01:00:11,380 --> 01:00:12,320
This is dumb.

1012
01:00:13,110 --> 01:00:19,180
You need that person that doesn't see
your hierarchy but will truly help you

1013
01:00:19,250 --> 01:00:23,650
think through stuff and is not going
to scare to say to you, you could have

1014
01:00:23,690 --> 01:00:25,000
handled that better, couldn't you?

1015
01:00:25,000 --> 01:00:27,370
Okay, or you handled that very well.

1016
01:00:27,370 --> 01:00:28,050
Do that again.

1017
01:00:29,130 --> 01:00:35,040
o that right hand is always a strong,
asset to any CTO, and from that

1018
01:00:35,040 --> 01:00:39,510
perspective, they're your first line
of defense in terms of any idea, any

1019
01:00:39,510 --> 01:00:40,930
strategy, any thought you're doing.

1020
01:00:41,920 --> 01:00:43,620
It's to, to throw it off of them.

1021
01:00:44,270 --> 01:00:45,280
What's their thoughts?

1022
01:00:45,560 --> 01:00:46,190
What do they get?

1023
01:00:46,470 --> 01:00:50,050
Because a right hand should never
be scared to go against you.

1024
01:00:50,050 --> 01:00:52,560
You should never be scared of
the consequences of completely

1025
01:00:52,560 --> 01:00:53,400
disagreeing with you.

1026
01:00:53,980 --> 01:00:57,320
Because it's through disagreement that
we figure out, I missed something.

1027
01:00:57,780 --> 01:00:59,020
Oh, you're right.

1028
01:00:59,040 --> 01:01:00,160
I completely missed that.

1029
01:01:00,560 --> 01:01:05,770
Or, this gets us 80% of the way,
but the last 20% is going to be

1030
01:01:05,770 --> 01:01:09,570
a real pain in the ass because
we've chosen wrongly at the start.

1031
01:01:09,580 --> 01:01:10,660
That type of stuff.

1032
01:01:10,990 --> 01:01:15,070
I liken it to the, If NASA leaves
the Earth atmosphere by half a degree

1033
01:01:15,070 --> 01:01:18,010
out, they're going to miss the moon
by hundreds of thousands of miles.

1034
01:01:18,410 --> 01:01:21,390
The small decisions at the start
can really make a big impact.

1035
01:01:22,140 --> 01:01:27,120
they're your first wave, and then again,
whether it's white papers, et cetera,

1036
01:01:27,420 --> 01:01:29,080
I always like to show my workings.

1037
01:01:29,080 --> 01:01:32,420
And to make sure that if we're deciding
we're going to use this software

1038
01:01:32,420 --> 01:01:35,880
or this platform, I'm going to show
them the three alternatives that we

1039
01:01:35,880 --> 01:01:39,400
debated, the feature sets, why we
didn't, and why we've chosen this one.

1040
01:01:39,410 --> 01:01:41,380
And it could be a mixture
of lots of things.

1041
01:01:42,050 --> 01:01:46,420
Favorable licensing, feature
sets, availability of up and

1042
01:01:46,420 --> 01:01:49,020
coming features that we know we're
going to get, whatever it is.

1043
01:01:49,020 --> 01:01:54,900
Make the argument be data-led,
not emotion-led, and make it

1044
01:01:54,900 --> 01:01:58,180
feel like there is a voice for
others to come in and collaborate.

1045
01:01:59,230 --> 01:02:01,690
But ultimately, you're the decision maker.

1046
01:02:01,840 --> 01:02:04,600
You're the one that's going
to be responsible for deciding

1047
01:02:05,050 --> 01:02:09,020
Postgres versus MySQL, to put it
down into a real simple example.

1048
01:02:09,437 --> 01:02:15,365
The Coke versus Pepsi also made me think
about some of the advice I heard elsewhere

1049
01:02:15,546 --> 01:02:22,555
about how when you're in a situation where
you can choose a path where the decision

1050
01:02:22,555 --> 01:02:26,874
is effectively reversible, there's really
no point spending too much time on that.

1051
01:02:26,910 --> 01:02:31,846
You can almost flip a coin between
Coke and Pepsi and then you try to

1052
01:02:31,946 --> 01:02:35,365
save your time for the ones where
you can't reverse them easily.

1053
01:02:35,919 --> 01:02:36,737
Is that good advice?

1054
01:02:36,801 --> 01:02:39,251
That is brilliant, absolutely brilliant.

