WEBVTT

1
00:00:00.120 --> 00:00:03.000
<v Speaker 1>Have you ever felt like you're just drowning in data?

2
00:00:03.720 --> 00:00:07.480
<v Speaker 1>You know, numbers everywhere, charts flashing, reports piling up, but

3
00:00:07.759 --> 00:00:10.119
<v Speaker 1>you're still thirsty for actual insight.

4
00:00:10.199 --> 00:00:15.359
<v Speaker 2>Oh. Absolutely, information overload, but insight starvation. It's a common feeling, exactly.

5
00:00:15.439 --> 00:00:18.039
<v Speaker 1>It's like you're collecting and collecting, but the real aha

6
00:00:18.440 --> 00:00:23.039
<v Speaker 1>moments feel just out of reach. That clarity you're craving

7
00:00:23.120 --> 00:00:26.199
<v Speaker 1>just gets lost in the noise. Well, today we're going

8
00:00:26.239 --> 00:00:28.160
<v Speaker 1>to dive deep into a guy that promises to help

9
00:00:28.239 --> 00:00:32.840
<v Speaker 1>cut through that clutter. Welcome to the deep dive. Our

10
00:00:32.880 --> 00:00:37.159
<v Speaker 1>focus is Mastering Tableau twenty twenty one implement Advanced Business

11
00:00:37.159 --> 00:00:39.200
<v Speaker 1>Intelligence by Meyer and Baldwin.

12
00:00:39.399 --> 00:00:41.640
<v Speaker 2>Right, and this isn't just a click here, do that

13
00:00:41.719 --> 00:00:45.679
<v Speaker 2>kind of book. It's about unlocking Tableau's real potential way

14
00:00:45.679 --> 00:00:47.039
<v Speaker 2>beyond basic dashboards.

15
00:00:47.119 --> 00:00:49.960
<v Speaker 1>Definitely. So our mission today is pretty clear. We want

16
00:00:50.000 --> 00:00:53.280
<v Speaker 1>to unpack how Tableau can take raw, sometimes really messy

17
00:00:53.359 --> 00:00:56.000
<v Speaker 1>data and turn it into genuinely powerful.

18
00:00:55.560 --> 00:00:58.520
<v Speaker 2>Stories, stories that communicate insights effectively.

19
00:00:58.640 --> 00:01:00.960
<v Speaker 1>Yes, and we'll look at making sure those stories perform well.

20
00:01:01.159 --> 00:01:02.560
<v Speaker 1>Nobody likes a slow dashboard.

21
00:01:02.679 --> 00:01:03.960
<v Speaker 2>H performance, it's key.

22
00:01:04.120 --> 00:01:07.480
<v Speaker 1>And then we'll get into some pretty cool surprising ways

23
00:01:07.519 --> 00:01:11.799
<v Speaker 1>to push what tableau can actually do. So let's start

24
00:01:11.799 --> 00:01:16.480
<v Speaker 1>with this idea of visual storytelling. Why is it more

25
00:01:16.519 --> 00:01:17.519
<v Speaker 1>than just pretty charts?

26
00:01:18.040 --> 00:01:21.439
<v Speaker 2>Well, what's really fascinating, and the book highlights this well,

27
00:01:21.640 --> 00:01:25.120
<v Speaker 2>is how these seemingly small design choices can make a

28
00:01:25.280 --> 00:01:29.079
<v Speaker 2>huge difference. It's, you know, the difference between your audience

29
00:01:29.120 --> 00:01:32.239
<v Speaker 2>instantly getting your point or just getting completely lost in

30
00:01:32.560 --> 00:01:33.799
<v Speaker 2>like visual clutter.

31
00:01:34.200 --> 00:01:36.519
<v Speaker 1>That's so true. We've all seen those dashboards that are

32
00:01:36.640 --> 00:01:40.200
<v Speaker 1>just a mess of rainbow explosion or something. The authors

33
00:01:40.239 --> 00:01:45.040
<v Speaker 1>give some really practical advice to avoid that, like reducing clutter.

34
00:01:45.480 --> 00:01:50.400
<v Speaker 1>First step, keep font choices simple. Sounds basic, Maybe.

35
00:01:50.120 --> 00:01:52.280
<v Speaker 2>It does, but it's often overlooked when people try to

36
00:01:52.319 --> 00:01:55.400
<v Speaker 2>get too creative. Simple is usually clearer, right.

37
00:01:55.680 --> 00:01:57.840
<v Speaker 1>And they also talk about lines like gridlines.

38
00:01:58.040 --> 00:02:00.319
<v Speaker 2>Yeah, use them as sparingly. They should be the most

39
00:02:00.439 --> 00:02:03.040
<v Speaker 2>muted thing on the chart, really faint, or as the

40
00:02:03.040 --> 00:02:04.959
<v Speaker 2>book says, you might even get rid of them altogether

41
00:02:05.040 --> 00:02:08.319
<v Speaker 2>if they're not helping. And another neat trick, for say

42
00:02:08.680 --> 00:02:12.879
<v Speaker 2>tall tables or lots of horizontal bars, use subtle bands

43
00:02:12.879 --> 00:02:15.599
<v Speaker 2>of color, maybe in groups of three to five.

