WEBVTT

1
00:00:00.080 --> 00:00:02.520
<v Speaker 1>It's a It's a really common challenge, isn't it. You

2
00:00:02.560 --> 00:00:05.799
<v Speaker 1>know your organization needs to leverage data, I mean properly

3
00:00:05.919 --> 00:00:08.720
<v Speaker 1>leverage it, but just figuring out where to start. How

4
00:00:08.720 --> 00:00:12.320
<v Speaker 1>do you even build that capability? It can feel well,

5
00:00:12.400 --> 00:00:13.919
<v Speaker 1>completely paralyzing sometimes.

6
00:00:14.080 --> 00:00:17.199
<v Speaker 2>Yeah, like staring at this huge ocean. You know there's

7
00:00:17.199 --> 00:00:21.440
<v Speaker 2>something valuable out there, but the map seems missing.

8
00:00:21.280 --> 00:00:25.559
<v Speaker 1>Exactly that feeling of being overwhelmed. The potential is massive,

9
00:00:25.600 --> 00:00:28.760
<v Speaker 1>we all see that, but the actual path yeah, not

10
00:00:28.839 --> 00:00:29.440
<v Speaker 1>always clear.

11
00:00:29.519 --> 00:00:32.280
<v Speaker 2>It's understandable. There's so much noise, so many options, it's

12
00:00:32.320 --> 00:00:34.119
<v Speaker 2>easy to get, you know, bogged down.

13
00:00:34.280 --> 00:00:36.759
<v Speaker 1>And that's exactly what we're diving deep into today. We

14
00:00:36.840 --> 00:00:39.200
<v Speaker 1>actually had a request from one of you asking for

15
00:00:39.240 --> 00:00:42.719
<v Speaker 1>a practical guide. Ah yeah, someone looking to strategically build

16
00:00:43.039 --> 00:00:47.280
<v Speaker 1>data science and analytics functions, but crucially without needing this

17
00:00:47.520 --> 00:00:50.560
<v Speaker 1>like huge upfront investment or getting lost in planning forever.

18
00:00:50.759 --> 00:00:51.640
<v Speaker 2>Right, makes sense.

19
00:00:52.200 --> 00:00:54.799
<v Speaker 1>So we're going to unpack the key insights from data

20
00:00:54.799 --> 00:00:58.439
<v Speaker 1>science and analytics strategy an emergent design approach. It's by

21
00:00:58.520 --> 00:00:59.840
<v Speaker 1>Kayla Shawadi.

22
00:00:59.439 --> 00:01:03.600
<v Speaker 2>And Alex good Choice. It offers a really useful perspective.

23
00:01:04.239 --> 00:01:07.040
<v Speaker 2>Our goal here is basically to pull out the actionable

24
00:01:07.079 --> 00:01:10.319
<v Speaker 2>advice from the book give you a clearer, maybe step

25
00:01:10.319 --> 00:01:13.359
<v Speaker 2>by step feel for how you can develop these capabilities.

26
00:01:13.920 --> 00:01:16.400
<v Speaker 2>And the key thing to remember really is that this

27
00:01:16.439 --> 00:01:19.239
<v Speaker 2>isn't about reaching some perfect end state overnight.

28
00:01:19.480 --> 00:01:20.280
<v Speaker 1>No, definitely not.

29
00:01:20.439 --> 00:01:22.879
<v Speaker 2>It's much more of a journey, you know, learning, adapting,

30
00:01:23.040 --> 00:01:24.120
<v Speaker 2>growing as you go.

31
00:01:24.319 --> 00:01:26.640
<v Speaker 1>Okay, so let's get into it. The book kicks off

32
00:01:26.680 --> 00:01:31.920
<v Speaker 1>by looking at why those traditional, very rigid data strategies

33
00:01:32.159 --> 00:01:33.599
<v Speaker 1>often just don't work out.

34
00:01:33.799 --> 00:01:35.359
<v Speaker 2>Yeah, they often fail to deliver.

35
00:01:35.519 --> 00:01:38.599
<v Speaker 1>It even points to that Harvard Business Review article is

36
00:01:38.680 --> 00:01:41.840
<v Speaker 1>data scientists still the sexiest job of the twenty first century?

37
00:01:41.920 --> 00:01:42.519
<v Speaker 1>Remember that one?

