1
00:00:00,000 --> 00:00:06,170
< Intro >

2
00:00:06,170 --> 00:00:09,334
– Welcome to another
insightful episode of Count Me In.

3
00:00:09,334 --> 00:00:12,730
Today, we're delving into the topic
of building a data-driven culture

4
00:00:12,730 --> 00:00:14,839
with our esteemed guest, Connie Siu,

5
00:00:14,839 --> 00:00:18,167
President of CDC Synectics Incorporated,

6
00:00:18,167 --> 00:00:19,834
and an accomplished author.

7
00:00:19,834 --> 00:00:22,000
Join us, as Connie shares her expertise

8
00:00:22,000 --> 00:00:26,060
on essential elements of data-driven
culture within an organization,

9
00:00:26,060 --> 00:00:29,430
and the significant impact it has
on today's business environment.

10
00:00:29,430 --> 00:00:32,167
Stay tuned, as we explore
key challenges faced

11
00:00:32,167 --> 00:00:34,430
during the transition,
and gain valuable insights

12
00:00:34,430 --> 00:00:37,334
on assessing the effectiveness
of a data-driven culture.

13
00:00:37,334 --> 00:00:40,250
This episode promises
to offer valuable insights,

14
00:00:40,250 --> 00:00:41,500
into the power of data-driven

15
00:00:41,500 --> 00:00:44,620
decision-making in shaping
organizational cultures

16
00:00:44,620 --> 00:00:47,260
and driving business success.

17
00:00:47,260 --> 00:00:48,000
Let's get started.

18
00:00:48,000 --> 00:00:54,500
< Music >

19
00:00:54,500 --> 00:00:56,000
Well, Connie, we want
to thank you so much

20
00:00:56,000 --> 00:00:58,667
for coming back on
the Count Me In podcast.

21
00:00:58,667 --> 00:01:01,500
And, today, we're going to be
talking about data-driven culture 

22
00:01:01,500 --> 00:01:03,000
and what that means.

23
00:01:03,000 --> 00:01:06,833
And, so, maybe, we can start off,
you can elaborate what constitutes

24
00:01:06,833 --> 00:01:10,090
having a data-driven culture
within an organization,

25
00:01:10,090 --> 00:01:14,167
and why is it essential,
especially, in business today?

26
00:01:15,333 --> 00:01:17,770
– That's a great start, Adam.

27
00:01:17,770 --> 00:01:22,350
Data-driven culture is the
consistent values and beliefs

28
00:01:22,350 --> 00:01:28,060
in distilling insights from data
to drive informed decision making,

29
00:01:28,060 --> 00:01:32,340
and that's happening across
the whole organization.

30
00:01:32,340 --> 00:01:34,666
And I would offer three characteristics

31
00:01:34,666 --> 00:01:39,166
that you can look for in, an organization,

32
00:01:39,166 --> 00:01:41,170
where there's a data-driven culture.

33
00:01:41,170 --> 00:01:46,060
The first one is you will
see individuals and teams

34
00:01:46,060 --> 00:01:48,500
actively asking themselves questions like,

35
00:01:48,500 --> 00:01:53,666
"What information we can draw on
to support and guide decisions."

36
00:01:54,189 --> 00:01:56,833
You will see consistent efforts devoted

37
00:01:56,833 --> 00:02:01,166
to pull relevant data to analyze an issue.

38
00:02:01,166 --> 00:02:05,899
And you will see open and frank
dialogues on understanding

39
00:02:05,899 --> 00:02:10,166
the root-cause of problem
by looking closely at KPIs.

40
00:02:10,833 --> 00:02:16,166
In terms of why it is essential
for businesses today,

41
00:02:16,166 --> 00:02:22,300
there are four factors, two
external and two internal,

42
00:02:22,300 --> 00:02:26,629
that are important to bear in mind.

43
00:02:26,629 --> 00:02:31,999
The first external factor is
competitive marketplace.

44
00:02:31,999 --> 00:02:34,499
Companies need focused strategies

45
00:02:34,499 --> 00:02:40,166
to target the right markets, to
differentiate themselves to compete,

46
00:02:40,166 --> 00:02:42,166
and they need the market intelligence

47
00:02:42,166 --> 00:02:45,666
to develop focused strategies.

48
00:02:45,666 --> 00:02:50,800
The second external factor
is digital transformation.

49
00:02:50,800 --> 00:02:53,499
The ability to adopt the right technologies

50
00:02:53,499 --> 00:02:57,999
to drive business outcomes is critical.

51
00:02:57,999 --> 00:02:59,999
Successful digital transformation

52
00:02:59,999 --> 00:03:03,999
involves using technology
to capture relevant data

53
00:03:03,999 --> 00:03:05,999
and analyze the results.

54
00:03:06,832 --> 00:03:10,332
To automate processes, for instance,

55
00:03:10,332 --> 00:03:15,110
companies need to know what
data is important and what's not.

56
00:03:16,166 --> 00:03:17,780
The internal factors:

57
00:03:17,780 --> 00:03:23,166
The first one is operational efficiency.

58
00:03:23,166 --> 00:03:25,500
Businesses need to be efficient today,

59
00:03:25,500 --> 00:03:30,319
and we are aware that costs
are going up, labor, materials.

60
00:03:30,319 --> 00:03:32,999
And with the current inflation,

61
00:03:32,999 --> 00:03:36,165
companies need to have a
good handle on the numbers.

