1
00:00:00,560 --> 00:00:05,120
Speaker 1: Well, after twenty five years of just googling stuff online

2
00:00:05,160 --> 00:00:07,720
Speaker 1: searches in the midst of a radical change.

3
00:00:08,000 --> 00:00:11,640
Speaker 2: Yeah, we're seeing the rise of conversational search and AI

4
00:00:11,800 --> 00:00:15,800
Speaker 2: generated search summaries, which have big implications for the seventy

5
00:00:15,800 --> 00:00:20,119
Speaker 2: billion dollar search engine optimization industry and the ability of

6
00:00:20,200 --> 00:00:23,200
Speaker 2: KIWI companies to find a global audience online.

7
00:00:23,239 --> 00:00:25,239
Speaker 1: This week on the Business of Tech, powered by two

8
00:00:25,239 --> 00:00:28,600
Speaker 1: Degrees Business, what businesses need to know about how search

9
00:00:28,720 --> 00:00:32,040
Speaker 1: is changing and how will measure audience engagement in the

10
00:00:32,080 --> 00:00:35,040
Speaker 1: world where clicks and impressions may mean a lot less

11
00:00:35,080 --> 00:00:35,839
Speaker 1: than they used to.

12
00:00:36,360 --> 00:00:39,440
Speaker 2: Our guest this week is Ryan McMillan from Atlas Digital,

13
00:00:39,720 --> 00:00:43,800
Speaker 2: a company that specializes in paid search and social media

14
00:00:44,240 --> 00:00:47,839
Speaker 2: marketing and advertising. The perfect guy to explain and make

15
00:00:48,000 --> 00:00:50,640
Speaker 2: sense off all of these changes that are coming to

16
00:00:50,760 --> 00:00:53,640
Speaker 2: search and what it means for the average QWI company

17
00:00:53,640 --> 00:00:55,440
Speaker 2: trying to make it on the global stage.

18
00:00:55,600 --> 00:00:58,120
Speaker 3: What I think we're going to see is probably less

19
00:00:58,200 --> 00:01:01,400
Speaker 3: use of a traditional search engine as people start to

20
00:01:01,400 --> 00:01:04,520
Speaker 3: find answers in different ways. Every year we hear something

21
00:01:04,560 --> 00:01:06,600
Speaker 3: that SEO is dead. Is SEO dead?

22
00:01:06,640 --> 00:01:08,920
Speaker 4: I'm not sure about. SEO is certainly a worth.

23
00:01:08,800 --> 00:01:11,600
Speaker 1: Changing stay tuned to find out everything you need to

24
00:01:11,680 --> 00:01:14,120
Speaker 1: know about the changing world of search engines.

25
00:01:14,400 --> 00:01:17,480
Speaker 2: First, though, if you hadn't noticed, we're about three months

26
00:01:17,520 --> 00:01:21,080
Speaker 2: away from the US presidential election and the miss and

27
00:01:21,280 --> 00:01:25,399
Speaker 2: disinformation is running thick and fast on social platforms like

28
00:01:25,720 --> 00:01:26,560
Speaker 2: x and Facebook.

29
00:01:26,800 --> 00:01:30,200
Speaker 1: It really moved up a gear following the assassination attempt

30
00:01:30,240 --> 00:01:34,640
Speaker 1: on former President Trump in Butler, Pennsylvania on July thirteenth.

31
00:01:35,240 --> 00:01:38,800
Speaker 2: Elin Musk's endorsement of Trump seemed to trigger a flood

32
00:01:39,040 --> 00:01:43,840
Speaker 2: of conservative slash Republican propaganda across his own platform and

33
00:01:43,959 --> 00:01:47,200
Speaker 2: counter offensives from liberal learning social media users.

34
00:01:47,360 --> 00:01:50,600
Speaker 1: Add in the escalating conflict in the Middle East, riots

35
00:01:50,600 --> 00:01:54,160
Speaker 1: in the UK, sweeping political change there, and the social

36
00:01:54,160 --> 00:01:57,080
Speaker 1: media scene seems like even more of a saspit than

37
00:01:57,120 --> 00:01:57,840
Speaker 1: it usually is.

38
00:01:58,160 --> 00:02:00,800
Speaker 2: So what's going on and what can we expect? Is

39
00:02:00,840 --> 00:02:03,920
Speaker 2: we approached that pivotal election when it comes to the

40
00:02:03,960 --> 00:02:06,000
Speaker 2: peddling of miss and disinformation.

41
00:02:06,320 --> 00:02:09,880
Speaker 1: I caught up with aut Associate professor in Communication Studies

42
00:02:09,880 --> 00:02:13,400
Speaker 1: and former BBC reporter doctor Helen Sissons to get some

43
00:02:13,440 --> 00:02:19,400
Speaker 1: perspective on what's going on. Hi, Helen, thank you for

44
00:02:19,480 --> 00:02:21,520
Speaker 1: joining us on the Business of Tech podcast.

45
00:02:21,760 --> 00:02:23,200
Speaker 5: It's a pleasure you've.

46
00:02:23,120 --> 00:02:27,280
Speaker 1: Joined us today to talk a little bit about misinformation

47
00:02:27,639 --> 00:02:32,160
Speaker 1: and how these new media technologies are playing a role

48
00:02:32,600 --> 00:02:36,880
Speaker 1: in the issue of disinformation and misinformation. And we saw

49
00:02:36,919 --> 00:02:41,160
Speaker 1: an explosion of that very recently with the attempted assassination

50
00:02:41,680 --> 00:02:43,200
Speaker 1: of Donald Trump.