1055
01:02:39,686 --> 01:02:43,266
sometimes it's hard to determine
which ones are  the big decisions

1056
01:02:43,266 --> 01:02:44,576
in which ones are the smaller ones.

1057
01:02:44,576 --> 01:02:48,446
Fortunately, in today's environment,
even from a language perspective,

1058
01:02:48,446 --> 01:02:51,716
like a perfect example is,
I'm a huge serverless fan.

1059
01:02:52,476 --> 01:02:55,556
I find that the vast majority of
companies that we interact with don't

1060
01:02:55,566 --> 01:03:01,136
need large servers, they can manage
with serverless and serverless has truly

1061
01:03:01,146 --> 01:03:05,026
removed the language decision away.

1062
01:03:05,606 --> 01:03:09,046
So we can write a handful
of endpoints in Java.

1063
01:03:09,596 --> 01:03:13,146
Or a handful of endpoints
in Go or Python or Node.

1064
01:03:14,046 --> 01:03:19,196
I don't have to be a single
or a couple of language shop.

1065
01:03:19,636 --> 01:03:22,836
I can be a polyglot environment using
the right tool for the right job.

1066
01:03:22,836 --> 01:03:26,256
So the consequences of me
saying, ' I'm going to write

1067
01:03:26,256 --> 01:03:27,626
these four endpoints in Java'.

1068
01:03:27,626 --> 01:03:28,716
Six months down the line.

1069
01:03:29,036 --> 01:03:30,536
Damn, I've got no more Java developers.

1070
01:03:31,176 --> 01:03:31,466
Okay.

1071
01:03:31,746 --> 01:03:33,886
Rewrite those four
endpoints in Node, please.

1072
01:03:33,926 --> 01:03:35,666
Because we've got a
team of Node developers.

1073
01:03:35,666 --> 01:03:39,606
That's a much easier thing to do than
saying, 'we have to rewrite this whole

1074
01:03:39,606 --> 01:03:43,346
architecture now because we've decided now
we're going to use a different language'.

1075
01:03:43,346 --> 01:03:47,531
So from that perspective, There
are certain safeguards that we can

1076
01:03:47,531 --> 01:03:53,011
put in, or insurance policies, to
make it easier to unwind without

1077
01:03:53,051 --> 01:03:54,451
the business feeling the effect.

1078
01:03:54,451 --> 01:03:55,442
that's solid advice.

1079
01:03:56,478 --> 01:04:00,305
Let's cover one more thing that I think
a lot of people and teams struggle

1080
01:04:00,305 --> 01:04:02,360
with, and that's documentation.

1081
01:04:02,405 --> 01:04:06,796
What advice would you give to
people who may have never seen

1082
01:04:07,005 --> 01:04:11,460
a team that actually has solid
docs or enjoys writing this docs?

1083
01:04:11,760 --> 01:04:13,787
How do you make that better?

1084
01:04:13,837 --> 01:04:17,477
Nobody likes to document, because they
always feel like it's holding them back

1085
01:04:17,527 --> 01:04:20,957
and you always get that engine that
says, 'But my code's obvious enough.

1086
01:04:20,957 --> 01:04:22,197
It doesn't need documentation.

1087
01:04:22,197 --> 01:04:22,977
Anybody can read it'.

1088
01:04:23,487 --> 01:04:25,267
And to a large extent, they're right.

1089
01:04:25,277 --> 01:04:25,757
for sure.

1090
01:04:26,287 --> 01:04:31,107
my sort of, advice on that perspective
is once your team gets to a significant

1091
01:04:31,107 --> 01:04:33,447
size, hire a technical writer.

1092
01:04:33,957 --> 01:04:36,977
They're one of the best
investments you could ever make.

1093
01:04:37,637 --> 01:04:42,137
they will, produce the documentation to
a level that's required for the business.

1094
01:04:42,387 --> 01:04:47,157
And when we need business, we're talking
about continuity, which is when we

1095
01:04:47,517 --> 01:04:53,717
change somebody in the team or we move
something, do we still have knowledge

1096
01:04:53,977 --> 01:04:55,757
of how that particular system worked?

1097
01:04:56,647 --> 01:04:59,687
And it doesn't usually have to be
that detailed from that perspective,

1098
01:04:59,727 --> 01:05:02,027
but you do need to have a trail.

1099
01:05:02,917 --> 01:05:07,857
The other thing that I would advise
is your head always goes to the

1100
01:05:07,857 --> 01:05:11,127
written word because that's just
the connections that we've made.