38
00:01:42.560 --> 00:01:46.000
<v Speaker 2>Oh? Yeah, that highlights some of the let's say, traps

39
00:01:46.040 --> 00:01:48.599
<v Speaker 2>for the unwary when setting up these functions.

40
00:01:48.719 --> 00:01:50.400
<v Speaker 1>So what are those main traps? What goes wrong with

41
00:01:50.439 --> 00:01:51.359
<v Speaker 1>the old school approach?

42
00:01:51.760 --> 00:01:55.120
<v Speaker 2>Well, one big one is rushing in hiring data scientists,

43
00:01:55.159 --> 00:01:58.439
<v Speaker 2>buying the flashy tech before really nailing down the business

44
00:01:58.480 --> 00:01:59.760
<v Speaker 2>problems they're supposed to solve.

45
00:02:00.200 --> 00:02:02.000
<v Speaker 1>This sexy java lure takes.

46
00:02:01.760 --> 00:02:05.519
<v Speaker 2>Over exactly, it distracts from the groundwork. And then there's

47
00:02:05.560 --> 00:02:10.120
<v Speaker 2>this great concept the book introduces, the corporate immune system.

48
00:02:10.319 --> 00:02:12.240
<v Speaker 1>The corporate immune system tell meymore.

49
00:02:12.360 --> 00:02:16.639
<v Speaker 2>It's that, you know, inherent organizational resistance to big changes,

50
00:02:16.680 --> 00:02:18.280
<v Speaker 2>like your body fighting off a cold.

51
00:02:18.360 --> 00:02:19.360
<v Speaker 1>Okay, I can picture that.

52
00:02:19.439 --> 00:02:24.199
<v Speaker 2>These big, top down, prescriptive strategies, they often trigger that

53
00:02:24.280 --> 00:02:28.800
<v Speaker 2>resistance because they feel disruptive, maybe even threatening, imposed from above.

54
00:02:29.000 --> 00:02:31.560
<v Speaker 1>Right. Yeah, we've probably all felt that pushback at some point.

55
00:02:32.159 --> 00:02:36.120
<v Speaker 1>So if that traditional, heavily planned way hits this wall,

56
00:02:36.719 --> 00:02:39.000
<v Speaker 1>what's the alternative? What does emergent design offer.

57
00:02:39.400 --> 00:02:43.280
<v Speaker 2>Emergent design is well, it's much more flexible, more responsive,

58
00:02:43.360 --> 00:02:47.080
<v Speaker 2>instead of these huge detailed plans that are probably outdated

59
00:02:47.120 --> 00:02:50.120
<v Speaker 2>the minute you print them. Sure, it focuses on setting

60
00:02:50.120 --> 00:02:52.840
<v Speaker 2>a clear direction and then learning and adapting as you go.

61
00:02:53.360 --> 00:02:56.840
<v Speaker 1>Iteratively, Well, that's about mapping every single step, more about

62
00:02:56.840 --> 00:02:59.400
<v Speaker 1>setting a course and then adjusting based on what you find.

63
00:02:59.439 --> 00:03:01.039
<v Speaker 2>Precisely, it's how much more evolutionary.

64
00:03:01.400 --> 00:03:04.560
<v Speaker 1>The book uses this really interesting analogy. It talks about

65
00:03:04.919 --> 00:03:09.280
<v Speaker 1>the strategist being more like a midwife rather than an expert.

66
00:03:09.759 --> 00:03:11.759
<v Speaker 1>What does that actually mean in practice?

67
00:03:12.000 --> 00:03:16.560
<v Speaker 2>It means, especially early on, your role is more about facilitation,

68
00:03:17.240 --> 00:03:20.520
<v Speaker 2>less about coming in saying I'm the expert, here are

69
00:03:20.560 --> 00:03:24.479
<v Speaker 2>the answers, and more about helping the business give birth

70
00:03:24.520 --> 00:03:27.919
<v Speaker 2>to its own solutions if you like. It's about engaging

71
00:03:27.919 --> 00:03:33.159
<v Speaker 2>with different departments, facilitating conversations to really understand their core challenges.

72
00:03:33.400 --> 00:03:35.719
<v Speaker 1>So it puts a big emphasis on problem finding.