44
00:02:15.520 --> 00:02:18.360
<v Speaker 1>Rows, like shading alternate groups exactly.

45
00:02:18.439 --> 00:02:21.000
<v Speaker 2>It helps guide the eye down The list segments things

46
00:02:21.080 --> 00:02:23.199
<v Speaker 2>nicely without adding more lines or noise.

47
00:02:23.680 --> 00:02:27.039
<v Speaker 1>It's clever. Okay, so less clutter. What about color? That's

48
00:02:27.080 --> 00:02:28.719
<v Speaker 1>another place things can go wrong fast?

49
00:02:28.759 --> 00:02:33.120
<v Speaker 2>Oh definitely. The book really hammers this. Use color intelligently,

50
00:02:33.400 --> 00:02:36.639
<v Speaker 2>keep it simple, keep it limited. Limited, maybe three to

51
00:02:36.800 --> 00:02:41.759
<v Speaker 2>five main hue variations. They quote Alberto Cairo actually pointing

52
00:02:41.800 --> 00:02:45.240
<v Speaker 2>out our visual working memories pretty limited. Too many colors

53
00:02:45.280 --> 00:02:46.240
<v Speaker 2>just overload us.

54
00:02:46.479 --> 00:02:48.639
<v Speaker 1>Makes sense. Our brains can only track so much.

55
00:02:48.479 --> 00:02:50.560
<v Speaker 2>At once, exactly. And you need to think about the

56
00:02:50.560 --> 00:02:54.319
<v Speaker 2>psychology of color. You know, red often means stop or bad,

57
00:02:54.479 --> 00:02:57.879
<v Speaker 2>Green means go or good. Use those conventions and.

58
00:02:57.840 --> 00:02:58.960
<v Speaker 1>Be colorblind friendly.

59
00:02:59.039 --> 00:03:02.560
<v Speaker 2>That's crucially critical for accessibility, and make sure everyone can

60
00:03:02.599 --> 00:03:05.560
<v Speaker 2>read your chart. And they also say use pure, really

61
00:03:05.599 --> 00:03:09.639
<v Speaker 2>bright colors sparingly. Well. If everything is highlighted, then nothing

62
00:03:09.680 --> 00:03:12.840
<v Speaker 2>is highlighted, right. Too many bright spots make it hard

63
00:03:12.840 --> 00:03:15.120
<v Speaker 2>to focus on the key message.

64
00:03:14.719 --> 00:03:17.759
<v Speaker 1>Got it? So use bright colors for emphasis only precisely.

65
00:03:17.960 --> 00:03:21.319
<v Speaker 2>And here's a subtle one. Choose color variations over different

66
00:03:21.360 --> 00:03:22.879
<v Speaker 2>shapes or symbols if you can.

67
00:03:23.280 --> 00:03:23.520
<v Speaker 1>Oh.

68
00:03:23.639 --> 00:03:26.759
<v Speaker 2>Yeah, apparently our brains find it easier and faster to

69
00:03:26.800 --> 00:03:30.639
<v Speaker 2>distinguish shades of color than to decode what different symbols mean.

70
00:03:31.159 --> 00:03:33.439
<v Speaker 2>It reduces that cognitive load we talked about.

71
00:03:33.680 --> 00:03:36.680
<v Speaker 1>Okay, less mental work for the audience, that's the goal.

72
00:03:36.960 --> 00:03:39.000
<v Speaker 1>And what about choose the right type of chart? They

73
00:03:39.039 --> 00:03:40.000
<v Speaker 1>mentioned pie charts?

74
00:03:40.120 --> 00:03:42.960
<v Speaker 2>Ah, yes, the infamous pie chart. The advice is pretty

75
00:03:42.960 --> 00:03:46.120
<v Speaker 2>clear use them sparingly.

76
00:03:46.039 --> 00:03:47.680
<v Speaker 1>Which many people don't, let's be honest.

77
00:03:47.879 --> 00:03:50.759
<v Speaker 2>True, but the book explains why they don't use screen

78
00:03:50.840 --> 00:03:54.240
<v Speaker 2>space very well, especially on rectangular screens, and it's often

79
00:03:54.319 --> 00:03:56.520
<v Speaker 2>genuinely hard to tell which slice is bigger if they're

80
00:03:56.560 --> 00:04:00.400
<v Speaker 2>close in size. Our eyes aren't great at comparing angles accurately,

81
00:04:00.719 --> 00:04:01.240
<v Speaker 2>so bar.

82
00:04:01.199 --> 00:04:03.599
<v Speaker 1>Charts are usually better for comparing amounts.

83
00:04:03.680 --> 00:04:08.240
<v Speaker 2>Almost always yes, for comparing magnitudes, bars are much clearer.