1101
01:05:11,127 --> 01:05:12,947
Video is documentation.

1102
01:05:12,947 --> 01:05:20,247
I'm more than happy for an engineer
to talk through a given area and just

1103
01:05:20,247 --> 01:05:22,237
record it and now that's available.

1104
01:05:23,052 --> 01:05:26,432
And given our modern AI tools, we
can transcribe that at a later date

1105
01:05:26,432 --> 01:05:27,532
in order to make it searchable.

1106
01:05:28,072 --> 01:05:30,932
that's a zero cost, transcription.

1107
01:05:30,982 --> 01:05:33,602
Particularly if you talk into Teams,
for example, they'll transcribe it

1108
01:05:33,602 --> 01:05:35,582
for you in real time, as will Google.

1109
01:05:35,642 --> 01:05:37,202
Boom, there, you've
done your documentation.

1110
01:05:37,302 --> 01:05:43,782
in order to make that a little bit easier,
and less interview-ish, is, at certain

1111
01:05:43,802 --> 01:05:48,842
points in the project, Get your team to
present what they've just done as part

1112
01:05:48,842 --> 01:05:54,162
of an educational video and let that Q&A
to happen back and forth and record that.

1113
01:05:54,162 --> 01:05:55,652
And now you've just documented it.

1114
01:05:55,652 --> 01:05:58,162
And once your team gets to a certain
size and your company gets to a

1115
01:05:58,162 --> 01:06:01,292
certain size and you need to be able
to have tighter documentation because

1116
01:06:01,292 --> 01:06:07,312
of compliance, because of, SOC 2
or ISO or whatever other compliant

1117
01:06:07,312 --> 01:06:08,812
framework you're trying to adhere to.

1118
01:06:09,122 --> 01:06:11,532
Let the technical writer
fill in the gaps for that.

1119
01:06:11,582 --> 01:06:16,942
But don't force everybody to crack
open Word and create a Word document.

1120
01:06:17,252 --> 01:06:19,812
Or create confluence pages, etc.

1121
01:06:20,472 --> 01:06:26,212
Documentation can be inherently
sourced through Jira bug tickets.

1122
01:06:26,212 --> 01:06:31,382
through, comments in tickets,
commit messages, video talks.

1123
01:06:31,492 --> 01:06:35,522
It also contributes to the overall
knowledge base of the company.

1124
01:06:35,522 --> 01:06:40,572
And my best advice would be to run
up a wiki in order to have that

1125
01:06:40,572 --> 01:06:45,612
wiki to be the glue between here's
the link to the GitLab repository.

1126
01:06:45,627 --> 01:06:46,887
Here's the Confluence one.

1127
01:06:46,887 --> 01:06:49,147
Here's to the SharePoint where
the videos were recorded.

1128
01:06:49,147 --> 01:06:50,027
Here's to the stuff.

1129
01:06:50,037 --> 01:06:52,647
and have that wiki be that
little bit of informal.

1130
01:06:52,647 --> 01:06:56,737
The conductor and the orchestra to
know where all the islands of data are.

1131
01:06:57,467 --> 01:07:01,707
Untl such times that you can dedicate
a role to that, to bring that data

1132
01:07:01,707 --> 01:07:03,577
together in a more formal manner.

1133
01:07:04,107 --> 01:07:07,967
Most of the time it's not
required in a formal package.

1134
01:07:08,207 --> 01:07:13,357
Gone are the days where I'm looking from
a 60 page PDF of the product manual.

1135
01:07:13,682 --> 01:07:18,546
I think the one thing I would add to that
is that, I've gotten a lot of mileage

1136
01:07:18,914 --> 01:07:24,978
from just setting one simple rule for
my teams that every major piece of,

1137
01:07:25,023 --> 01:07:30,114
functionality or of work really that
we do, that requires people to talk to

1138
01:07:30,114 --> 01:07:33,069
each other to see what it's actually
going to look like, how it's going to

1139
01:07:33,096 --> 01:07:38,205
get implemented needs to result in a
small design doc that basically says,

1140
01:07:38,241 --> 01:07:39,732
'okay, this is what we set out to do'.

1141
01:07:40,232 --> 01:07:42,069
This is roughly how we expect it to work.

1142
01:07:42,096 --> 01:07:43,523
This is what we're going to do.