73
00:03:35.639 --> 00:03:39.479
<v Speaker 2>Huge emphasis really digging deep to uncover what keeps people

74
00:03:39.599 --> 00:03:40.280
<v Speaker 2>up at night, and.

75
00:03:40.240 --> 00:03:43.080
<v Speaker 1>It actually suggests asking that specific question, right, what keeps

76
00:03:43.120 --> 00:03:47.080
<v Speaker 1>you up at night? Go ask sales operations HR yes, get.

77
00:03:46.879 --> 00:03:50.280
<v Speaker 2>Them to articulate their main pain points. Sandra Hogan, who's

78
00:03:50.319 --> 00:03:52.800
<v Speaker 2>quoted in the book, even suggests starting right at the top,

79
00:03:53.039 --> 00:03:53.439
<v Speaker 2>ask the.

80
00:03:53.400 --> 00:03:55.719
<v Speaker 1>CEO, I understand the being picture first.

81
00:03:55.520 --> 00:03:59.759
<v Speaker 2>Exactly, get the CEO's vision the main organizational challenges, and

82
00:03:59.759 --> 00:04:02.599
<v Speaker 2>then you can tailor those conversations with other leaders to

83
00:04:02.639 --> 00:04:04.120
<v Speaker 2>make sure everything aligns so.

84
00:04:04.080 --> 00:04:06.639
<v Speaker 1>You ensure the data stuff you do actually tackles the

85
00:04:06.639 --> 00:04:07.960
<v Speaker 1>most critical business needs.

86
00:04:08.360 --> 00:04:13.000
<v Speaker 2>Right and crucially, the book stresses showing value early tangible

87
00:04:13.039 --> 00:04:16.959
<v Speaker 2>results through what they call proofs of value, not just

88
00:04:17.000 --> 00:04:20.920
<v Speaker 2>proofs of concept, not just technical feasibility no Criicknapier makes

89
00:04:20.920 --> 00:04:23.040
<v Speaker 2>this point well in the book. You need to demonstrate

90
00:04:23.079 --> 00:04:26.959
<v Speaker 2>actual business benefit quick wins that show this approach works.

91
00:04:27.240 --> 00:04:30.000
<v Speaker 1>Okay, that makes a lot of sense. So you found

92
00:04:30.040 --> 00:04:34.319
<v Speaker 1>some key problems using this collaborative method. Now you need

93
00:04:34.519 --> 00:04:37.959
<v Speaker 1>the tools, the infrastructure, the team. The book talks about

94
00:04:37.959 --> 00:04:40.959
<v Speaker 1>building the data analytics stack. Can you break that down?

95
00:04:41.079 --> 00:04:42.480
<v Speaker 1>What are the basic pieces?

96
00:04:42.839 --> 00:04:45.360
<v Speaker 2>Sure, at a high level, you've got a few key stages.

97
00:04:45.560 --> 00:04:48.319
<v Speaker 2>First is data ingestion getting the data in from all

98
00:04:48.319 --> 00:04:51.360
<v Speaker 2>the different source systems okay. Then data storage keeping it

99
00:04:51.399 --> 00:04:54.879
<v Speaker 2>safe and accessible. Then data processing that's the cleaning transforming,

100
00:04:54.879 --> 00:04:56.480
<v Speaker 2>getting it ready for analysis.

101
00:04:56.000 --> 00:04:58.160
<v Speaker 1>A messy part, often often the messiest.

102
00:04:58.560 --> 00:05:02.000
<v Speaker 2>Then the analysis itself, finding the insights, answering the questions,

103
00:05:02.360 --> 00:05:05.000
<v Speaker 2>and finally data consumption how those insights get to the

104
00:05:05.000 --> 00:05:09.120
<v Speaker 2>people who need them, reports, dashboards, apps, whatever it is.

105
00:05:09.480 --> 00:05:12.720
<v Speaker 1>And the book really hammers home the incrementally part of

106
00:05:12.759 --> 00:05:14.639
<v Speaker 1>building this right. Start with what you've got.

107
00:05:14.800 --> 00:05:19.839
<v Speaker 2>Absolutely, leverage existing tools, existing infrastructure as much as you can. Initially,

108
00:05:20.560 --> 00:05:23.000
<v Speaker 2>don't feel you need the shiniest new thing right away,

109
00:05:23.639 --> 00:05:24.920
<v Speaker 2>so resist.