51
00:02:43,320 --> 00:02:45,360
Speaker 5: Well, first of all, Ben, you're absl right. What we

52
00:02:45,400 --> 00:02:50,440
Speaker 5: saw post the assassination attempt was quite alarming, and it

53
00:02:50,480 --> 00:02:53,320
Speaker 5: has been obviously the subject of discussion, and we saw

54
00:02:53,320 --> 00:02:56,120
Speaker 5: some quite senior people obviously piling in really with some

55
00:02:56,160 --> 00:02:59,840
Speaker 5: really with some very concerning and not fact checked, no

56
00:03:00,040 --> 00:03:02,840
Speaker 5: evidence to support what they were saying, accusing various people

57
00:03:02,919 --> 00:03:05,160
Speaker 5: that came from both sides. We had, you know, obviously

58
00:03:05,160 --> 00:03:08,680
Speaker 5: the usual Deep State is involved in that actually was

59
00:03:08,880 --> 00:03:12,560
Speaker 5: a Donald Trump construction, and so you know, the idea

60
00:03:12,639 --> 00:03:14,720
Speaker 5: that the deep state is after him and has been

61
00:03:14,800 --> 00:03:18,080
Speaker 5: and he's the victim has become a sort of mantra

62
00:03:18,160 --> 00:03:21,480
Speaker 5: almost and we saw that within almost minutes of the

63
00:03:21,919 --> 00:03:26,240
Speaker 5: of the attempt. We also saw quite senior people pointing

64
00:03:26,240 --> 00:03:28,679
Speaker 5: the finger at Joe Biden, which of course is incendiary

65
00:03:28,800 --> 00:03:31,520
Speaker 5: and quite and very concerning. And then we saw on

66
00:03:31,560 --> 00:03:33,920
Speaker 5: the other side, oh, this has been a you know,

67
00:03:34,160 --> 00:03:37,280
Speaker 5: Trump's own supporters. They're doing this to get the sympathy

68
00:03:37,400 --> 00:03:41,080
Speaker 5: of people, and that again is very concerning because you know,

69
00:03:41,120 --> 00:03:44,520
Speaker 5: we see them both sides sort of piling in and

70
00:03:44,840 --> 00:03:46,960
Speaker 5: we're in. It's very concerning for the people. If we've

71
00:03:46,960 --> 00:03:49,440
Speaker 5: taken it from the people who were on social media,

72
00:03:49,440 --> 00:03:52,560
Speaker 5: who are not politicians, who are not activists, it begins

73
00:03:52,560 --> 00:03:55,840
Speaker 5: to break down their trust. It's very It is very

74
00:03:56,280 --> 00:03:59,320
Speaker 5: It stokes fear, It stokes concern, and it stokes who

75
00:03:59,400 --> 00:04:02,440
Speaker 5: can I trust? Who's everyone out there is behaving in

76
00:04:02,480 --> 00:04:05,440
Speaker 5: a way that makes it that makes me feel unsafe.

77
00:04:05,880 --> 00:04:10,120
Speaker 5: So it is really bad for everyone's a sense of

78
00:04:10,120 --> 00:04:13,800
Speaker 5: security within our democracies. And it has been coined by

79
00:04:13,920 --> 00:04:16,440
Speaker 5: one of the top researchers in this area, Claire Wardle,

80
00:04:16,480 --> 00:04:20,400
Speaker 5: who's now Brown University, as information disorder. And now we

81
00:04:20,440 --> 00:04:23,920
Speaker 5: always had obviously, rumors and misinformation have always been around,

82
00:04:24,080 --> 00:04:27,599
Speaker 5: it's just the level and so the ability of social

83
00:04:27,600 --> 00:04:30,360
Speaker 5: media to disseminate them in a way that hasn't been

84
00:04:30,400 --> 00:04:34,360
Speaker 5: possible before. Misinformation has always been around, it is because

85
00:04:34,400 --> 00:04:38,560
Speaker 5: of social media that it's now on steroids. And people believe,

86
00:04:39,160 --> 00:04:41,320
Speaker 5: you know, half of New Zealanders believe that they're seeing

87
00:04:41,360 --> 00:04:45,640
Speaker 5: misinformation every day. It's not necessarily the truth. But when

88
00:04:45,720 --> 00:04:49,039
Speaker 5: half New Zealanders think that they're seeing false information every

89
00:04:49,080 --> 00:04:52,640
Speaker 5: day and ninety percent think they're seeing it monthly, then

90
00:04:52,680 --> 00:04:56,320
Speaker 5: we've got a trust issue here with our information ecosystem.

91
00:04:56,600 --> 00:04:59,680
Speaker 1: Let's go to the issue of the moment. We talked

92
00:04:59,720 --> 00:05:03,520
Speaker 1: pret Flea about the Trump possessionation attempt and the resulting

93
00:05:03,920 --> 00:05:08,960
Speaker 1: frenzy of information following that. How does that kind of

94
00:05:09,040 --> 00:05:12,240
Speaker 1: encapsulate where we are today, where we're more likely to

95
00:05:12,240 --> 00:05:15,279
Speaker 1: believe whatever kind of comes through the social media channels,

96
00:05:15,800 --> 00:05:19,600
Speaker 1: and maybe even less likely to believe the factual reporting

97
00:05:19,640 --> 00:05:21,440
Speaker 1: that comes through mainstream media news.

98
00:05:22,279 --> 00:05:26,279
Speaker 5: As I said, you've always had this, the misinformation, the peddlers,

99
00:05:26,320 --> 00:05:27,839
Speaker 5: and you've always had But then actually they were sitting

100
00:05:27,839 --> 00:05:30,640
Speaker 5: in the pub talking to your two or three people, going,

101
00:05:30,680 --> 00:05:32,360
Speaker 5: you know, I think the moon landings of the fake.