84
00:04:08.680 --> 00:04:11.639
<v Speaker 2>So putting it all together, these design tips aren't just

85
00:04:11.680 --> 00:04:13.280
<v Speaker 2>about making things look nice.

86
00:04:13.479 --> 00:04:18.040
<v Speaker 1>No, it's about effectiveness, making the data understandable and impactful exactly.

87
00:04:18.079 --> 00:04:19.480
<v Speaker 2>It elevates the insight itself.

88
00:04:19.560 --> 00:04:22.120
<v Speaker 1>Okay, so that's the visual side the art, But what

89
00:04:22.199 --> 00:04:26.199
<v Speaker 1>about the engine behind it? Tableau's magic isn't just skin deep.

90
00:04:26.519 --> 00:04:28.319
<v Speaker 1>How does it actually handle all that data?

91
00:04:28.360 --> 00:04:30.920
<v Speaker 2>Great question? And how do we make sure our dashboards

92
00:04:30.959 --> 00:04:32.399
<v Speaker 2>are fast, not just pretty.

93
00:04:32.639 --> 00:04:33.879
<v Speaker 1>Yeah, let's get under the hood.

94
00:04:34.120 --> 00:04:37.720
<v Speaker 2>Okay, So understanding Tableau's core engines is really key here.

95
00:04:38.279 --> 00:04:41.920
<v Speaker 2>The big one, especially in recent versions, is the Hyper

96
00:04:42.040 --> 00:04:45.600
<v Speaker 2>Data Handling Engine Hyper. Yeah. Hyper. Think of it as

97
00:04:45.639 --> 00:04:49.000
<v Speaker 2>a highly specialized database engine. It's designed to do several

98
00:04:49.040 --> 00:04:53.000
<v Speaker 2>things at once, really efficiently, general database stuff, data ingestion,

99
00:04:53.240 --> 00:04:55.399
<v Speaker 2>and analytics, all simultaneous.

100
00:04:55.480 --> 00:04:59.800
<v Speaker 1>WHOA okay, simultaneous. How does it manage that without slowing down?

101
00:05:00.079 --> 00:05:03.000
<v Speaker 2>It's incredibly efficient. Actually, the book mentions it can use

102
00:05:03.000 --> 00:05:06.040
<v Speaker 2>like ninety nine percent of available CPUs. It uses this

103
00:05:06.120 --> 00:05:11.360
<v Speaker 2>technique called morsel driven parallelization. Imagine breaking a huge task

104
00:05:11.480 --> 00:05:15.000
<v Speaker 2>into tiny little pieces or morsels, and giving each piece

105
00:05:15.079 --> 00:05:16.920
<v Speaker 2>to a different worker to do at the same time,

106
00:05:17.360 --> 00:05:21.160
<v Speaker 2>like a super organized team exactly so it can handle transactions,

107
00:05:21.319 --> 00:05:24.240
<v Speaker 2>load new data, and run complex queries all at once

108
00:05:24.360 --> 00:05:25.519
<v Speaker 2>incredibly fast.

109
00:05:25.680 --> 00:05:27.879
<v Speaker 1>That sounds amazing for anyone who's stared at a loading

110
00:05:27.879 --> 00:05:31.519
<v Speaker 1>spinner for too long. So Hyper processes the data, then

111
00:05:31.600 --> 00:05:35.639
<v Speaker 1>how does Tableau actually turn that processed data into the

112
00:05:35.759 --> 00:05:36.399
<v Speaker 1>charts we see?

113
00:05:36.439 --> 00:05:40.480
<v Speaker 2>Ah, that's where VISCUOL comes in VIZQL Visual Query Language.

114
00:05:40.519 --> 00:05:40.839
<v Speaker 1>Okay.

115
00:05:40.959 --> 00:05:44.279
<v Speaker 2>Viscal is the component that translates your actions like dragging

116
00:05:44.319 --> 00:05:47.839
<v Speaker 2>and dropping fields into data queries and then renders the

117
00:05:47.920 --> 00:05:51.079
<v Speaker 2>visual results the charts. And it does this entirely in memory.

118
00:05:51.160 --> 00:05:52.800
<v Speaker 1>In memory, so it's fast, very.

119
00:05:52.600 --> 00:05:55.439
<v Speaker 2>Fast, and super flexible. The real power is that you,

120
00:05:55.759 --> 00:05:59.240
<v Speaker 2>the analyst, can change the underlying query just by say,

121
00:05:59.680 --> 00:06:03.319
<v Speaker 2>drag digging a field from the measures area to the dimensions.

122
00:06:02.800 --> 00:06:05.199
<v Speaker 1>Area without writing SQL code exactly.

123
00:06:05.319 --> 00:06:08.399
<v Speaker 2>Viskalel rewrites the query for you on the fly. It

124
00:06:08.439 --> 00:06:12.399
<v Speaker 2>allows for that really fluid, spontaneous exploration of data. You

125
00:06:12.399 --> 00:06:15.079
<v Speaker 2>see something interesting, you drag a field and boom new you.