110
00:05:24.560 --> 00:05:26.639
<v Speaker 1>The urge to buy everything immediately.

111
00:05:26.759 --> 00:05:31.040
<v Speaker 2>Definitely, And a key principle is matching your data ingestion

112
00:05:31.120 --> 00:05:34.240
<v Speaker 2>frequency how often you pull data in with what you

113
00:05:34.399 --> 00:05:36.279
<v Speaker 2>actually need for analysis.

114
00:05:36.560 --> 00:05:39.600
<v Speaker 1>Meaning if you only report weekly, maybe you don't need

115
00:05:39.639 --> 00:05:42.160
<v Speaker 1>real time data streams for everything exactly.

116
00:05:42.279 --> 00:05:46.120
<v Speaker 2>Batch processing might be fine, but there's an important caveat.

117
00:05:46.120 --> 00:05:51.480
<v Speaker 2>The book mentions operational alerts think manufacturing IoT sensors.

118
00:05:51.040 --> 00:05:53.959
<v Speaker 1>Ah, right, where you need to know now if something's wrong.

119
00:05:54.120 --> 00:05:57.199
<v Speaker 2>Yes, for those critical operational systems, real time data is

120
00:05:57.240 --> 00:05:58.360
<v Speaker 2>obviously essential.

121
00:05:58.519 --> 00:06:00.639
<v Speaker 1>Got it. And then there's the team structure. Sure, the

122
00:06:00.680 --> 00:06:04.639
<v Speaker 1>book mentions BI developers, analysts, data scientists, data engineers, architects.

123
00:06:04.879 --> 00:06:06.600
<v Speaker 1>Sometimes those lines feel a bit fuzzy.

124
00:06:06.759 --> 00:06:10.000
<v Speaker 2>They really can be, especially in smaller teams or organizations

125
00:06:10.040 --> 00:06:13.279
<v Speaker 2>just starting out. But broadly, think of BI folks as

126
00:06:13.399 --> 00:06:14.639
<v Speaker 2>focused on the what.

127
00:06:14.759 --> 00:06:18.399
<v Speaker 1>Happened reporting dashboards right looking back.

128
00:06:18.720 --> 00:06:21.759
<v Speaker 2>Data scientists are more about the why and what might

129
00:06:21.800 --> 00:06:25.160
<v Speaker 2>happen next. They dig deeper, build predictive.

130
00:06:24.720 --> 00:06:27.720
<v Speaker 1>Models, forecasting, understanding drivers exactly.

131
00:06:28.040 --> 00:06:32.199
<v Speaker 2>And data engineers or architects they're the builders. They create

132
00:06:32.279 --> 00:06:36.160
<v Speaker 2>and maintain the underlying infrastructure, the data pipelines, making sure

133
00:06:36.199 --> 00:06:38.279
<v Speaker 2>everything is reliable and performs well.

134
00:06:38.399 --> 00:06:39.519
<v Speaker 1>The foundation, the foundation.

135
00:06:39.879 --> 00:06:42.560
<v Speaker 2>But yeah, the book acknowledges it's often blurry. People wear

136
00:06:42.639 --> 00:06:43.399
<v Speaker 2>multiple hats.

137
00:06:43.480 --> 00:06:47.079
<v Speaker 1>That clarification helps. And you mentioned predictive models. The book

138
00:06:47.160 --> 00:06:52.319
<v Speaker 1>touches on supervised versus unsupervised machine learning. Quick explanation for

139
00:06:52.439 --> 00:06:53.800
<v Speaker 1>US Dragon free.

140
00:06:53.600 --> 00:06:57.439
<v Speaker 2>If possible, Okay, let's try. Supervised learning is like like

141
00:06:57.560 --> 00:07:00.399
<v Speaker 2>learning with a teacher or flash cards. You have data

142
00:07:00.399 --> 00:07:02.000
<v Speaker 2>that's already labeled with the right.

143
00:07:01.920 --> 00:07:05.079
<v Speaker 1>Answer, like emails marked spam or not spam.

144
00:07:04.839 --> 00:07:08.120
<v Speaker 2>Perfect example. The algorithm learns from those labels to classify

145
00:07:08.199 --> 00:07:12.360
<v Speaker 2>new emails or regression. Another supervised type predicting a continuous

146
00:07:12.439 --> 00:07:15.879
<v Speaker 2>number like a house price or sales figures based on

147
00:07:15.959 --> 00:07:17.560
<v Speaker 2>labeled historical data.