102
00:05:33,839 --> 00:05:38,280
Speaker 5: Now you've got the ability for that one particular person,

103
00:05:38,320 --> 00:05:40,400
Speaker 5: who perhaps only had an audience of three or four

104
00:05:40,440 --> 00:05:43,560
Speaker 5: in the pub, to have built up an audience on

105
00:05:43,640 --> 00:05:47,719
Speaker 5: social media across the world of many thousands of people.

106
00:05:47,800 --> 00:05:53,320
Speaker 5: So what social media allowed is people with perhaps unusual

107
00:05:53,440 --> 00:05:58,320
Speaker 5: views extreme in some cases extreme views to find a

108
00:05:58,680 --> 00:06:01,600
Speaker 5: to find like minded people to spread conspiracy theories. Now,

109
00:06:01,600 --> 00:06:04,800
Speaker 5: the attraction of conspiracy theories is that they make a

110
00:06:04,880 --> 00:06:09,400
Speaker 5: complicated world easy to understand. According to research from Victoria

111
00:06:09,480 --> 00:06:12,680
Speaker 5: University in twenty nineteen, so even before the pandemic, So

112
00:06:12,720 --> 00:06:15,800
Speaker 5: it's probably more now eight point nine percent of New

113
00:06:15,880 --> 00:06:19,600
Speaker 5: Zealanders believe the moon landings were faked. Social media is

114
00:06:19,640 --> 00:06:26,080
Speaker 5: allowed what we've been minority opinions to become mainstream, and

115
00:06:26,120 --> 00:06:28,360
Speaker 5: that has divided us. So that means that when you

116
00:06:28,400 --> 00:06:33,040
Speaker 5: get a critical event, instead of pulling together, there is

117
00:06:33,080 --> 00:06:35,800
Speaker 5: an opportunity for division to be stoked.

118
00:06:36,080 --> 00:06:38,920
Speaker 1: Is that because you know, there are these minority opinions,

119
00:06:39,120 --> 00:06:45,720
Speaker 1: but they somehow hitched themselves to majority beliefs. QAnon for example,

120
00:06:45,839 --> 00:06:51,279
Speaker 1: started out as this very fringe belief around deep state

121
00:06:51,440 --> 00:06:55,560
Speaker 1: and child abuse rings and then it somehow became hitched

122
00:06:55,640 --> 00:06:59,400
Speaker 1: to the majority Republican party in some way. So how

123
00:06:59,440 --> 00:07:02,440
Speaker 1: does that happened and how does social media enable these

124
00:07:02,480 --> 00:07:06,080
Speaker 1: fringe beliefs to then become hitched two majority believes?

125
00:07:06,360 --> 00:07:09,640
Speaker 5: Well, they need to be amplified by people who have

126
00:07:10,040 --> 00:07:14,160
Speaker 5: large platforms. Right, Sometimes you know a view can be

127
00:07:14,200 --> 00:07:17,360
Speaker 5: seeded on somewhere like four Channel or whatever, circulate there

128
00:07:17,360 --> 00:07:20,000
Speaker 5: for a while, and then it can jump to mainstream

129
00:07:20,040 --> 00:07:23,559
Speaker 5: social media, you know, Facebook, by somebody who's a member

130
00:07:23,560 --> 00:07:26,360
Speaker 5: of both those kinds of groups, and then it begins

131
00:07:26,400 --> 00:07:28,600
Speaker 5: to get shared. So then it's kind of an organic

132
00:07:28,680 --> 00:07:31,600
Speaker 5: sort of thing, but often it is actually something that

133
00:07:31,720 --> 00:07:36,120
Speaker 5: needs to be amplified. So for example, a fringe British

134
00:07:36,280 --> 00:07:39,960
Speaker 5: far right party a few years ago put out a

135
00:07:40,040 --> 00:07:45,280
Speaker 5: video three videos asserting that a particular incident that was

136
00:07:45,320 --> 00:07:50,920
Speaker 5: actually in Holland between two young boys was an attack

137
00:07:51,000 --> 00:07:56,520
Speaker 5: by a Muslim on a crippled white kids. And this

138
00:07:56,640 --> 00:07:58,840
Speaker 5: was on a fringe British site. Not didn't pick up

139
00:07:58,920 --> 00:08:03,320
Speaker 5: much until Donald Trump tweeted it, and when he retweeted it,

140
00:08:03,320 --> 00:08:07,160
Speaker 5: it's obviously became viral and it was fact checked again

141
00:08:07,160 --> 00:08:09,800
Speaker 5: by the wa she'd post, among others who and the

142
00:08:09,920 --> 00:08:12,880
Speaker 5: Dutch embassy had to come in and say, neither of

143
00:08:12,960 --> 00:08:16,800
Speaker 5: these boys involved in this video are either Muslim or immigrant.

144
00:08:16,840 --> 00:08:19,960
Speaker 5: They were both born and bred in the Netherlands. The

145
00:08:20,040 --> 00:08:22,560
Speaker 5: threat is real and this is very worrying because it's

146
00:08:22,600 --> 00:08:25,200
Speaker 5: divisive and it can lead to people getting hurt.

147
00:08:25,720 --> 00:08:27,880
Speaker 1: We saw that with the christ Church attacks right his

148
00:08:28,120 --> 00:08:32,760
Speaker 1: manifesto was called the Great Replacement I believe Great theory, Yes,

149
00:08:32,840 --> 00:08:37,640
Speaker 1: Great Replacement theory, which is a conspiracy theory that isn't

150
00:08:37,679 --> 00:08:40,400
Speaker 1: backed by science and doesn't have any basis and truth,

151
00:08:40,480 --> 00:08:43,200
Speaker 1: but was used as a justification for one of the

152
00:08:43,200 --> 00:08:47,360
Speaker 1: most horrific acts of violence in New Zealand's history. So

153
00:08:47,840 --> 00:08:49,880
Speaker 1: you know, and that is kind of this meta narrative.