62
00:03:36,165 --> 00:03:42,332
The second internal factor is the
need to treat data as a strategic asset.

63
00:03:43,332 --> 00:03:45,830
Every business has tons of data.

64
00:03:45,830 --> 00:03:51,140
Imagine if you can mine
the data for intelligence,

65
00:03:51,140 --> 00:03:55,165
they will uncover lots of opportunities
to make all kinds of improvements,

66
00:03:55,165 --> 00:03:59,832
such as targeting
high-margin niche markets.

67
00:03:59,832 --> 00:04:01,832
So these four factors require

68
00:04:01,832 --> 00:04:06,499
an appreciation of making smart
choices from data analytics.

69
00:04:07,209 --> 00:04:10,909
It is more important, than ever,
to build a data-driven culture.

70
00:04:11,998 --> 00:04:16,998
– Yes, I think those are some great
factors to take into consideration,

71
00:04:16,998 --> 00:04:18,949
especially, if you recognize
that your organization

72
00:04:18,949 --> 00:04:21,270
doesn't have that data-driven culture.

73
00:04:21,270 --> 00:04:24,498
Maybe we can talk about
some key challenges

74
00:04:24,498 --> 00:04:27,340
that organizations face when
they're trying to transition to that.

75
00:04:27,340 --> 00:04:29,165
Because it's not something
that happens overnight,

76
00:04:29,165 --> 00:04:30,332
something that you can turn a switch

77
00:04:30,332 --> 00:04:32,320
and say, "Hey, we're a data-driven culture."

78
00:04:32,320 --> 00:04:34,665
It's something that
builds over time, I'm sure.

79
00:04:34,998 --> 00:04:39,199
– Yes, there are two key
challenges I'd like to share.

80
00:04:39,199 --> 00:04:43,165
The first one is the lack
of technical capabilities.

81
00:04:43,832 --> 00:04:45,665
And when I say technical capabilities,

82
00:04:45,665 --> 00:04:53,165
they include the skills to identify what
data, or KPIs, are relevant to look at.

83
00:04:53,165 --> 00:04:56,331
They include skills to
analyze the numbers.

84
00:04:56,331 --> 00:05:01,165
For instance, how do you know you
have achieved efficiency improvement?

85
00:05:01,165 --> 00:05:06,665
What would you look at to
monitor process performance?

86
00:05:07,165 --> 00:05:09,498
Do you want to look at the
results on a weekly basis

87
00:05:09,498 --> 00:05:13,831
or it makes better sense to compare
month over month changes.

88
00:05:14,570 --> 00:05:16,760
And there are many data
points you can look at,

89
00:05:16,760 --> 00:05:19,498
but not all of them are relevant.

90
00:05:20,331 --> 00:05:23,665
Once you have the data, you
need the tools to capture,

91
00:05:23,665 --> 00:05:26,331
compile, and analyze them.

92
00:05:26,331 --> 00:05:32,498
And many companies are still using
legacy systems that are not integrated.

93
00:05:32,498 --> 00:05:37,664
So it is a tedious and often very
frustrating exercise to extract the data.

94
00:05:38,998 --> 00:05:45,419
And to overcome that lack of technical
capabilities, start with training.

95
00:05:45,419 --> 00:05:46,759
Training the fundamental skills

96
00:05:46,759 --> 00:05:52,164
on asking good questions to identify
what data do we need to look at.

97
00:05:52,164 --> 00:05:54,998
Training on the skills to analyze an issue.

98
00:05:55,830 --> 00:05:57,331
And I would suggest train everyone

99
00:05:57,331 --> 00:06:02,110
from the executives to people
working on the front line.

100
00:06:02,110 --> 00:06:05,080
We don't need to train everyone
to be a data scientist,

101
00:06:05,080 --> 00:06:09,919
but we do need them to have the
basic skills to ask good questions.

102
00:06:09,919 --> 00:06:15,831
To understand what they need to look
at, and become good problem solvers.

103
00:06:15,831 --> 00:06:18,910
And in terms of the legacy systems,

104
00:06:18,910 --> 00:06:22,250
there's only so much you
can do patching them.

105
00:06:22,250 --> 00:06:26,164
Eventually you need to
invest in modern technologies,

106
00:06:26,164 --> 00:06:28,831
and there are so many
options out there today,

107
00:06:28,831 --> 00:06:35,164
and there's no need, and
I want to emphasize this,

108
00:06:35,164 --> 00:06:39,997
it's not necessary to invest in
the most comprehensive ERP.

109
00:06:39,997 --> 00:06:46,497
The key is to find the right applications
that meet your business needs.

110
00:06:46,997 --> 00:06:52,497
Now, the second challenge I'd like
to talk about is the lack of buy-in.

111
00:06:52,497 --> 00:06:55,497
When you don't have the support
of the senior management team

112
00:06:55,497 --> 00:06:59,997
and the middle managers, it is
very difficult to make that shift.

113
00:07:00,664 --> 00:07:02,664
Now, middle managers are accountable

114
00:07:02,664 --> 00:07:04,497
for the team's performance.

115
00:07:04,497 --> 00:07:08,664
So that fear of poor results is natural

116
00:07:08,664 --> 00:07:11,497
because they reflect on
their leadership skills,

117
00:07:11,497 --> 00:07:14,669
and no one wants to look bad.

118
00:07:14,669 --> 00:07:17,830
When middle managers shy
away from results reporting,

119
00:07:17,830 --> 00:07:22,164
they tend to do the minimal,
just what is needed.