148
00:07:17.600 --> 00:07:18.720
<v Speaker 1>So you know what you're aiming for.

149
00:07:19.079 --> 00:07:22.800
<v Speaker 2>Yes, you have a target. Unsupervised learning is more like exploring.

150
00:07:22.839 --> 00:07:25.920
<v Speaker 2>You don't have labels. The algorithm's job is to find

151
00:07:26.000 --> 00:07:29.480
<v Speaker 2>hidden patterns or structures in the data itself.

152
00:07:29.240 --> 00:07:32.839
<v Speaker 1>Like grouping customers together based on behavior exactly.

153
00:07:33.079 --> 00:07:36.680
<v Speaker 2>Customer segmentation is a classic example, finding groups you didn't

154
00:07:36.720 --> 00:07:39.800
<v Speaker 2>even know existed. The book also gives a nod to

155
00:07:40.079 --> 00:07:44.240
<v Speaker 2>deep learning, especially for unstructured stuff like text and images, right,

156
00:07:44.279 --> 00:07:47.519
<v Speaker 2>the complex stuff, but it also wisely points out it's

157
00:07:47.519 --> 00:07:50.519
<v Speaker 2>not always the magic bullet, especially for standard you know,

158
00:07:50.839 --> 00:07:53.560
<v Speaker 2>tabular data. Sometimes simpler models work better.

159
00:07:53.800 --> 00:07:57.480
<v Speaker 1>Okay, useful distinction. So you're building the tech, assembling the team,

160
00:07:58.079 --> 00:08:01.319
<v Speaker 1>but the book is very clear technology isn't the whole story.

161
00:08:01.600 --> 00:08:05.120
<v Speaker 1>You need to cultivate a data driven culture. What does

162
00:08:05.160 --> 00:08:06.120
<v Speaker 1>that actually look like?

163
00:08:06.399 --> 00:08:09.199
<v Speaker 2>It means creating an environment where using data to make

164
00:08:09.240 --> 00:08:12.279
<v Speaker 2>decisions isn't just like a special project, It's just how

165
00:08:12.360 --> 00:08:15.480
<v Speaker 2>things are done everywhere, ingrained in the day to day exactly.

166
00:08:15.800 --> 00:08:18.319
<v Speaker 2>Technology is just the tool. The real shift is in

167
00:08:18.319 --> 00:08:21.480
<v Speaker 2>people's mindset and behavior, and the book suggests several ways

168
00:08:21.519 --> 00:08:24.879
<v Speaker 2>to nurture this. One is finding and growing talent.

169
00:08:24.759 --> 00:08:27.120
<v Speaker 1>Internally, not just hiring externally.

170
00:08:27.399 --> 00:08:32.360
<v Speaker 2>Right, maybe offer introductory courses, say in Python, across different departments.

171
00:08:32.519 --> 00:08:35.519
<v Speaker 2>You might uncover hidden talent, people who are really keen

172
00:08:35.559 --> 00:08:39.000
<v Speaker 2>on data but never have the chance upscill your existing folks.

173
00:08:39.080 --> 00:08:41.559
<v Speaker 1>I like that, democratizing it a bit. The book also

174
00:08:41.600 --> 00:08:46.039
<v Speaker 1>talks about internal hackathons like it Lassians ship it days,

175
00:08:46.320 --> 00:08:48.240
<v Speaker 1>but maybe smaller scale to start.

176
00:08:48.440 --> 00:08:51.960
<v Speaker 2>Yeah, even mini hackathons can be fantastic, great for learning,

177
00:08:52.120 --> 00:08:55.759
<v Speaker 2>trying new things, getting different teams, collaborating on data problems.

178
00:08:56.000 --> 00:08:59.399
<v Speaker 2>Plus they can be fun, create some buzz definitely. And

179
00:08:59.440 --> 00:09:02.120
<v Speaker 2>then there are things things like communities of practice, do

180
00:09:02.240 --> 00:09:06.559
<v Speaker 2>heat a catapo shares experiences on this, or data champions networks.

181
00:09:06.639 --> 00:09:09.440
<v Speaker 2>Kyl Ash mentions this ways to connect people interested in

182
00:09:09.519 --> 00:09:14.039
<v Speaker 2>data across the organization, spreading the knowledge and enthusiasm precisely.