154
00:08:49,960 --> 00:08:52,480
Speaker 1: And like you say, if you believe in the truth

155
00:08:52,520 --> 00:08:56,640
Speaker 1: of the big narrative, then it doesn't necessarily matter if

156
00:08:56,720 --> 00:09:01,080
Speaker 1: every single piece of smaller evidence is true or not,

157
00:09:01,960 --> 00:09:06,680
Speaker 1: because it feeds into your confirmation bias. Right. Do you

158
00:09:06,679 --> 00:09:11,360
Speaker 1: think that Donald Trump intentionally weaponized Twitter in order to

159
00:09:11,440 --> 00:09:16,680
Speaker 1: gain power and influence or do you think that he

160
00:09:16,720 --> 00:09:18,880
Speaker 1: would have found some other way if it wasn't Twitter.

161
00:09:20,240 --> 00:09:24,000
Speaker 5: I think several things came together to help Donald Trump.

162
00:09:24,360 --> 00:09:27,280
Speaker 5: One of them, of course, was the constant undermining of

163
00:09:27,280 --> 00:09:30,280
Speaker 5: the mainstream media that went alongside that. The media was

164
00:09:30,320 --> 00:09:33,280
Speaker 5: a bit slow, there was some hand ringing what have you,

165
00:09:33,320 --> 00:09:35,559
Speaker 5: before they actually came out and said and called Donald

166
00:09:35,559 --> 00:09:38,959
Speaker 5: Trump a spreader of misinformation and in fact a liar. Eventually,

167
00:09:39,360 --> 00:09:41,800
Speaker 5: the media didn't want to do that. They would rather

168
00:09:41,840 --> 00:09:44,720
Speaker 5: have taken each thing and fact checked it seriously and

169
00:09:44,760 --> 00:09:48,640
Speaker 5: put it against people were not there for that. And

170
00:09:48,679 --> 00:09:51,000
Speaker 5: at the same time, he was going fake news. Every

171
00:09:51,040 --> 00:09:53,640
Speaker 5: time there was a he had a rally and he

172
00:09:53,640 --> 00:09:56,360
Speaker 5: would point at individual journalism, call your fake new CNN

173
00:09:56,400 --> 00:09:59,960
Speaker 5: fake news, trying to undermine trust. So it went along

174
00:10:00,160 --> 00:10:03,720
Speaker 5: side again it was they just trust me, don't trust them.

175
00:10:04,000 --> 00:10:07,920
Speaker 5: He was speaking to these people who felt disenfranchised, felt

176
00:10:07,960 --> 00:10:11,360
Speaker 5: that they'd been forgotten, and so social media allowed him

177
00:10:11,400 --> 00:10:14,640
Speaker 5: to reach out to these little disparate groups all over

178
00:10:14,679 --> 00:10:16,960
Speaker 5: the place and bring them together for his base. So

179
00:10:17,000 --> 00:10:19,720
Speaker 5: he was talking to them and it helped him a lot.

180
00:10:19,800 --> 00:10:22,880
Speaker 5: That the mainstream media didn't quite know how to deal

181
00:10:22,920 --> 00:10:26,400
Speaker 5: with this in the beginning, and again amplified many times

182
00:10:26,480 --> 00:10:29,960
Speaker 5: his messages which were nonsense, you know, many of them,

183
00:10:30,240 --> 00:10:32,439
Speaker 5: but taking it seriously, and it became like his car crash.

184
00:10:32,640 --> 00:10:34,760
Speaker 5: I don't know about you, but I was every morning

185
00:10:35,240 --> 00:10:37,360
Speaker 5: wake up wonder what Donald Trump had said. I was

186
00:10:37,400 --> 00:10:39,640
Speaker 5: as bad as anybody else. And this is a big

187
00:10:39,640 --> 00:10:42,920
Speaker 5: mistake that often the liberal elite can make, is not

188
00:10:43,000 --> 00:10:47,360
Speaker 5: to take seriously people who behave and act differently. They

189
00:10:47,400 --> 00:10:49,520
Speaker 5: don't think who were they talking to? And then look

190
00:10:49,559 --> 00:10:54,040
Speaker 5: at that carefully. So social media allowed him to communicate

191
00:10:54,080 --> 00:10:57,520
Speaker 5: with a set of people that had been disenfranchised, and

192
00:10:57,600 --> 00:10:59,960
Speaker 5: he's still talking to those people and they're still there.

193
00:11:00,320 --> 00:11:03,880
Speaker 1: Are we seeing this in New Zealand? Is it similar

194
00:11:03,880 --> 00:11:07,400
Speaker 1: but on a different scale? Has our political structure with

195
00:11:07,600 --> 00:11:12,600
Speaker 1: MMP rather than just having a two party system per se?

196
00:11:12,679 --> 00:11:14,640
Speaker 1: It is kind of but not to the same extent

197
00:11:14,679 --> 00:11:17,360
Speaker 1: as the US. Has that inoculated us a little bit

198
00:11:17,840 --> 00:11:21,720
Speaker 1: against the extent to which it's happened in the US.

199
00:11:22,480 --> 00:11:23,920
Speaker 1: Or are we just as bad?