120
00:07:22,164 --> 00:07:24,319
Essentially they create an alignment

121
00:07:24,319 --> 00:07:30,664
where there's little incentive for
the team to embrace analytics.

122
00:07:30,664 --> 00:07:33,569
Now, when we look at the
senior management team,

123
00:07:33,569 --> 00:07:37,169
when there's no buy-in
from them, on analytics,

124
00:07:37,169 --> 00:07:42,009
you tend to see an authoritative
management style.

125
00:07:42,009 --> 00:07:49,663
Top-down decisions will become
directives for the teams to execute.

126
00:07:49,663 --> 00:07:57,470
And in this situation, the efforts made
on analytics are not valued at all.

127
00:07:57,470 --> 00:08:00,770
To overcome the lack of support,

128
00:08:00,770 --> 00:08:04,490
start with understanding
what the dynamics is today

129
00:08:04,490 --> 00:08:06,710
and find your champion.

130
00:08:06,710 --> 00:08:11,330
That champion could be a
team leader for a small group,

131
00:08:11,330 --> 00:08:16,080
a middle manager, or an executive.

132
00:08:16,080 --> 00:08:20,830
Someone who is receptive to analytics,

133
00:08:20,830 --> 00:08:25,663
open to discussing results, and
also willing to devote the time

134
00:08:25,663 --> 00:08:28,996
and effort to data analytics.

135
00:08:29,663 --> 00:08:32,163
And once you have that champion,

136
00:08:32,163 --> 00:08:37,163
pick a problem to tackle and develop
a game plan, and that game plan

137
00:08:37,163 --> 00:08:40,996
has got to be practical, for folks
who will be doing the work.

138
00:08:40,996 --> 00:08:42,496
Include, in your game plan;

139
00:08:42,496 --> 00:08:43,830
1. How you're going to capture the data.

140
00:08:43,830 --> 00:08:45,830
2. What tool you're going to use.

141
00:08:45,830 --> 00:08:48,496
3. Who is going to do the analysis.

142
00:08:48,496 --> 00:08:52,663
4. What forum you're going to bring
folks together to discuss the results.

143
00:08:52,663 --> 00:08:56,830
5. Who is going to make
decision on what action to take,

144
00:08:56,830 --> 00:09:00,060
and implement the improvements.

145
00:09:00,060 --> 00:09:03,180
And, then, go through the
cycle of monitoring the results

146
00:09:03,180 --> 00:09:05,829
and refine your changes.

147
00:09:05,829 --> 00:09:10,663
So those are the key points on
overcoming the lack of buy-in.

148
00:09:10,663 --> 00:09:13,850
– Yes, that's a big one, is making
sure you have that proponent,

149
00:09:13,850 --> 00:09:17,996
that person, who can help lead
the change in the organization.

150
00:09:17,996 --> 00:09:19,560
Because unless that's
coming from the top-down,

151
00:09:19,560 --> 00:09:23,850
it's very difficult to drive
that change in the culture.

152
00:09:23,850 --> 00:09:27,329
– Yes, definitely, and one
thing I forgot to mention is

153
00:09:27,329 --> 00:09:30,610
share your success stories with
as many groups as you can.

154
00:09:30,610 --> 00:09:35,496
Because the more you can broadcast
how analytics will help improving

155
00:09:35,496 --> 00:09:38,663
business outcomes,
you will build momentum

156
00:09:38,663 --> 00:09:41,130
and excitement around analytics.

157
00:09:41,663 --> 00:09:43,380
– Now, one thing I
wanted to circle back to,

158
00:09:43,380 --> 00:09:45,470
you were mentioning legacy systems,

159
00:09:45,470 --> 00:09:49,329
and how it's hard to connect things
and there's a lot of manual data.

160
00:09:49,329 --> 00:09:51,180
Maybe we can talk a little bit about

161
00:09:51,180 --> 00:09:54,162
how companies should
strategically invest that money.

162
00:09:54,162 --> 00:09:56,079
Especially if you're a medium
to small-sized business,

163
00:09:56,079 --> 00:09:58,496
it's not always easy to
implement new systems,

164
00:09:58,496 --> 00:09:59,910
you might not have the capital.

165
00:09:59,910 --> 00:10:02,162
But you want to strategically invest
that money so that you can have 

166
00:10:02,162 --> 00:10:03,890
the right systems in place,

167
00:10:03,890 --> 00:10:06,329
to foster that data culture
we've been talking about.

168
00:10:06,662 --> 00:10:11,920
– Yes, there are three areas
I would offer for consideration.

169
00:10:11,920 --> 00:10:16,110
The first one is to build the
capabilities within the organization.

170
00:10:16,110 --> 00:10:20,662
So that goes back to training
employees on the skills that they need.

171
00:10:20,662 --> 00:10:24,329
To ask good questions to
identify what data they need.

172
00:10:24,329 --> 00:10:27,870
Train them on how to analyze results,

173
00:10:27,870 --> 00:10:29,670
with the skills they will take ownership

174
00:10:29,670 --> 00:10:33,329
on the data capture and analysis.

175
00:10:33,329 --> 00:10:36,662
The second area to
invest in is technology.

176
00:10:36,662 --> 00:10:39,240
The key is to find
the right technologies.