183
00:09:14.279 --> 00:09:17.720
<v Speaker 2>It all feeds into building broader data literacy, which the

184
00:09:17.720 --> 00:09:20.200
<v Speaker 2>book highlights as absolutely critical.

185
00:09:19.879 --> 00:09:22.559
<v Speaker 1>And not just for the data team right. Ian Jackman

186
00:09:22.639 --> 00:09:24.919
<v Speaker 1>and zanvan Wick emphasize.

187
00:09:24.320 --> 00:09:27.559
<v Speaker 2>This, No, absolutely not. Data literacy is for everyone. It's

188
00:09:27.600 --> 00:09:31.440
<v Speaker 2>about understanding data, being able to interpret it, question it,

189
00:09:31.519 --> 00:09:34.240
<v Speaker 2>think critically about its quality and how it's being used.

190
00:09:34.720 --> 00:09:37.879
<v Speaker 2>Sandra Hogan calls it empowering the business community.

191
00:09:38.240 --> 00:09:41.519
<v Speaker 1>And there's that key phrase. Data doesn't make decisions.

192
00:09:41.639 --> 00:09:47.039
<v Speaker 2>People do crucial point data informs, but humans decide, which

193
00:09:47.080 --> 00:09:50.120
<v Speaker 2>brings in the need for critical thinking skills across the board.

194
00:09:50.559 --> 00:09:54.320
<v Speaker 2>Tim van Gelder's definition is great the art of being right,

195
00:09:54.840 --> 00:09:56.919
<v Speaker 2>partly by considering how you might.

196
00:09:56.799 --> 00:10:00.159
<v Speaker 1>Be wrong, thinking critically about the data itself. Okay, makes

197
00:10:00.200 --> 00:10:05.159
<v Speaker 1>sense now, shifting gears slightly, actually choosing the technology. That

198
00:10:05.279 --> 00:10:09.440
<v Speaker 1>landscape can feel incredibly complex. Any practical guidance from the book.

199
00:10:09.279 --> 00:10:11.559
<v Speaker 2>Yeah, it offers some solid advice. First, maybe start with

200
00:10:11.600 --> 00:10:13.879
<v Speaker 2>cloud providers. Your organization is already.

201
00:10:13.600 --> 00:10:15.759
<v Speaker 1>Familiar with less friction that way usually. Yeah.

202
00:10:15.799 --> 00:10:20.200
<v Speaker 2>It mentions the Big three AWS, AZURE, GCP and notes

203
00:10:20.240 --> 00:10:23.600
<v Speaker 2>that since many orgs use Microsoft Tools. Azure often feels

204
00:10:23.639 --> 00:10:25.960
<v Speaker 2>like a natural starting point for the bi side of things,

205
00:10:26.240 --> 00:10:30.120
<v Speaker 2>but a really important theme is interoperability. Design things to

206
00:10:30.159 --> 00:10:34.759
<v Speaker 2>be modular, API driven avoid getting locked into one vendor.

207
00:10:34.840 --> 00:10:36.399
<v Speaker 2>Ian Jackman really stresses this.

208
00:10:36.559 --> 00:10:39.559
<v Speaker 1>Future proofing essentially exactly needs change.

209
00:10:39.799 --> 00:10:42.240
<v Speaker 2>De Weeeda Katapou gives an example of her org moving

210
00:10:42.279 --> 00:10:45.639
<v Speaker 2>from Redshift to Snowflake because Redshift struggled with their Jason

211
00:10:45.759 --> 00:10:49.279
<v Speaker 2>data needs at the time. You need that flexibility, good example.

212
00:10:49.360 --> 00:10:52.879
<v Speaker 2>What else, usability for the team is key. Heima Prosade

213
00:10:52.960 --> 00:10:55.480
<v Speaker 2>makes this point. If the tools are clunky or hard

214
00:10:55.480 --> 00:10:58.480
<v Speaker 2>for your data folks to use, adoption just won't happen.

215
00:10:58.600 --> 00:11:00.600
<v Speaker 1>Makes sense, Keep the users happy.