200
00:11:24,600 --> 00:11:26,800
Speaker 5: More people in New Zealand believe the moon landing was

201
00:11:26,840 --> 00:11:29,640
Speaker 5: faked than in America, So I think we've got to

202
00:11:29,640 --> 00:11:32,280
Speaker 5: be very careful not to believe where this little country

203
00:11:32,280 --> 00:11:33,960
Speaker 5: at the end of the world that nobody is going

204
00:11:33,960 --> 00:11:38,000
Speaker 5: to target, nobody cares about. It's not true. We're very strategic,

205
00:11:38,200 --> 00:11:40,880
Speaker 5: we are valuable. The fact that we are a liberal

206
00:11:40,960 --> 00:11:44,320
Speaker 5: democracy makes us very attractive as a target for some

207
00:11:44,480 --> 00:11:49,040
Speaker 5: misinformation peddlers, and there's been some excellent work done by

208
00:11:49,520 --> 00:11:53,280
Speaker 5: Stuff Circuit on the soft power of China. I think

209
00:11:53,280 --> 00:11:56,640
Speaker 5: the disinformation campaign at the last election didn't find as

210
00:11:56,720 --> 00:11:59,839
Speaker 5: much misinformation as we thought there might be, but work

211
00:12:00,160 --> 00:12:03,320
Speaker 5: that's been done in New Zealand has shown that we

212
00:12:03,360 --> 00:12:06,680
Speaker 5: are very concerned about misinformation. New Zealanders are and believe

213
00:12:06,720 --> 00:12:09,320
Speaker 5: we see quite a lot of it, which just that

214
00:12:09,400 --> 00:12:12,840
Speaker 5: belief undermines our trust in our information sources. So we

215
00:12:12,920 --> 00:12:15,040
Speaker 5: need to work very hard to build trust and the

216
00:12:15,040 --> 00:12:19,040
Speaker 5: trust research has done out of aut Media Muliati and

217
00:12:19,080 --> 00:12:23,240
Speaker 5: Greg Treadwell has shown that this is not going away,

218
00:12:23,679 --> 00:12:26,360
Speaker 5: but we can do things about it. Media, literacy, education,

219
00:12:26,559 --> 00:12:28,199
Speaker 5: there is a lot of hope, you know, there is

220
00:12:28,240 --> 00:12:30,040
Speaker 5: a lot of hope. People are concerned, they want to

221
00:12:30,120 --> 00:12:33,160
Speaker 5: learn and we can help. So yeah, I don't think

222
00:12:33,160 --> 00:12:37,079
Speaker 5: we should be all doom and gloom raising awareness and

223
00:12:37,200 --> 00:12:40,640
Speaker 5: knowing what has been a human pattern. It's just that

224
00:12:40,760 --> 00:12:43,640
Speaker 5: social media has allowed that pattern and a lot of

225
00:12:43,800 --> 00:12:46,520
Speaker 5: a lot of extra miles, so to speak.

226
00:12:50,160 --> 00:12:54,479
Speaker 2: So being no easy answers from hell and there. It's complex,

227
00:12:55,240 --> 00:13:00,319
Speaker 2: it's societal, it's interwoven with technological change and control of

228
00:13:00,480 --> 00:13:04,120
Speaker 2: those platforms. It's only going to get worse, and AI

229
00:13:04,360 --> 00:13:08,040
Speaker 2: is just another element that's going to confound things even more.

230
00:13:08,200 --> 00:13:11,040
Speaker 1: Yeah, it wasn't the most positive sounding of interviews, but

231
00:13:11,160 --> 00:13:12,720
Speaker 1: at the end of the day, I think what we

232
00:13:12,760 --> 00:13:14,920
Speaker 1: do need to remember is people do like to speculate

233
00:13:14,960 --> 00:13:18,760
Speaker 1: and get excited about things, but social media can just

234
00:13:18,840 --> 00:13:21,880
Speaker 1: take that speculation and excitement and turn it into something

235
00:13:21,920 --> 00:13:26,240
Speaker 1: pretty nasty pretty quickly. So media literacy is really key,

236
00:13:26,559 --> 00:13:28,959
Speaker 1: and just making sure that we're all you know, if

237
00:13:29,000 --> 00:13:31,600
Speaker 1: you hear somebody talking about something, you have that little

238
00:13:31,679 --> 00:13:35,240
Speaker 1: voice that goes, just check that one maybe, and reminding

239
00:13:35,240 --> 00:13:37,240
Speaker 1: your kids that it's really important to do the same

240
00:13:37,280 --> 00:13:38,360
Speaker 1: as well. That's vital.

241
00:13:38,800 --> 00:13:40,880
Speaker 2: Yeah, I think it's going to be interesting in the

242
00:13:40,920 --> 00:13:45,360
Speaker 2: whole endorsement of Trump by I wouldn't say, you know,

243
00:13:45,480 --> 00:13:48,960
Speaker 2: a widespread contingent of tech leaders in Silicon Valley, but

244
00:13:48,960 --> 00:13:51,199
Speaker 2: definitely some of the powerful ones. Peter Thiel, who was

245
00:13:51,240 --> 00:13:54,679
Speaker 2: a big backer last time, but most notably Elon Musk

246
00:13:55,400 --> 00:13:59,960
Speaker 2: within hours of the assassination attempt endorsing Trump to US

247
00:14:00,160 --> 00:14:03,480
Speaker 2: one hundred and ninety two million followers. Incredibly powerful and

248
00:14:03,480 --> 00:14:05,880
Speaker 2: that's why Trump is basically said in the last week,

249
00:14:06,400 --> 00:14:09,480
Speaker 2: I sort of need to support electric vehicles because of Elon,

250
00:14:10,240 --> 00:14:12,679
Speaker 2: And so he's starting to see this really which we haven't.