1143
01:07:43,605 --> 01:07:47,114
Just having that written down
instead of having that in Slack

1144
01:07:47,150 --> 01:07:49,306
messages and Jira tickets And stuff.

1145
01:07:49,796 --> 01:07:50,456
Completely.

1146
01:07:50,766 --> 01:07:55,636
Even if it's part of a wiki, like
you said, tends to work really well.

1147
01:07:55,706 --> 01:07:59,516
It gets outdated, obviously, when
you've got newer versions, but

1148
01:07:59,516 --> 01:08:02,816
at least gives you a snapshot in
time of, this is why we did that.

1149
01:08:02,866 --> 01:08:03,406
Oh, wow.

1150
01:08:03,446 --> 01:08:03,846
Okay.

1151
01:08:05,356 --> 01:08:05,866
yes.

1152
01:08:05,866 --> 01:08:09,166
and having somebody stand in front
of a whiteboard, draw a picture, and

1153
01:08:09,171 --> 01:08:10,716
then somebody take a picture of it.

1154
01:08:11,226 --> 01:08:12,187
That's documentation too.

1155
01:08:12,262 --> 01:08:12,716
little time.

1156
01:08:13,551 --> 01:08:17,731
It doesn't have to be somebody going back
into draw.io to recreate that whiteboard.

1157
01:08:17,821 --> 01:08:18,201
Why?

1158
01:08:18,491 --> 01:08:19,621
it's doing well as it is.

1159
01:08:19,631 --> 01:08:20,681
Just leave it as an image.

1160
01:08:21,031 --> 01:08:23,831
Make sure it's indexed though so we
know how to find it and what we're

1161
01:08:23,831 --> 01:08:25,371
about to do when we click on that image.

1162
01:08:26,526 --> 01:08:27,166
Absolutely.

1163
01:08:27,216 --> 01:08:32,616
Some of the last chapters of your
book, talk about the company growth and

1164
01:08:32,646 --> 01:08:37,756
potential acquisition and how to make
it as, smooth as possible from again,

1165
01:08:37,766 --> 01:08:42,736
the perspective of the CTO, what would
be the most important message on that

1166
01:08:42,736 --> 01:08:44,716
front that you would, want to send

1167
01:08:44,736 --> 01:08:49,936
When you are in a very
growth-orientated company,

1168
01:08:50,076 --> 01:08:52,176
inevitably the company will evolve.

1169
01:08:52,216 --> 01:08:58,316
Now it'll either be acquired, It will
acquire other companies, it will go

1170
01:08:58,316 --> 01:09:00,656
through a growth phase, of course it will.

1171
01:09:01,366 --> 01:09:04,786
And what you're doing there is making
sure that all the decisions that you're

1172
01:09:04,786 --> 01:09:08,246
making are going to outlive your tenure.

1173
01:09:09,106 --> 01:09:12,846
so any decision that you make, you've
got to make sure that the person

1174
01:09:12,846 --> 01:09:16,196
coming behind you is going to feel
confident that, 'hey, you made the right

1175
01:09:16,206 --> 01:09:17,566
decision, I can take this on further'.

1176
01:09:17,566 --> 01:09:18,056
Okay.

1177
01:09:18,056 --> 01:09:24,036
And planning that far ahead takes,
takes discipline and it takes hardness.

1178
01:09:24,686 --> 01:09:28,816
for me, I've worked a lot in the
private equity space being a CTO

1179
01:09:28,816 --> 01:09:29,916
for a private equity company.

1180
01:09:29,916 --> 01:09:31,136
So what does that basically mean?

1181
01:09:31,516 --> 01:09:33,276
It means that, you're going to be sold.

1182
01:09:33,856 --> 01:09:36,666
At a given date, because that's how
private equity makes their money,

1183
01:09:37,606 --> 01:09:40,186
and that means you're going to be
sold to either another private equity

1184
01:09:40,186 --> 01:09:44,356
company or another company that's
going to consume you as part of that.

1185
01:09:45,216 --> 01:09:50,106
Now, what I love about being a CTO in
a private equity world is that at some

1186
01:09:50,106 --> 01:09:53,596
point, and I talk about this in the book
as well, is you're going to go through

1187
01:09:54,166 --> 01:09:55,826
a process that's called due diligence.

1188
01:09:56,636 --> 01:10:02,326
And that's when somebody comes
in and effectively, rates

1189
01:10:02,386 --> 01:10:03,286
everything that you've done.