216
00:11:00.360 --> 00:11:03.080
<v Speaker 2>And don't forget the ongoing effort to keep the lights

217
00:11:03.080 --> 00:11:06.519
<v Speaker 2>on or ktlo work needed to maintain any data product

218
00:11:06.600 --> 00:11:08.639
<v Speaker 2>or platform that needs factoring in.

219
00:11:08.840 --> 00:11:11.440
<v Speaker 1>Right, it's not just build and forget and the process

220
00:11:11.480 --> 00:11:12.639
<v Speaker 1>for choosing, it's just.

221
00:11:12.720 --> 00:11:16.720
<v Speaker 2>The structured way brainstorm your decision criteria collectively, get input

222
00:11:16.720 --> 00:11:20.759
<v Speaker 2>from everyone affected. Then maybe use something like pair wise comparison,

223
00:11:20.840 --> 00:11:24.639
<v Speaker 2>comparing options side by side on each specific criterion to

224
00:11:24.679 --> 00:11:26.240
<v Speaker 2>help make a rational choice.

225
00:11:26.399 --> 00:11:31.240
<v Speaker 1>Okay, methodical approach. So we've got tech team culture selection.

226
00:11:32.159 --> 00:11:36.879
<v Speaker 1>What about governance and ethics that feels increasingly important, hugely important.

227
00:11:37.159 --> 00:11:40.080
<v Speaker 2>The book frames data, governance and ethics not as nice

228
00:11:40.120 --> 00:11:43.440
<v Speaker 2>to have as anymore, but absolute, must have, non negotiable

229
00:11:43.480 --> 00:11:46.519
<v Speaker 2>pretty much. Governance provides the rules of the road. The

230
00:11:46.600 --> 00:11:52.639
<v Speaker 2>processes rolls responsibilities for managing data well, ensuring quality, security compliance.

231
00:11:53.240 --> 00:11:59.240
<v Speaker 2>Think DAMA definitions, managing availability, usability, integrity. Security sounds potentially

232
00:11:59.279 --> 00:12:02.360
<v Speaker 2>massive that it can be, But the advice, echoing fvirus

233
00:12:02.399 --> 00:12:06.200
<v Speaker 2>Hamden's experience is start small. Build governance based on the

234
00:12:06.200 --> 00:12:08.919
<v Speaker 2>problems you actually have right now. Don't try to boil

235
00:12:08.960 --> 00:12:12.519
<v Speaker 2>the ocean on day one. Establish data stewards, maybe data

236
00:12:12.559 --> 00:12:16.200
<v Speaker 2>councils to manage specific areas. The road organically, yes, and

237
00:12:16.240 --> 00:12:19.120
<v Speaker 2>ethics is woven through this. Especially with AI, bias and

238
00:12:19.159 --> 00:12:22.240
<v Speaker 2>fairness are critical. You have to minimize bias to build trust.

239
00:12:22.519 --> 00:12:25.240
<v Speaker 2>There's an IBM stat mentioned about how crucial trust is

240
00:12:25.240 --> 00:12:26.039
<v Speaker 2>for AI adoption.

241
00:12:26.440 --> 00:12:28.159
<v Speaker 1>Bias can creep in easily.

242
00:12:28.200 --> 00:12:31.559
<v Speaker 2>Very easily. The book uses a great phrase data does

243
00:12:31.600 --> 00:12:34.600
<v Speaker 2>not speak for itself. It depends on perspective, how it

244
00:12:34.679 --> 00:12:38.159
<v Speaker 2>was collected, how it's interpreted. The example given is how

245
00:12:38.200 --> 00:12:41.159
<v Speaker 2>different Sydney suburbs might look depending on which data points

246
00:12:41.200 --> 00:12:42.279
<v Speaker 2>you choose to highlight right.

247
00:12:42.360 --> 00:12:43.799
<v Speaker 1>Context matters hugely.

248
00:12:44.039 --> 00:12:47.720
<v Speaker 2>Plus, you have regulations like GDPR, data privacy rights, the

249
00:12:47.799 --> 00:12:50.960
<v Speaker 2>right to be informed, access erasure and so on. These

250
00:12:51.000 --> 00:12:51.759
<v Speaker 2>have to be built in.

251
00:12:51.960 --> 00:12:54.720
<v Speaker 1>So responsible AI needs thinking about from the start.