251
00:14:12,920 --> 00:14:15,200
Speaker 2: We knew it was happening, but haven't seen it to

252
00:14:15,200 --> 00:14:18,760
Speaker 2: this extent. That sort of the tech bro tech elite,

253
00:14:19,240 --> 00:14:23,720
Speaker 2: conservative leaning elites anyway intertwined with Trump. They've all got

254
00:14:23,720 --> 00:14:26,480
Speaker 2: super packs. They're giving him money. There's got to be

255
00:14:26,520 --> 00:14:30,360
Speaker 2: some payoff for that. So, for instance, on crypto, Trump

256
00:14:30,360 --> 00:14:34,120
Speaker 2: has gone from saying cryptos a scam to basically saying, oh, yeah,

257
00:14:34,160 --> 00:14:36,160
Speaker 2: I think I'm going to I'm going to see assets,

258
00:14:36,240 --> 00:14:40,400
Speaker 2: crypto assets and use it to payoff national debt. So

259
00:14:40,960 --> 00:14:43,320
Speaker 2: he's done a one ad on that. But you know,

260
00:14:43,720 --> 00:14:48,240
Speaker 2: is a crypto sector going to see a dream run

261
00:14:48,400 --> 00:14:53,040
Speaker 2: of you know, no regulation and government endorsement of their

262
00:14:53,080 --> 00:14:54,800
Speaker 2: way of doing things. I doubt it. You know, Trump

263
00:14:54,840 --> 00:15:00,720
Speaker 2: is just so sort of self interested, He's got authoritarian tendencies,

264
00:15:00,760 --> 00:15:02,120
Speaker 2: and crypto goes against that.

265
00:15:02,960 --> 00:15:05,960
Speaker 1: Yeah, I think you use the term payoff there, So

266
00:15:06,120 --> 00:15:08,120
Speaker 1: maybe we'll just let that's it.

267
00:15:15,760 --> 00:15:18,920
Speaker 2: So anyway, that's what's going on with social media and

268
00:15:18,920 --> 00:15:21,800
Speaker 2: the fight for control of the political narrative, which is

269
00:15:21,920 --> 00:15:25,880
Speaker 2: intensifying with the US elections looming. But there's another big

270
00:15:25,920 --> 00:15:28,920
Speaker 2: movement underway in the world of online search and the

271
00:15:29,080 --> 00:15:30,520
Speaker 2: advertising that goes with it.

272
00:15:30,680 --> 00:15:33,440
Speaker 1: The online search engine market was worth around two hundred

273
00:15:33,520 --> 00:15:37,360
Speaker 1: billion last year, a market dominated by Google. Even here

274
00:15:37,360 --> 00:15:40,320
Speaker 1: in New Zealand. Google's local accounts show it generated the

275
00:15:40,320 --> 00:15:43,920
Speaker 1: best part of one billion dollars in sales here, much

276
00:15:43,960 --> 00:15:47,240
Speaker 1: of which will be through businesses paying to feature at

277
00:15:47,280 --> 00:15:49,000
Speaker 1: the top of Google search results.

278
00:15:49,200 --> 00:15:52,720
Speaker 2: Those paid links are everything for a retailer or a

279
00:15:52,760 --> 00:15:57,800
Speaker 2: financial services company trying to attract customers. But AI is

280
00:15:57,960 --> 00:16:01,920
Speaker 2: changing search. It became most evident last year with conversational

281
00:16:02,040 --> 00:16:05,960
Speaker 2: Search debuting and Microsoft's being search engine powered by the

282
00:16:06,000 --> 00:16:08,240
Speaker 2: same tech underpinning chat GPT.

283
00:16:08,600 --> 00:16:12,000
Speaker 1: Then a few months ago, Google debut it's aioverview future,

284
00:16:12,280 --> 00:16:14,640
Speaker 1: which will try to answer your search query, not simply

285
00:16:14,720 --> 00:16:17,840
Speaker 1: with a bunch of relevant links, but a short blurb

286
00:16:17,840 --> 00:16:19,360
Speaker 1: attempting to answer your question.

287
00:16:19,600 --> 00:16:23,640
Speaker 2: So featuring in that AI generated blurb will increasingly be

288
00:16:23,760 --> 00:16:27,560
Speaker 2: the place to be in search results. That is implications

289
00:16:28,160 --> 00:16:31,360
Speaker 2: for the type of content website owners create to appeal

290
00:16:31,800 --> 00:16:33,280
Speaker 2: to Google search algorithms.

291
00:16:33,440 --> 00:16:36,400
Speaker 1: At List Digital is a Wellington based paid search and

292
00:16:36,480 --> 00:16:39,840
Speaker 1: social media startup that helps other tech startups find an

293
00:16:39,920 --> 00:16:42,000
Speaker 1: audience in the noisy world of search.

294
00:16:42,360 --> 00:16:45,680
Speaker 2: It's founder, Ryan McMillan, has helped secure some impressive results

295
00:16:45,680 --> 00:16:49,480
Speaker 2: for fast growing companies like Chersey's and Jasper. He's been

296
00:16:49,520 --> 00:16:52,080
Speaker 2: thinking a lot about the future of the search business

297
00:16:52,160 --> 00:16:55,000
Speaker 2: as Google Bing and others embrace AI.

298
00:16:55,440 --> 00:16:58,400
Speaker 1: This one is worth listening to for anyone trying to

299
00:16:58,400 --> 00:17:02,720
Speaker 1: find an audience online. Here's at list Digital's Ryan McMillan.

300
00:17:06,200 --> 00:17:08,919
Speaker 2: Brian, thanks so much for coming on the business of

301
00:17:09,080 --> 00:17:11,240
Speaker 2: Tech Now. We met probably a few months ago for

302
00:17:11,240 --> 00:17:12,840
Speaker 2: the first time. It was on a red Eye flight

303
00:17:13,200 --> 00:17:16,400
Speaker 2: to Wellington, and I was really taken by the conversation.