126
00:06:15.279 --> 00:06:17.759
<v Speaker 1>That drag and drop power is definitely a huge part

127
00:06:17.800 --> 00:06:23.000
<v Speaker 1>of Tableau's appeal. But okay, powerful engine, great visualization tool.

128
00:06:23.600 --> 00:06:25.199
<v Speaker 1>What if the data you feed it is.

129
00:06:25.600 --> 00:06:29.759
<v Speaker 2>Well a mess ah, the perennial problem, garbage in, garbage

130
00:06:29.800 --> 00:06:30.560
<v Speaker 2>out right right?

131
00:06:30.839 --> 00:06:32.879
<v Speaker 1>What does the book say about cleaning up that raw

132
00:06:33.000 --> 00:06:35.800
<v Speaker 1>data before it even gets to hyper or viscoel.

133
00:06:35.959 --> 00:06:39.319
<v Speaker 2>That's where Tableau Prep Builder comes into play. It's described

134
00:06:39.319 --> 00:06:42.759
<v Speaker 2>as a newer member of the Tableau family specifically designed

135
00:06:42.800 --> 00:06:46.360
<v Speaker 2>for data preparation. Prep Builder, and the great thing about

136
00:06:46.360 --> 00:06:49.199
<v Speaker 2>prep is that it's visual, just like Tableau Desktop. You

137
00:06:49.199 --> 00:06:51.800
<v Speaker 2>don't write scripts in the dark. You actually see each

138
00:06:51.839 --> 00:06:54.959
<v Speaker 2>step of your data cleaning process laid out visually.

139
00:06:54.720 --> 00:06:57.639
<v Speaker 1>So you can literally watch the data transform like a flow.

140
00:06:57.519 --> 00:07:01.680
<v Speaker 2>Chart pretty much. Yeah, you connect steps for cleaning, joining tables, pivoting,

141
00:07:01.759 --> 00:07:05.079
<v Speaker 2>data aggregating, even running scripts if you need to. It

142
00:07:05.120 --> 00:07:07.759
<v Speaker 2>makes that whole process much more transparent and intuitive.

143
00:07:08.079 --> 00:07:11.399
<v Speaker 1>That sounds incredibly useful because data prep can take forever.

144
00:07:11.600 --> 00:07:14.639
<v Speaker 2>It really can. The book mentions data prep can account

145
00:07:14.639 --> 00:07:18.439
<v Speaker 2>for up to sixty percent, zero percent of the entire

146
00:07:18.519 --> 00:07:19.920
<v Speaker 2>data mining effort sometimes.

147
00:07:19.959 --> 00:07:22.759
<v Speaker 1>Wow, okay, so streamlining that with a visual tool like

148
00:07:22.800 --> 00:07:23.759
<v Speaker 1>prep builder.

149
00:07:23.639 --> 00:07:27.800
<v Speaker 2>Is huge, absolutely invaluable, says massive amounts of time, potential errors.

150
00:07:27.839 --> 00:07:30.160
<v Speaker 1>All right, So let's assume we've used prep, We've got

151
00:07:30.199 --> 00:07:33.279
<v Speaker 1>clean data, we've got hyper and visquil working their magic.

152
00:07:33.519 --> 00:07:37.920
<v Speaker 1>We're building our dashboard and Tableau desktop. But it's still slow.

153
00:07:38.079 --> 00:07:40.079
<v Speaker 1>Users are complaining. What can we do? Then?

154
00:07:40.240 --> 00:07:45.879
<v Speaker 2>Yeah, performance tuning crucial step. The book offers several good strategies.

155
00:07:45.959 --> 00:07:48.879
<v Speaker 2>One big one is using data extracts effectively.

156
00:07:49.000 --> 00:07:51.079
<v Speaker 1>Extras like saving a local copy.

157
00:07:50.879 --> 00:07:54.120
<v Speaker 2>Of the data exactly. Tableau extracts are often much faster

158
00:07:54.199 --> 00:07:56.759
<v Speaker 2>to query than connecting live to a database, especially a

159
00:07:56.759 --> 00:08:00.920
<v Speaker 2>complex one. The book points out that extracts are always flattened.

160
00:08:01.000 --> 00:08:02.079
<v Speaker 1>Flattened meaning what.

161
00:08:02.240 --> 00:08:05.920
<v Speaker 2>It means Tableau pre processes the data, handling any joins

162
00:08:05.959 --> 00:08:09.879
<v Speaker 2>or complexities, and stores it in a single optimized table

163
00:08:09.879 --> 00:08:13.680
<v Speaker 2>structure within the extract file. This makes querying super fast.

164
00:08:14.000 --> 00:08:18.600
<v Speaker 2>The advice is maybe start building your dashboard using a

165
00:08:18.600 --> 00:08:21.199
<v Speaker 2>small extract, just a sample of the data, so it's

166
00:08:21.240 --> 00:08:23.279
<v Speaker 2>really responsive while you design.