177
00:10:39,240 --> 00:10:42,162
Some companies will
spend thousands of dollars

178
00:10:42,162 --> 00:10:46,662
and potentially millions to
invest in state-of-the-art ERP.

179
00:10:46,662 --> 00:10:51,162
But, yet, they might be using 10%
or even 5% of the functionalities.

180
00:10:51,162 --> 00:10:54,829
So any way you look at it, they're
not going to get the ROI on that.

181
00:10:54,920 --> 00:10:58,329
And there are lots of smaller
applications out there,

182
00:10:58,329 --> 00:11:02,680
cloud solutions, for instance,
today, that are very affordable.

183
00:11:02,680 --> 00:11:06,010
And for smaller businesses, they
might want to focus on those

184
00:11:06,010 --> 00:11:08,828
and hone in on what are
the biggest functions

185
00:11:08,828 --> 00:11:13,662
that you need from that application,
and that's the best way to go.

186
00:11:13,662 --> 00:11:16,870
And you want to make sure,
also, the tool is easy to use.

187
00:11:16,870 --> 00:11:20,162
Those big ERPs, generally,
are clunky to use.

188
00:11:20,162 --> 00:11:24,662
So the smaller and simpler the
tool is, you get better user adoption.

189
00:11:24,662 --> 00:11:27,162
Because when users
use a tool haphazardly,

190
00:11:27,162 --> 00:11:30,495
you end up with incorrect
and inaccurate data.

191
00:11:30,495 --> 00:11:35,019
The third area to invest in,
it's got to be time and effort.

192
00:11:35,019 --> 00:11:37,820
It takes time to do the work, capture data,

193
00:11:37,820 --> 00:11:42,870
compile it, analyze, discuss, take
action, make improvements, et cetera.

194
00:11:42,870 --> 00:11:45,828
So it's not something that you want
the staff to do it for one month,

195
00:11:45,828 --> 00:11:47,910
put it aside for a few months
and come back to it.

196
00:11:47,910 --> 00:11:48,995
It doesn't work that way.

197
00:11:48,995 --> 00:11:51,700
To build that culture,
you got to be consistent

198
00:11:51,700 --> 00:11:56,328
and put in the time,
regularly, to build that habit.

199
00:11:56,328 --> 00:12:00,250
So when you invest your time
and effort in these three areas,

200
00:12:00,250 --> 00:12:05,970
technical capabilities, technology,
and time and effort to build a habit.

201
00:12:05,970 --> 00:12:08,995
You will build a confidence
for your teams,

202
00:12:08,995 --> 00:12:13,661
hopefully, across the organization,
to make a shift to a data-driven culture.

203
00:12:14,161 --> 00:12:16,328
– Yes, no, that's great advice.

204
00:12:16,328 --> 00:12:17,328
But when you think about all the data

205
00:12:17,328 --> 00:12:20,328
that we have in organizations,
it can be very difficult.

206
00:12:20,328 --> 00:12:23,828
And all that data is not
necessarily quality data.

207
00:12:23,828 --> 00:12:27,970
The old adage "Garbage in, garbage out".

208
00:12:27,970 --> 00:12:30,661
How can organizations
ensure that they possess

209
00:12:30,661 --> 00:12:32,389
a complete set of accurate data?

210
00:12:32,389 --> 00:12:34,860
And some of that time that you 
were talking about putting in,

211
00:12:34,860 --> 00:12:37,328
does that include cleaning up the data?

212
00:12:38,320 --> 00:12:41,161
– That's an excellent question, Adam.

213
00:12:41,161 --> 00:12:44,370
Quality data is a challenge
for many companies,

214
00:12:44,370 --> 00:12:48,161
and it's nice to have
accurate and complete data.

215
00:12:48,161 --> 00:12:53,990
But, in reality, most companies
still have a lot of work to do.

216
00:12:53,990 --> 00:12:57,470
Of course, you can clean and
correct your historical data,

217
00:12:57,470 --> 00:13:02,220
in your systems, but it is
usually a painful exercise

218
00:13:02,220 --> 00:13:05,994
and often the game might
not worth the efforts.

219
00:13:05,994 --> 00:13:09,750
So if you, indeed, need to make
decisions based on historical data,

220
00:13:09,750 --> 00:13:11,994
I would suggest a couple of things.

221
00:13:11,994 --> 00:13:16,389
The first one is to understand
where your data deficiencies are,

222
00:13:16,389 --> 00:13:20,161
and incorporate assumptions
in your analysis.

223
00:13:20,161 --> 00:13:22,890
Develop the worst-case
and the best-case scenario,

224
00:13:22,890 --> 00:13:24,494
so you have the bookends.

225
00:13:24,494 --> 00:13:28,661
And when you apply your business
savviness to your numbers, 

226
00:13:28,661 --> 00:13:30,470
you make better decisions.

227
00:13:30,470 --> 00:13:33,540
For example, when Covid hit, 2020,

228
00:13:33,540 --> 00:13:36,494
we know that in the
second half of the year,

229
00:13:36,494 --> 00:13:40,170
the shipping costs went sky high.

230
00:13:40,170 --> 00:13:45,850
So if you include the cost for
those six to eight months, in 2020,

231
00:13:45,850 --> 00:13:49,790
when you want to deduce the average
margin cost for your portfolio,

232
00:13:49,790 --> 00:13:52,160
you know the numbers will be out.

233
00:13:52,160 --> 00:13:56,100
But you know the reasons, and
you can explain the anomalies.