304
00:17:16,440 --> 00:17:19,440
Speaker 2: We talked about the tech landscape, startups which we're both

305
00:17:19,600 --> 00:17:24,200
Speaker 2: passionate about, and search engine optimization and started talking about

306
00:17:24,200 --> 00:17:28,520
Speaker 2: this revolution that's happening with generative AI, conversational search, which

307
00:17:28,560 --> 00:17:30,200
Speaker 2: is changing the game. So we're going to get into

308
00:17:30,280 --> 00:17:32,840
Speaker 2: that and what it means for New Zealand businesses going

309
00:17:32,880 --> 00:17:36,520
Speaker 2: global trying to find people online to buy their wares.

310
00:17:37,080 --> 00:17:39,520
Speaker 2: But tell us a little bit first about your business

311
00:17:39,560 --> 00:17:40,400
Speaker 2: at list Digital.

312
00:17:41,160 --> 00:17:41,400
Speaker 4: Yeah.

313
00:17:41,440 --> 00:17:44,240
Speaker 3: Sure, So I started at list Digital about five and

314
00:17:44,280 --> 00:17:46,679
Speaker 3: a half years ago, and fundamentally what we do is

315
00:17:46,720 --> 00:17:50,040
Speaker 3: we specialize in paid search like Google Ads, paid social

316
00:17:50,200 --> 00:17:53,600
Speaker 3: like LinkedIn or Facebook advertising, and then search engine optimization

317
00:17:53,680 --> 00:17:55,919
Speaker 3: as well, which obviously we're here to talk about today.

318
00:17:56,400 --> 00:17:59,480
Speaker 3: So some of our clients seventy five percent of them

319
00:18:00,040 --> 00:18:02,359
Speaker 3: to B SAS companies, with the other twenty five percent

320
00:18:02,440 --> 00:18:04,600
Speaker 3: being B two C SAS or B two C tech.

321
00:18:05,080 --> 00:18:06,919
Speaker 3: And what we find with almost all of the B

322
00:18:07,000 --> 00:18:10,160
Speaker 3: to B SaaS companies is that they're looking to expand internationally.

323
00:18:10,240 --> 00:18:13,480
Speaker 3: So from an SEO perspective, that can already introduce a

324
00:18:13,600 --> 00:18:16,400
Speaker 3: number of questions that need to be answered and ideally

325
00:18:16,440 --> 00:18:19,920
Speaker 3: thought of fairly early in the piece. For example, your

326
00:18:19,960 --> 00:18:24,320
Speaker 3: domain structure is something that's quite important to consider, as

327
00:18:24,440 --> 00:18:27,720
Speaker 3: is the localization of language that you'll be using. Do

328
00:18:27,760 --> 00:18:29,800
Speaker 3: you need to talk in different languages? Do you need

329
00:18:29,840 --> 00:18:32,760
Speaker 3: to colloquialize the language. These are just a few of

330
00:18:32,760 --> 00:18:35,840
Speaker 3: the things that we set out to answer quite early

331
00:18:35,880 --> 00:18:38,280
Speaker 3: on and then work through building a strategy in order

332
00:18:38,320 --> 00:18:39,920
Speaker 3: to make best use of that.

333
00:18:40,160 --> 00:18:42,199
Speaker 2: So I think the perception out there is you know

334
00:18:42,320 --> 00:18:45,000
Speaker 2: search you know was in the late nineties, there were

335
00:18:45,000 --> 00:18:48,439
Speaker 2: early search engines. Google came along with the power of

336
00:18:48,520 --> 00:18:53,440
Speaker 2: its way of ranking pages and its algorithms changed the game,

337
00:18:53,600 --> 00:18:58,320
Speaker 2: and then nothing really happened fundamentally for the next twenty

338
00:18:58,720 --> 00:19:01,480
Speaker 2: five years. There was a lot of great innovation behind

339
00:19:01,520 --> 00:19:05,040
Speaker 2: the scenes, but search the experience that white screen with

340
00:19:05,160 --> 00:19:09,760
Speaker 2: the search bar didn't really change and people love Google.

341
00:19:10,359 --> 00:19:13,160
Speaker 2: Early twenty twenty three, we saw Being, which had really

342
00:19:13,160 --> 00:19:18,440
Speaker 2: been struggling in search, team up with open Ai integrating

343
00:19:19,320 --> 00:19:23,640
Speaker 2: GPT into the Being search engine. A lot of interest

344
00:19:23,680 --> 00:19:26,399
Speaker 2: in that, and that was really the start of conversational search,

345
00:19:26,560 --> 00:19:30,760
Speaker 2: an auto generated summary that tries to answer your question. So, Ryan,

346
00:19:31,280 --> 00:19:33,600
Speaker 2: what is actual conversational search?

347
00:19:33,600 --> 00:19:34,359
Speaker 4: How does it work?

348
00:19:34,880 --> 00:19:39,040
Speaker 3: The ability of conversational search is, first of all, can

349
00:19:39,560 --> 00:19:44,000
Speaker 3: understand using natural language processing the intent behind what you're

350
00:19:44,040 --> 00:19:46,639
Speaker 3: looking for at a much deeper level. It's not about

351
00:19:46,680 --> 00:19:49,679
Speaker 3: going what a related keyword someone might talk about, but

352
00:19:49,800 --> 00:19:52,679
Speaker 3: actually really understanding, hey, what does this person want to know?

353
00:19:53,440 --> 00:19:55,920
Speaker 3: And then on top of that, it has this sort

354
00:19:55,920 --> 00:20:00,560
Speaker 3: of multifaceted, dynamic approach where you can continue to on

355
00:20:01,320 --> 00:20:05,679
Speaker 3: the questions you're asking and almost like it's mimicking actually

356
00:20:05,680 --> 00:20:07,639
Speaker 3: talking to a person. You know, I want to buy

357
00:20:07,760 --> 00:20:09,480
Speaker 3: Nike shoes, and it shows you a bunch, Well, I

358
00:20:09,480 --> 00:20:11,240
Speaker 3: want to see all the red ones. And then now

359
00:20:11,240 --> 00:20:12,680
Speaker 3: you've got all the red ones, and you can really

360
00:20:12,720 --> 00:20:14,720
Speaker 3: build on that, and you can even loop back later

361
00:20:14,840 --> 00:20:17,360
Speaker 3: like remember those those red shoes you showed me? Can

362
00:20:17,400 --> 00:20:19,840
Speaker 3: you show me the first two? Because I want to

363
00:20:19,880 --> 00:20:22,320
Speaker 3: see those two and so the reality is that you're

364
00:20:22,359 --> 00:20:27,040
Speaker 3: almost starting to get this customer service experience from conversational search.