167
00:08:23.160 --> 00:08:25.800
<v Speaker 1>Ah, so you're not waiting for the full data set

168
00:08:25.839 --> 00:08:27.879
<v Speaker 1>to load every time you tweak something right.

169
00:08:28.160 --> 00:08:30.920
<v Speaker 2>Then once you're happy with the design, you can point

170
00:08:30.920 --> 00:08:33.559
<v Speaker 2>it to the full extract or the live connection if needed.

171
00:08:33.919 --> 00:08:37.480
<v Speaker 1>Smart What about filters they always seem to slow things down.

172
00:08:38.200 --> 00:08:40.279
<v Speaker 2>They can, but it depends on the type of filter.

173
00:08:40.360 --> 00:08:43.159
<v Speaker 2>This is a key distinction the book makes okay. Dimension

174
00:08:43.240 --> 00:08:46.720
<v Speaker 2>filters and measure filters, the most common types, actually improve

175
00:08:46.799 --> 00:08:50.240
<v Speaker 2>performance usually now because they limit the amount of data

176
00:08:50.320 --> 00:08:53.679
<v Speaker 2>pulled from the source before Tableau does the visualization work.

177
00:08:54.360 --> 00:08:56.600
<v Speaker 2>Less data processed means faster results.

178
00:08:56.759 --> 00:08:57.279
<v Speaker 1>Makes sense.

179
00:08:57.320 --> 00:09:01.240
<v Speaker 2>But there's another type called table calculation filter. These are different.

180
00:09:01.360 --> 00:09:03.559
<v Speaker 2>They run after all the data has been retrieved and

181
00:09:03.559 --> 00:09:04.720
<v Speaker 2>calculated AH.

182
00:09:04.639 --> 00:09:06.879
<v Speaker 1>So they don't reduce the initial data load exactly.

183
00:09:06.919 --> 00:09:10.000
<v Speaker 2>They just hide marks that are already calculated, so they're

184
00:09:10.159 --> 00:09:12.960
<v Speaker 2>useful for analysis, but they don't give you that initial

185
00:09:13.000 --> 00:09:15.919
<v Speaker 2>performance boost. Knowing the difference is important.

186
00:09:16.080 --> 00:09:19.279
<v Speaker 1>Definitely good tip. Anything else on a dashboard design itself

187
00:09:19.279 --> 00:09:20.039
<v Speaker 1>for speed.

188
00:09:20.039 --> 00:09:24.120
<v Speaker 2>Yeah, keep it simple, basically, avoid overcrowding every single chart.

189
00:09:24.159 --> 00:09:25.879
<v Speaker 2>Every filter adds to the calculation.

190
00:09:25.960 --> 00:09:28.759
<v Speaker 1>Load less it's more again pretty much.

191
00:09:29.159 --> 00:09:31.759
<v Speaker 2>Also, fixing the dashboard size setting it to a specific

192
00:09:31.799 --> 00:09:35.279
<v Speaker 2>resolution rather than automatic can help prevent Tableaux from having

193
00:09:35.320 --> 00:09:39.639
<v Speaker 2>to constantly recalculate layouts for different screens. Okay, and maybe

194
00:09:39.679 --> 00:09:44.600
<v Speaker 2>software advice, but important. Set expectations with your users. If

195
00:09:44.600 --> 00:09:47.080
<v Speaker 2>a complex dashboard will take a few seconds to load,

196
00:09:47.320 --> 00:09:47.759
<v Speaker 2>let them know.

197
00:09:48.039 --> 00:09:51.480
<v Speaker 1>Manage expectations. Yeah. And hardware does it matter much?

198
00:09:51.600 --> 00:09:52.440
<v Speaker 2>It does play a role.

199
00:09:52.519 --> 00:09:53.000
<v Speaker 1>Yeah.

200
00:09:53.120 --> 00:09:55.639
<v Speaker 2>The book mentions things like having enough ram at least

201
00:09:55.679 --> 00:09:58.639
<v Speaker 2>eight gigle bay for Mac for example, and a decent

202
00:09:58.720 --> 00:10:02.279
<v Speaker 2>graphics card like an GPU can help with rendering speed

203
00:10:02.360 --> 00:10:06.399
<v Speaker 2>using accelerated graphics. And one for the developers, remember the

204
00:10:06.480 --> 00:10:09.440
<v Speaker 2>run update command F nine on Windows. You can pause

205
00:10:09.480 --> 00:10:12.039
<v Speaker 2>automatic updates, make a bunch of changes, and then hit

206
00:10:12.240 --> 00:10:15.080
<v Speaker 2>F nine to refresh only when you're ready. Saves constant

207
00:10:15.120 --> 00:10:15.600
<v Speaker 2>re rendering.

208
00:10:15.679 --> 00:10:19.960
<v Speaker 1>Nice practical tips. Okay, so we've covered design, data prep,

209
00:10:20.159 --> 00:10:24.279
<v Speaker 1>the engine performance. Let's push further. How does Tableau help

210
00:10:24.360 --> 00:10:27.039
<v Speaker 1>us get deeper insights beyond the basics.