234
00:13:56,100 --> 00:14:01,494
The second option is you can exclude
those data points from your analysis.

235
00:14:01,494 --> 00:14:05,040
Now, the second part to make decisions

236
00:14:05,040 --> 00:14:07,494
from historical data, as you mentioned.

237
00:14:07,494 --> 00:14:12,720
If you have the time and manpower
to do the data cleanup, you can do it.

238
00:14:12,720 --> 00:14:15,860
But I would suggest to be very selective

239
00:14:15,860 --> 00:14:20,827
on how much you want to do because
you don't want to get into a spiral.

240
00:14:20,827 --> 00:14:22,890
Now, that's historical data.

241
00:14:22,890 --> 00:14:27,080
Going forward, though, you have
more control on the data quality,

242
00:14:27,080 --> 00:14:30,430
and there are two parts
to that, to build good data.

243
00:14:30,430 --> 00:14:34,370
The first part is to
capture meaningful data.

244
00:14:34,370 --> 00:14:38,660
The second part is have good data input.

245
00:14:38,660 --> 00:14:40,660
Let's look at the first part first.

246
00:14:40,660 --> 00:14:43,993
Capture meaningful data;
so that is training your staff

247
00:14:43,993 --> 00:14:46,820
to have the skills to ask quick questions,

248
00:14:46,820 --> 00:14:50,660
so they know what data
they need to go after.

249
00:14:50,660 --> 00:14:53,160
And, essentially, when
they're good at that,

250
00:14:53,160 --> 00:14:57,660
they will become filters for
capturing meaningful data.

251
00:14:57,660 --> 00:15:01,160
Now, the second part
is good data capture.

252
00:15:01,827 --> 00:15:06,660
What is most critical here is to
have the tool that is easy to use.

253
00:15:08,399 --> 00:15:12,950
Think about a worker working in
the site, on a construction site.

254
00:15:12,950 --> 00:15:16,493
They have limited amount
of time to enter data,

255
00:15:16,493 --> 00:15:18,889
and you got to make it easy for them.

256
00:15:18,889 --> 00:15:22,993
Use drop-down lists, for instance,
minimize the guesswork.

257
00:15:22,993 --> 00:15:25,270
And if they're working out in the rain,

258
00:15:25,270 --> 00:15:28,493
you're asking them to enter
20 data fields on a screen,

259
00:15:28,493 --> 00:15:31,110
that's not going to happen.

260
00:15:31,110 --> 00:15:34,660
So you want to ask for the
minimum amount of input,

261
00:15:34,660 --> 00:15:37,320
and that goes back to
ask for what is relevant.

262
00:15:37,320 --> 00:15:41,639
Forget about what's not relevant
because it doesn't make sense

263
00:15:41,639 --> 00:15:44,199
for them to do all that work.

264
00:15:44,199 --> 00:15:47,759
And you asked about the
building trust in data, too,

265
00:15:47,759 --> 00:15:50,550
and I would like to address that part.

266
00:15:50,550 --> 00:15:54,160
On how do you get people
build trust in the data

267
00:15:54,160 --> 00:15:57,659
and therefore the output
that you generate from it?

268
00:15:58,326 --> 00:16:02,180
One best approach I suggest
is to look at the results

269
00:16:02,180 --> 00:16:05,493
and do reasonableness tests.

270
00:16:05,493 --> 00:16:08,326
For example, you can
use a subset of the data

271
00:16:08,326 --> 00:16:13,649
and use that to verify the margins
for select skills of your portfolio,

272
00:16:13,649 --> 00:16:16,449
and share the analytics with
as many people as possible.

273
00:16:16,449 --> 00:16:19,690
Because the more pair of
eyes you get on the results,

274
00:16:19,690 --> 00:16:23,826
you get better feedback, and
you can tweak your analysis.

275
00:16:24,089 --> 00:16:26,326
The idea is not to go for perfection

276
00:16:26,326 --> 00:16:29,159
because you don't
want analysis paralysis.

277
00:16:29,659 --> 00:16:30,870
– Definitely, you don't want that,

278
00:16:30,870 --> 00:16:35,326
and I think it's so easy when there's
so much data to get lost in the details.

279
00:16:35,326 --> 00:16:38,949
And you can't talk about big data, you
can't talk about massive sets of data,

280
00:16:38,949 --> 00:16:43,840
without talking about generative
AI tools like ChatGPT.

281
00:16:43,840 --> 00:16:45,930
The ones that everybody's talking about.

282
00:16:45,930 --> 00:16:48,992
But in a lot of these tools, the ERP
systems that you're mentioning,

283
00:16:48,992 --> 00:16:50,790
a lot of them are incorporating

284
00:16:50,790 --> 00:16:55,269
those types of generative AI to help
you with the analysis of the data.

285
00:16:55,269 --> 00:16:58,992
So we've talked about how important
it is to have good data in your system.

286
00:16:58,992 --> 00:17:02,326
Now, how can these tools help be a tool?

287
00:17:02,326 --> 00:17:03,790
Obviously, they're not the end all, be all,

288
00:17:03,790 --> 00:17:06,659
because with all AI you need
HI, Human Intelligence,

289
00:17:06,659 --> 00:17:08,659
to make sure that they work together.

290
00:17:08,659 --> 00:17:11,490
But how can these tools
help with reliable insights,

291
00:17:11,490 --> 00:17:13,826
especially, with the power
of AI that's out there?