1190
01:10:03,286 --> 01:10:07,066
And that's where I like to say, somebody's
gonna come in and say, Hey, which

1191
01:10:07,066 --> 01:10:08,606
asshole thought this was a good idea?

1192
01:10:09,326 --> 01:10:11,756
And you're gonna have to put your
hand up and say, I'm the asshole.

1193
01:10:12,386 --> 01:10:13,536
I thought it was a good idea.

1194
01:10:14,546 --> 01:10:19,936
So by having that level of,
oversight and review that you know

1195
01:10:19,936 --> 01:10:21,966
is coming, it keeps you more honest.

1196
01:10:21,966 --> 01:10:26,196
It makes sure that you make decisions
that are going to stand up to scrutiny.

1197
01:10:26,686 --> 01:10:30,306
Because if you're a CTO in a
private company, who's really

1198
01:10:30,306 --> 01:10:31,516
going to question your decisions?

1199
01:10:31,516 --> 01:10:34,146
CEO's not going to come in and
say, Hey, why did you choose

1200
01:10:34,146 --> 01:10:35,256
this database over this database?

1201
01:10:35,256 --> 01:10:36,466
They don't have the time for that.

1202
01:10:36,466 --> 01:10:38,006
They probably don't have
the knowledge to do that.

1203
01:10:38,006 --> 01:10:43,451
But during acquisition time, There will
be experts that will come in and they

1204
01:10:43,451 --> 01:10:46,421
will know the reasons why you should
have done one over the other, etc.

1205
01:10:47,301 --> 01:10:49,601
Or the pain points of
having one over the other.

1206
01:10:49,601 --> 01:10:52,831
And they're going to ask you
to justify that, and what

1207
01:10:52,841 --> 01:10:54,551
your plans were for that, etc.

1208
01:10:54,881 --> 01:11:00,271
And now you truly have to be standing up
and take stock of what you decided to do.

1209
01:11:00,911 --> 01:11:04,851
I like to make sure that when I'm
mentoring and teaching and helping other

1210
01:11:04,851 --> 01:11:08,046
CTOs that, Stop thinking about today.

1211
01:11:08,596 --> 01:11:11,376
Start thinking about tomorrow.

1212
01:11:11,496 --> 01:11:12,646
And how is this going to look?

1213
01:11:12,776 --> 01:11:18,916
Did you choose to say, for example,
go with, AWS's best cloud practices?

1214
01:11:19,506 --> 01:11:22,616
Or did you decide that you knew best
and you were going to go this way?

1215
01:11:23,206 --> 01:11:26,676
Which one of the two is going to be easier
defended if you're going to go this way?

1216
01:11:27,216 --> 01:11:29,966
Or which one do you think is
going to be easily google-able?

1217
01:11:30,771 --> 01:11:32,951
For a new engineer coming in
trying to figure out how to

1218
01:11:32,951 --> 01:11:35,171
help manage your ecosystem.

1219
01:11:35,591 --> 01:11:37,551
don't always go your own path.

1220
01:11:37,651 --> 01:11:41,061
Sometimes go a path where
everybody else is trodden.

1221
01:11:41,811 --> 01:11:43,721
Because there's support for that.

1222
01:11:43,721 --> 01:11:49,661
That doesn't mean you can't innovate and
you can't change and tweak and evolve.

1223
01:11:49,661 --> 01:11:51,361
But you're always doing everything.

1224
01:11:51,871 --> 01:11:55,451
For the behalf of the company,
you do not want to be the reason

1225
01:11:55,451 --> 01:11:56,831
that salaries cannot be paid.

1226
01:11:57,881 --> 01:11:58,471
for sure.

1227
01:12:01,576 --> 01:12:05,956
This has been an absolute pleasure,
I think I personally learned quite a

1228
01:12:05,956 --> 01:12:11,656
bit, and for anybody who wants to learn
more about Alan's book, it's called

1229
01:12:11,696 --> 01:12:15,796
"Think Like a CTO", it's published by
Manning, and it's available right now.

1230
01:12:15,796 --> 01:12:20,106
Alan, thank you so much, and hopefully
see you next time for your next

1231
01:12:20,106 --> 01:12:22,506
book, and thank you for sharing.

1232
01:12:22,944 --> 01:12:24,124
Miko, it was an absolute pleasure.

1233
01:12:24,124 --> 01:12:24,664
Thank you, sir.

1234
01:12:24,926 --> 01:12:25,356
Thank you.