211
00:10:27.360 --> 00:10:30.159
<v Speaker 2>Yeah, moving into more advanced analytics. It's one thing to

212
00:10:30.200 --> 00:10:33.039
<v Speaker 2>see totals or averages, but what if you need to

213
00:10:33.080 --> 00:10:37.360
<v Speaker 2>compare a value to say, a category average that isn't

214
00:10:37.360 --> 00:10:37.960
<v Speaker 2>even shown on.

215
00:10:38.000 --> 00:10:40.679
<v Speaker 1>Your charge exactly? Those more complex comparisons.

216
00:10:40.799 --> 00:10:44.840
<v Speaker 2>That's where level of detailed calculations or LODs come in.

217
00:10:45.159 --> 00:10:46.960
<v Speaker 2>They are incredibly powerful.

218
00:10:47.039 --> 00:10:49.320
<v Speaker 1>Alu D's I've heard of those, but they can seem

219
00:10:49.360 --> 00:10:50.200
<v Speaker 1>a bit intimidating.

220
00:10:50.480 --> 00:10:53.159
<v Speaker 2>They can be at first. Yeah, but the core idea

221
00:10:53.240 --> 00:10:57.080
<v Speaker 2>is this. Normally Tableau calculates measures based only on the

222
00:10:57.080 --> 00:10:59.559
<v Speaker 2>dimensions you have physically dragged into your view.

223
00:10:59.639 --> 00:11:02.799
<v Speaker 1>Right, sales per region if region is in the view exactly.

224
00:11:03.120 --> 00:11:07.039
<v Speaker 2>But LOD calculations, let you Tellcablo hey for this calculation,

225
00:11:07.360 --> 00:11:09.519
<v Speaker 2>calculate it at a different level in detail than what's

226
00:11:09.559 --> 00:11:10.200
<v Speaker 2>currently shown.

227
00:11:10.279 --> 00:11:13.240
<v Speaker 1>Ah, so you can override the default behavior precisely.

228
00:11:13.440 --> 00:11:16.720
<v Speaker 2>The book gives a handyamonic for the three main types fixed,

229
00:11:16.960 --> 00:11:21.720
<v Speaker 2>include and exclude. Fixed lets you specify exactly which dimensions

230
00:11:21.759 --> 00:11:25.120
<v Speaker 2>to use for the calculation, regardless of the view. In

231
00:11:25.240 --> 00:11:27.679
<v Speaker 2>slury d lets you add dimensions to the calculation that

232
00:11:27.720 --> 00:11:31.559
<v Speaker 2>aren't in the view, and excluid lets you remove dimensions

233
00:11:31.600 --> 00:11:33.200
<v Speaker 2>from the calculation that are in the view.

234
00:11:34.000 --> 00:11:37.399
<v Speaker 1>Okay, so you can do things like calculate the total

235
00:11:37.440 --> 00:11:40.720
<v Speaker 1>sales for an entire product category and show that value

236
00:11:40.759 --> 00:11:44.320
<v Speaker 1>next to the sales for each individual subcategory.

237
00:11:43.679 --> 00:11:46.600
<v Speaker 2>Exactly even if the main category isn't a dimension in

238
00:11:46.639 --> 00:11:50.159
<v Speaker 2>your chart. Or you could calculate the average customer order

239
00:11:50.279 --> 00:11:53.720
<v Speaker 2>value ignoring the specific products in the view. It unlocks

240
00:11:53.799 --> 00:11:55.720
<v Speaker 2>really sophisticated analysis that.

241
00:11:55.720 --> 00:11:59.360
<v Speaker 1>Sounds super flexible, a real game changer for asking complex questions,

242
00:11:59.399 --> 00:12:03.480
<v Speaker 1>definitely and beyond calculations within Tableau itself. The book talks

243
00:12:03.519 --> 00:12:07.159
<v Speaker 1>about extending Tableau's capabilities, right, Yeah, connecting it to other tools.

244
00:12:07.200 --> 00:12:10.120
<v Speaker 2>Absolutely. Tableau isn't just a closed box anymore. It's becoming

245
00:12:10.120 --> 00:12:12.840
<v Speaker 2>more of an open platform. One way is through the

246
00:12:12.840 --> 00:12:15.120
<v Speaker 2>Tableau Extensions API API.

247
00:12:15.200 --> 00:12:16.519
<v Speaker 1>So connecting other software.

248
00:12:16.639 --> 00:12:19.360
<v Speaker 2>Yeah, it allows third party developers or even your own

249
00:12:19.360 --> 00:12:23.039
<v Speaker 2>company to build tools that run directly inside a Tableau dashboard.

250
00:12:23.159 --> 00:12:24.440
<v Speaker 1>Like what kind of tools.