292
00:17:14,326 --> 00:17:20,825
– ChatGPT has really created
a big rave out there with AI.

293
00:17:20,825 --> 00:17:26,992
And with ChatGPT and AI-driven
insights, data quality is very important.

294
00:17:26,992 --> 00:17:30,320
And back in March, earlier this year,

295
00:17:30,320 --> 00:17:34,950
OpenAI did share that the
fourth generation of GPT,

296
00:17:34,950 --> 00:17:39,992
on average, makes up
stuff 20% of the time.

297
00:17:39,992 --> 00:17:43,310
And you heard about
ChatGPT hallucination,

298
00:17:43,310 --> 00:17:46,325
generating outputs based
on wrong information.

299
00:17:46,325 --> 00:17:48,800
And I'd also like to mention
a couple of articles

300
00:17:48,800 --> 00:17:54,159
that Microsoft had to take down what
they claimed were unsupervised

301
00:17:54,159 --> 00:17:58,100
AI-generated articles
on the travel website.

302
00:17:58,100 --> 00:18:00,720
One of the articles was recommendations

303
00:18:00,720 --> 00:18:05,890
for travelers visiting Ottawa,
in Canada, our capital city,

304
00:18:05,890 --> 00:18:08,658
and they suggested that you
got to visit the food bank

305
00:18:08,658 --> 00:18:10,289
with an empty stomach.

306
00:18:10,289 --> 00:18:13,992
And the second article
was a recommendation

307
00:18:13,992 --> 00:18:17,492
for visitors going to Montreal, in Canada,

308
00:18:17,492 --> 00:18:21,658
and one of the suggestions was
you got to try mouth-watering

309
00:18:21,658 --> 00:18:24,158
dishes such as McDonald's hamburger.

310
00:18:24,158 --> 00:18:27,325
So you got to be careful about how

311
00:18:27,325 --> 00:18:29,825
you're looking at the
AI-generated, outputs.

312
00:18:29,825 --> 00:18:35,658
Do your fact checking and judgment
as well, see if it makes sense.

313
00:18:35,658 --> 00:18:39,158
Because if you just use what is presented,

314
00:18:39,158 --> 00:18:42,825
you could make poor decisions and
potentially exposing the company

315
00:18:42,825 --> 00:18:46,325
to legal and non-compliance risks.

316
00:18:46,325 --> 00:18:50,991
Now, you talk about using
AI to generate content

317
00:18:50,991 --> 00:18:54,780
and incorporating part of
that into in-house tools.

318
00:18:54,780 --> 00:18:58,920
Using AI based on internal data set

319
00:18:58,920 --> 00:19:02,120
is probably somewhat, quote-unquote,

320
00:19:02,120 --> 00:19:06,170
could be more reliable
when you have quality data.

321
00:19:06,170 --> 00:19:08,825
But the same thing is
you need to make sure

322
00:19:08,825 --> 00:19:12,500
that what you fit in is reasonable.

323
00:19:12,500 --> 00:19:16,991
Also, check the performance of your
AI model, and there are metrics 

324
00:19:16,991 --> 00:19:19,380
out there that you can look at now,

325
00:19:19,380 --> 00:19:24,100
looking at the accuracy,
precision, and F1 score, et cetera.

326
00:19:24,100 --> 00:19:27,991
So you need to be careful of
how you are using that model.

327
00:19:27,991 --> 00:19:30,110
And if you look at a lot
of articles out there,

328
00:19:30,110 --> 00:19:33,658
now, they're talking about
companies are diving into AI

329
00:19:33,658 --> 00:19:39,658
but, yet, not all of them are deploying
them in a big scale, at this point.

330
00:19:39,658 --> 00:19:44,324
Because of the concerns
about the accuracy

331
00:19:44,324 --> 00:19:47,658
and how data could be misused,

332
00:19:47,658 --> 00:19:51,950
and also generating output that
could be misguiding decisions.

333
00:19:52,158 --> 00:19:53,760
– Yes, that's a really good point.

334
00:19:53,760 --> 00:19:56,991
Things to always keep in mind
when using any generative AI.

335
00:19:56,991 --> 00:19:58,658
Now, what if there's a listener

336
00:19:58,658 --> 00:20:00,158
listening to this conversation, right now,

337
00:20:00,158 --> 00:20:03,330
and they're like, "Connie, I've done
all the points that you've made.

338
00:20:03,330 --> 00:20:06,340
All the points you've made I've
implemented in my organization."

339
00:20:06,340 --> 00:20:08,491
Now, how can they
assess the effectiveness

340
00:20:08,491 --> 00:20:12,809
of this new data-driven culture that
they've created in their organization?

341
00:20:13,157 --> 00:20:15,324
– There are three things
they could look for

342
00:20:15,324 --> 00:20:19,020
to assess the effectiveness
of their data-driven culture.

343
00:20:19,020 --> 00:20:24,491
The first thing is
enthusiasm around analytics.

344
00:20:24,491 --> 00:20:28,991
Are people asking good questions to
verify observations, that's one thing.

345
00:20:28,991 --> 00:20:31,491
When people just share
data or share information,

346
00:20:31,491 --> 00:20:36,491
they ask for justification,
and verification for those.

347
00:20:36,491 --> 00:20:39,600
Are they asking good
questions to pinpoint problems?

348
00:20:39,600 --> 00:20:42,657
Are they getting clarity on work ideas?