252
00:12:54.919 --> 00:12:58.000
<v Speaker 2>Absolutely. Doctor Lean's point in the book is about building

253
00:12:58.000 --> 00:13:02.120
<v Speaker 2>in diversity from the beginning to encourage ethical discussions. Frame

254
00:13:02.159 --> 00:13:05.519
<v Speaker 2>it using the language your organization understands. Maybe it's risk management,

255
00:13:05.519 --> 00:13:09.360
<v Speaker 2>maybe it's core company values, maybe it's sustainable development goals.

256
00:13:09.679 --> 00:13:11.919
<v Speaker 2>Find the hook to embed ethical thinking.

257
00:13:12.000 --> 00:13:14.879
<v Speaker 1>Okay, that brings a lot together. So wrapping this up,

258
00:13:15.240 --> 00:13:18.960
<v Speaker 1>what's the single biggest takeaway from this whole emergent design

259
00:13:19.000 --> 00:13:21.720
<v Speaker 1>approach to building data capabilities?

260
00:13:22.080 --> 00:13:24.840
<v Speaker 2>I think the core message is it's a process. It's

261
00:13:24.840 --> 00:13:27.159
<v Speaker 2>not a project with a neat finish line. It's this

262
00:13:27.360 --> 00:13:32.639
<v Speaker 2>ongoing emergent journey of learning, adapting, and constantly collaborating between

263
00:13:32.679 --> 00:13:34.679
<v Speaker 2>the tech side and the business side.

264
00:13:34.399 --> 00:13:38.720
<v Speaker 1>A continuous evolution exactly. And for someone listening feeling maybe inspired,

265
00:13:38.720 --> 00:13:41.440
<v Speaker 1>but still a bit unsure where to start, what's one

266
00:13:41.519 --> 00:13:43.679
<v Speaker 1>concrete next step you'd suggest based.

267
00:13:43.399 --> 00:13:46.519
<v Speaker 2>On this, Start with the conversations, use the what keeps

268
00:13:46.519 --> 00:13:50.159
<v Speaker 2>you up at night. Question Talk to different teams, Find

269
00:13:50.200 --> 00:13:53.440
<v Speaker 2>one specific nagging business problem that data might.

270
00:13:53.320 --> 00:13:55.720
<v Speaker 1>Help with just one to start just one.

271
00:13:56.000 --> 00:13:59.679
<v Speaker 2>Then focus on delivering a small, targeted proof of value

272
00:13:59.720 --> 00:14:03.159
<v Speaker 2>for that problem. Show a tangible win, however small, that

273
00:14:03.200 --> 00:14:04.120
<v Speaker 2>builds momentum.

274
00:14:04.200 --> 00:14:07.679
<v Speaker 1>Great advice, and finally, a last thought for our listener

275
00:14:07.720 --> 00:14:10.120
<v Speaker 1>to maybe chew on after this deep dive.

276
00:14:10.360 --> 00:14:13.720
<v Speaker 2>Maybe reflect on that idea of the ever changing reality

277
00:14:13.879 --> 00:14:17.600
<v Speaker 2>borrowing from Claudiosubora. Things will change the market, the business,

278
00:14:17.720 --> 00:14:20.240
<v Speaker 2>the tech. An emergent approach accepts that.

279
00:14:20.399 --> 00:14:22.240
<v Speaker 1>It embraces the flux right.

280
00:14:22.399 --> 00:14:25.399
<v Speaker 2>So think about how you can foster that ongoing dialogue,

281
00:14:25.399 --> 00:14:28.600
<v Speaker 2>that iterative adaptation in your organization to make sure your

282
00:14:28.679 --> 00:14:31.799
<v Speaker 2>data efforts stay aligned and keep delivering value as things

283
00:14:31.840 --> 00:14:32.879
<v Speaker 2>inevitably shift.

284
00:14:33.080 --> 00:14:35.600
<v Speaker 1>A really powerful way to think about it, Fantastic Insights,

285
00:14:35.600 --> 00:14:37.639
<v Speaker 1>Thanks so much for unpacking all of that with us today,

286
00:14:37.720 --> 00:14:40.159
<v Speaker 1>My pleasure. We really hope this deep dive gives you,

287
00:14:40.320 --> 00:14:43.639
<v Speaker 1>our listener, a clearer path forward on your own data journey.

288
00:14:43.759 --> 00:14:44.639
<v Speaker 1>Thanks for joining us