251
00:12:24.279 --> 00:12:27.759
<v Speaker 2>The book mentions things like a sinky chart extension. Sinki

252
00:12:27.799 --> 00:12:31.039
<v Speaker 2>diagrams are complex to build manually in Tableau, so an

253
00:12:31.080 --> 00:12:34.120
<v Speaker 2>extension can save a ton of time or maybe custom

254
00:12:34.200 --> 00:12:38.159
<v Speaker 2>rate that capabilities or specialized statistical analysis tools.

255
00:12:38.200 --> 00:12:41.759
<v Speaker 1>Okay, so extending the functionality. What about AI and machine learning?

256
00:12:42.320 --> 00:12:43.399
<v Speaker 1>That's everywhere now.

257
00:12:43.440 --> 00:12:47.000
<v Speaker 2>Right, and Tableau integrated that with Einstein Discovery, which the

258
00:12:47.039 --> 00:12:50.399
<v Speaker 2>book flags is a major feature introduced around the twenty

259
00:12:50.440 --> 00:12:51.919
<v Speaker 2>twenty one point one version.

260
00:12:51.720 --> 00:12:53.720
<v Speaker 1>Einstein Discovery from Salesforce.

261
00:12:53.879 --> 00:12:58.799
<v Speaker 2>Yeah, leveraging Salesforce's AI capabilities, it essentially brings built in

262
00:12:58.919 --> 00:13:02.799
<v Speaker 2>machine learning and real time prediction directly into your Tableau workflow.

263
00:13:02.960 --> 00:13:04.399
<v Speaker 1>How would that work in practice?

264
00:13:04.720 --> 00:13:07.639
<v Speaker 2>Well, imagine you have a supply chain dashboard. You could

265
00:13:07.679 --> 00:13:11.000
<v Speaker 2>point Einstein Discovery at your data and it might automatically

266
00:13:11.000 --> 00:13:14.080
<v Speaker 2>identify the key factors driving delivery times. Okay, and not

267
00:13:14.159 --> 00:13:17.039
<v Speaker 2>just identify them, but actually build a predictive model on

268
00:13:17.080 --> 00:13:20.120
<v Speaker 2>the fly and even suggest actions you could take based

269
00:13:20.120 --> 00:13:23.720
<v Speaker 2>on the model to improve shipment times. It's bringing predictive

270
00:13:23.799 --> 00:13:25.159
<v Speaker 2>power to the business user.

271
00:13:25.519 --> 00:13:29.759
<v Speaker 1>Wow, that is powerful. Moving from just reporting the past

272
00:13:30.320 --> 00:13:33.279
<v Speaker 1>to actively predicting and influencing the future.

273
00:13:33.440 --> 00:13:34.120
<v Speaker 2>That's the goal.

274
00:13:34.320 --> 00:13:37.559
<v Speaker 1>And for users who are comfortable with coding like data scientists,

275
00:13:37.799 --> 00:13:39.200
<v Speaker 1>can they bring their own models in.

276
00:13:39.440 --> 00:13:44.360
<v Speaker 2>Yes, definitely. Tableau has integrations with R and Python, two

277
00:13:44.360 --> 00:13:47.639
<v Speaker 2>of the most popular languages for data science. This lets

278
00:13:47.679 --> 00:13:51.919
<v Speaker 2>you run R or Python scripts directly from Tableau calculations,

279
00:13:52.159 --> 00:13:55.320
<v Speaker 2>so you can perform really advanced analyzes that go way

280
00:13:55.360 --> 00:13:57.399
<v Speaker 2>beyond Tableau's built in.

281
00:13:57.480 --> 00:13:58.879
<v Speaker 1>Function like what sort of things?

282
00:13:59.000 --> 00:14:02.600
<v Speaker 2>All sorts? The book examples like calculating sentiment analysis on

283
00:14:02.639 --> 00:14:05.559
<v Speaker 2>techt data. They actually use dialogue from Lord of the

284
00:14:05.600 --> 00:14:07.919
<v Speaker 2>Rings as a fun example data set for that.

285
00:14:08.200 --> 00:14:09.639
<v Speaker 1>Ah, really that's cool.

286
00:14:09.720 --> 00:14:14.600
<v Speaker 2>Yeah, Or generating random numbers for simulations, performing complex statistical tests,

287
00:14:14.879 --> 00:14:19.559
<v Speaker 2>running sophisticated machine learning models like linear regression for prediction. Basically,

288
00:14:19.600 --> 00:14:22.519
<v Speaker 2>anything you can code in R or Python you can

289
00:14:22.559 --> 00:14:25.000
<v Speaker 2>potentially integrate into your Tableau visualization.

290
00:14:25.360 --> 00:14:28.120
<v Speaker 1>So it really opens the door to highly specialized, cutting

291
00:14:28.200 --> 00:14:30.799
<v Speaker 1>edge analytics right within the Tableau environment.

292
00:14:31.080 --> 00:14:34.759
<v Speaker 2>It really does. It bridges the gap between visual analytics

293
00:14:34.960 --> 00:14:36.759
<v Speaker 2>and advanced statistical modeling.