349
00:20:42,657 --> 00:20:45,824
When your boss tell you that,
"Oh, we got to be efficient."

350
00:20:45,824 --> 00:20:50,150
And right away, if you hear someone
ask, "What do you mean by efficiency?

351
00:20:50,150 --> 00:20:51,850
Can you be more specific about it?"

352
00:20:51,850 --> 00:20:54,157
Because once you hone
in on those specifics,

353
00:20:54,157 --> 00:20:58,324
it will help you to identify,
"Ah, you're talking about speed

354
00:20:58,324 --> 00:21:01,157
of the process or errors
that we're making

355
00:21:01,157 --> 00:21:05,490
that will help you to identify data to
capture, and therefore, there are KPIs

356
00:21:05,510 --> 00:21:09,324
that you need to hone in
for doing your analysis.

357
00:21:09,324 --> 00:21:13,520
The second thing you'll
look for is transparency.

358
00:21:13,520 --> 00:21:16,490
When you have a solid
data-driven culture,

359
00:21:16,490 --> 00:21:20,824
people are very receptive
to what the data present.

360
00:21:20,824 --> 00:21:23,490
They're very objective and impartial

361
00:21:23,490 --> 00:21:26,490
when it comes to interpreting the results,

362
00:21:26,490 --> 00:21:28,824
and they're ready to
share the information.

363
00:21:28,824 --> 00:21:30,830
No reservation about it, good and bad.

364
00:21:30,830 --> 00:21:33,657
Let's just look at it and be
open about discussing

365
00:21:33,657 --> 00:21:37,490
what that means and how
do we need to respond.

366
00:21:37,490 --> 00:21:41,350
The third thing you can look
for is that trust in each other.

367
00:21:41,350 --> 00:21:44,990
When people are very comfortable
in sharing results openly,

368
00:21:44,990 --> 00:21:47,657
and they're very forthcoming,
focusing on issues

369
00:21:47,657 --> 00:21:54,823
rather than personal attacks
or pinpointing blames.

370
00:21:54,823 --> 00:21:57,157
People when they're
not afraid to speak up,

371
00:21:57,157 --> 00:21:59,657
you can see that you
have a data-driven culture,

372
00:21:59,657 --> 00:22:05,323
that people are very forthcoming and,
in fact, collaborating well together.

373
00:22:05,823 --> 00:22:09,429
Now, you've also asked about
the second part of that question,

374
00:22:09,429 --> 00:22:13,490
the culture, whether, fostering
better decision-making.

375
00:22:13,490 --> 00:22:15,490
I would put the onus on the champion.

376
00:22:15,490 --> 00:22:18,200
We talked about the champion before.

377
00:22:18,200 --> 00:22:21,823
As the champion, he needs to reinforce

378
00:22:21,823 --> 00:22:27,820
that discipline is in place to turn data
into actions and improvements.

379
00:22:27,820 --> 00:22:32,323
He also needs to pay attention
to whether folks are committing

380
00:22:32,323 --> 00:22:37,490
to measurements, and analysis,

381
00:22:37,490 --> 00:22:42,670
and he will need to observe
how they make decisions.

382
00:22:42,670 --> 00:22:47,279
The champion also needs to monitor
if the capabilities are in place.

383
00:22:47,279 --> 00:22:51,656
You got to give people the tools
to do the work, track the impact.

384
00:22:51,656 --> 00:22:55,989
Be able to have the time
allocated to discuss the results,

385
00:22:55,989 --> 00:22:59,156
take action, and then
continue to monitor it,

386
00:22:59,156 --> 00:23:03,489
and I would make a comment on, this.

387
00:23:03,489 --> 00:23:07,280
Random improvements
are often short-lived,

388
00:23:07,280 --> 00:23:09,823
but evidence-based
improvements are sustainable

389
00:23:09,823 --> 00:23:14,989
because they indeed tackle a problem
that is important to the business.

390
00:23:15,890 --> 00:23:19,600
– That's really important, and as
you have this data-driven culture,

391
00:23:19,600 --> 00:23:22,520
and you'll be able to
see things more quickly.

392
00:23:22,520 --> 00:23:25,669
How important is it to swiftly
act on these new insights

393
00:23:25,669 --> 00:23:28,489
that you're gaining more quickly,

394
00:23:28,489 --> 00:23:30,650
as you're seeing the data
and seeing the big picture

395
00:23:30,650 --> 00:23:33,156
but in a better way than you were before?

396
00:23:33,540 --> 00:23:38,140
– It is super important because doing
the analytics is just part of the work.

397
00:23:38,140 --> 00:23:44,230
Turning that analytics into action
and follow-through is very important.

398
00:23:44,230 --> 00:23:48,360
And I'd like to share
a story on Alan Mulally

399
00:23:48,360 --> 00:23:52,810
who is the former CEO
of Ford Motor Company.

400
00:23:52,810 --> 00:23:57,440
When he joined Ford in
2006 and became the CEO,

401
00:23:57,440 --> 00:24:02,610
Ford had lost $17 billion in
the previous fiscal year.

402
00:24:02,610 --> 00:24:03,989
And over the course of eight years,

403
00:24:03,989 --> 00:24:08,322
what he had done was he turned
the weekly executive team meeting

404
00:24:08,322 --> 00:24:10,822
into a collaboration exercise.

405
00:24:10,822 --> 00:24:13,989
Executives will come to the
meeting with the numbers,

406
00:24:13,989 --> 00:24:17,580
with the issues, table it openly,

407
00:24:17,580 --> 00:24:24,322
and ask for advice and insight ideas
on what they can do about them.