294
00:14:36.919 --> 00:14:40.200
<v Speaker 1>Wow, what incredible journey we've been on today. We've really

295
00:14:40.200 --> 00:14:42.679
<v Speaker 1>covered a lot of ground. We certainly have started with

296
00:14:42.720 --> 00:14:46.200
<v Speaker 1>the art, you know, visual storytelling, making sure dashboards are clear,

297
00:14:46.480 --> 00:14:48.960
<v Speaker 1>clutter free, using color wisely mm hmm.

298
00:14:49.360 --> 00:14:50.600
<v Speaker 2>Getting those foundations right.

299
00:14:50.840 --> 00:14:53.519
<v Speaker 1>Then we looked under the hood at the engines Hyper

300
00:14:53.559 --> 00:14:57.240
<v Speaker 1>for speed, vizquil for flexibility so important.

301
00:14:57.360 --> 00:15:01.000
<v Speaker 2>Absolutely understanding how it works helps you use it better.

302
00:15:00.919 --> 00:15:04.840
<v Speaker 1>Than tackling messy data with Tableau prep builder. That visual

303
00:15:04.879 --> 00:15:06.159
<v Speaker 1>workflow seems.

304
00:15:05.879 --> 00:15:08.759
<v Speaker 2>Key, save so much time and headache.

305
00:15:08.440 --> 00:15:13.279
<v Speaker 1>And optimizing performance using extracts smartly understanding those filter types,

306
00:15:13.360 --> 00:15:14.799
<v Speaker 1>keeping dashboards lean.

307
00:15:15.000 --> 00:15:18.360
<v Speaker 2>Crucial for user adoption. Slow dashboards just don't get.

308
00:15:18.360 --> 00:15:21.679
<v Speaker 1>Used, definitely not. And finally diving into the really advanced

309
00:15:21.720 --> 00:15:26.000
<v Speaker 1>stuff led calculations for deeper analysis and pushing boundaries with

310
00:15:26.080 --> 00:15:30.440
<v Speaker 1>extensions Einstein Discovery, AI and even R and Python integration.

311
00:15:30.600 --> 00:15:32.919
<v Speaker 2>Yeah, it's quite the toolkit, it really is. You know,

312
00:15:33.039 --> 00:15:36.080
<v Speaker 2>if you connect all these pieces together, what we've really

313
00:15:36.080 --> 00:15:39.159
<v Speaker 2>described isn't just a list of software features. It's more

314
00:15:39.200 --> 00:15:42.559
<v Speaker 2>like a complete ecosystem. Ecosystem I like that. Yeah, an

315
00:15:42.600 --> 00:15:45.960
<v Speaker 2>ecosystem designed to empower you not just to see your data,

316
00:15:46.039 --> 00:15:49.639
<v Speaker 2>but to really understand it, clean it up, properly, analyze

317
00:15:49.639 --> 00:15:51.919
<v Speaker 2>it deeply, and even use it to predict what might

318
00:15:51.960 --> 00:15:52.679
<v Speaker 2>happen next.

319
00:15:52.919 --> 00:15:55.840
<v Speaker 1>So it moves you from just passively looking at reports

320
00:15:56.240 --> 00:15:59.440
<v Speaker 1>to actively discovering insights and making informed.

321
00:15:59.120 --> 00:16:03.519
<v Speaker 2>Decisions exactly, active data discovery, actionable intelligence. That's the promise.

322
00:16:04.000 --> 00:16:08.039
<v Speaker 1>So thinking about our listeners, what does all this mean

323
00:16:08.159 --> 00:16:11.000
<v Speaker 1>for you, the person wanting to get better with data?

324
00:16:11.440 --> 00:16:14.799
<v Speaker 1>The book itself says something interesting near the end that

325
00:16:15.000 --> 00:16:17.559
<v Speaker 1>data mining isn't really a project you finish.

326
00:16:17.600 --> 00:16:19.960
<v Speaker 2>Right says it does not cease upon the completion of

327
00:16:19.960 --> 00:16:22.879
<v Speaker 2>a particular project, but continues for the life of the business.

328
00:16:23.320 --> 00:16:25.240
<v Speaker 2>It's an ongoing process.

329
00:16:24.799 --> 00:16:27.679
<v Speaker 1>An ongoing process. So the question for you is what

330
00:16:27.879 --> 00:16:31.159
<v Speaker 1>data story is waiting inside your data? What insights are

331
00:16:31.200 --> 00:16:33.799
<v Speaker 1>you hoping to uncover, either for work or just out

332
00:16:33.799 --> 00:16:37.720
<v Speaker 1>of curiosity, And how can these powerful but as we've seen,

333
00:16:38.039 --> 00:16:41.320
<v Speaker 1>pretty accessible tableau features help you find that story, shape it,

334
00:16:41.679 --> 00:16:43.480
<v Speaker 1>and share it in a way that really makes an impact.

335
00:16:43.840 --> 00:16:44.759
<v Speaker 1>Something to think about