408
00:24:24,322 --> 00:24:27,590
That's a big contrast to his predecessor.

409
00:24:27,590 --> 00:24:30,909
What it used to happen is the expectation

410
00:24:30,909 --> 00:24:34,989
was, "You don't come into
this executive team meeting

411
00:24:34,989 --> 00:24:37,309
without a solution to your problem."

412
00:24:37,309 --> 00:24:41,155
So what happened then is there's
no incentive to share issues

413
00:24:41,155 --> 00:24:46,822
and that's really forcing, in a way,
guiding people to work in silos.

414
00:24:46,822 --> 00:24:51,322
When Alan had his first
executive team meeting,

415
00:24:51,322 --> 00:24:53,655
after he began to CEO, he was shocked

416
00:24:53,655 --> 00:24:58,110
when he looked at the dashboards
that folks brought to the meeting.

417
00:24:58,110 --> 00:25:01,155
There were hardly any red lights.

418
00:25:01,155 --> 00:25:03,100
If you think about the dashboards;

419
00:25:03,100 --> 00:25:06,155
the green light, red lights, there
were hardly any red lights.

420
00:25:06,155 --> 00:25:08,655
And the first question he
posed to the team was,

421
00:25:08,655 --> 00:25:13,140
"Folks, we know the
company is losing money.

422
00:25:13,140 --> 00:25:17,988
How can we only have a few
red lights on this dashboard?"

423
00:25:17,988 --> 00:25:20,655
So you can see that he really
turned the company around

424
00:25:20,655 --> 00:25:25,155
when he exercised the regiment
of come and bring the results, 

425
00:25:25,155 --> 00:25:29,488
whatever it is, green, red,
yellow, bring them all in.

426
00:25:29,560 --> 00:25:31,655
You just need to identify
what the issues are.

427
00:25:31,655 --> 00:25:34,988
If you have some ideas on what
you're going to do with them,

428
00:25:34,988 --> 00:25:37,488
let's share them, openly, with the team.

429
00:25:37,488 --> 00:25:39,940
Others will have ideas, or experience,

430
00:25:39,940 --> 00:25:44,190
or people with a skill set that
will be able to offer some help.

431
00:25:44,190 --> 00:25:46,820
So he really changed that whole culture,

432
00:25:46,820 --> 00:25:52,321
driving the data-driven culture home
by actively promoting that every week.

433
00:25:52,321 --> 00:25:58,840
So big kudo to him, when he
retired from Ford in 2014,

434
00:25:58,840 --> 00:26:00,920
Ford was a money-making machine.

435
00:26:00,920 --> 00:26:05,279
It had a profit of $7 billion when he retired.

436
00:26:05,279 --> 00:26:07,321
So that speaks volume to his leadership

437
00:26:07,321 --> 00:26:09,659
and how he changed that
whole culture around.

438
00:26:09,659 --> 00:26:14,960
So it also illustrates that, yes,
you do the analytics is one part.

439
00:26:14,960 --> 00:26:18,155
But getting the folks together
to talk about the results openly,

440
00:26:18,155 --> 00:26:22,870
no hidden agenda, "Let's be open and
honest about it, what's happening?"

441
00:26:22,870 --> 00:26:25,488
And let's solve and
tackle the issues together.

442
00:26:25,488 --> 00:26:27,154
– That's so important,

443
00:26:27,154 --> 00:26:29,488
and some people call those
the fierce conversations.

444
00:26:29,488 --> 00:26:32,960
Those conversations that
may make you uncomfortable,

445
00:26:32,960 --> 00:26:36,488
but they're super important
to having open and honest, 

446
00:26:36,488 --> 00:26:39,040
and making your organization successful.

447
00:26:39,321 --> 00:26:43,654
– Yes, definitely, because you can
only do so much doing analytics.

448
00:26:43,654 --> 00:26:46,539
And, yes, you can have
a center of excellence,

449
00:26:46,539 --> 00:26:50,488
building intelligence in
this particular group.

450
00:26:50,488 --> 00:26:54,821
But if you don't have that arena
for people to talk about it openly

451
00:26:54,821 --> 00:26:58,630
and share the information,
it's not going to help a whole lot.

452
00:26:58,821 --> 00:27:00,988
– Well, this has been a
wonderful conversation.

453
00:27:00,988 --> 00:27:02,720
Thank you so much for
sharing your insights

454
00:27:02,720 --> 00:27:05,700
and the importance of having
that data-driven culture, Connie.

455
00:27:05,700 --> 00:27:07,870
Thank you so much for
coming back on the podcast.

456
00:27:07,870 --> 00:27:10,321
– Happy to be here, and
thanks for having me.

457
00:27:10,321 --> 00:27:12,487
< Outro >

458
00:27:12,487 --> 00:27:13,987
– This has been Count Me In,

459
00:27:13,987 --> 00:27:18,154
IMA's podcast, providing you with the
latest perspectives of thought leaders,

460
00:27:18,154 --> 00:27:20,154
from the accounting
and finance profession.

461
00:27:20,154 --> 00:27:22,770
If you like what you heard,
and you'd like to be counted in

462
00:27:22,770 --> 00:27:25,154
for more relevant accounting
and finance education,

463
00:27:25,154 --> 00:27:31,654
visit IMA's website at www.imanet.org.