365
00:20:27,119 --> 00:20:29,520
Speaker 3: Like it means that we can go a lot deeper

366
00:20:29,560 --> 00:20:31,800
Speaker 3: with the searches that we're doing, and off the back

367
00:20:31,840 --> 00:20:35,800
Speaker 3: of it probably not only answer the questions we've got better,

368
00:20:35,920 --> 00:20:38,280
Speaker 3: but also walk away with a lot more than we

369
00:20:38,320 --> 00:20:41,200
Speaker 3: sort of bargained for as well. For example, you could

370
00:20:41,280 --> 00:20:43,639
Speaker 3: look at buying Nike shoes, choose the red ones, and

371
00:20:43,640 --> 00:20:45,640
Speaker 3: then go, now, I really want to run this half

372
00:20:45,680 --> 00:20:48,159
Speaker 3: marathon in about six months, what's a training plan? And

373
00:20:48,400 --> 00:20:50,879
Speaker 3: you can really build that and have that all stitched

374
00:20:50,960 --> 00:20:53,679
Speaker 3: up in a few minutes versus right now, you'd need

375
00:20:53,720 --> 00:20:56,359
Speaker 3: to google half marathon training plan and it wouldn't cater

376
00:20:56,440 --> 00:20:58,720
Speaker 3: to you, it wouldn't have your past search experience, it

377
00:20:58,720 --> 00:21:00,920
Speaker 3: wouldn't know who you are, and it wouldn't be out

378
00:21:00,920 --> 00:21:04,399
Speaker 3: of therefore, adjust that information to cater to yourself. So

379
00:21:04,680 --> 00:21:06,320
Speaker 3: it's a really exciting space to be in.

380
00:21:07,160 --> 00:21:09,880
Speaker 2: Yeah, it's pretty cool. And what's your take on how

381
00:21:09,960 --> 00:21:14,160
Speaker 2: this is changing search engine optimization and your own business?

382
00:21:14,760 --> 00:21:17,640
Speaker 3: It is hugely changing the game. And what I think

383
00:21:17,680 --> 00:21:20,000
Speaker 3: we're going to see is probably less use of a

384
00:21:20,080 --> 00:21:23,600
Speaker 3: traditional search engine as people start to find answers in

385
00:21:23,640 --> 00:21:26,520
Speaker 3: different ways, and by the very nature of that, it

386
00:21:26,520 --> 00:21:30,320
Speaker 3: will also change the SEO industry itself. Do I think

387
00:21:30,359 --> 00:21:33,280
Speaker 3: that that will be the end of SEO? Certainly not.

388
00:21:33,440 --> 00:21:35,600
Speaker 3: But I think SEO as we know it is going

389
00:21:35,640 --> 00:21:38,760
Speaker 3: to change, which I suppose is something quite cliche every

390
00:21:38,760 --> 00:21:41,640
Speaker 3: time every year we hear something that SEO is dead.

391
00:21:42,600 --> 00:21:45,080
Speaker 4: Is SEO dead? I'm not sure? But SEO is certainly

392
00:21:45,119 --> 00:21:45,840
Speaker 4: always changing.

393
00:21:46,520 --> 00:21:48,800
Speaker 2: So what does that mean for the types of content?

394
00:21:49,359 --> 00:21:53,040
Speaker 2: Because ultimately your job now is to try and feature

395
00:21:53,080 --> 00:21:56,639
Speaker 2: in that auto generated summary you want. For instance, if

396
00:21:56,640 --> 00:22:00,680
Speaker 2: your campman do a clothing company and some enters tell

397
00:22:00,720 --> 00:22:04,240
Speaker 2: me the best clothing for a trip to root Burn.

398
00:22:04,440 --> 00:22:06,760
Speaker 2: If I'm hiking the root Burn, what gear do I need?

399
00:22:07,400 --> 00:22:10,480
Speaker 2: Katmandu wants to be in that summary at the top

400
00:22:10,520 --> 00:22:12,960
Speaker 2: of Google. So what do you need to do now

401
00:22:13,000 --> 00:22:14,920
Speaker 2: to make sure that your stuff gets there.

402
00:22:15,200 --> 00:22:19,040
Speaker 3: The experiential nature is probably the most important piece of

403
00:22:19,560 --> 00:22:21,600
Speaker 3: SEO at the moment, and in my opinion, it's the

404
00:22:21,640 --> 00:22:27,240
Speaker 3: way that we really find that competitive advantage against llm's

405
00:22:27,280 --> 00:22:30,320
Speaker 3: which your AI like Gemini, which will be able to

406
00:22:30,359 --> 00:22:33,040
Speaker 3: give your answers and result in people not visiting your

407
00:22:33,040 --> 00:22:38,080
Speaker 3: website at all. I think with the Gemini responses, really

408
00:22:38,080 --> 00:22:41,520
Speaker 3: what they're doing is synthesizing all of the information on

409
00:22:41,560 --> 00:22:43,800
Speaker 3: the web and then using that to obviously generate an

410
00:22:43,840 --> 00:22:47,000
Speaker 3: answer based on the patterns and data that the model's

411
00:22:47,000 --> 00:22:49,840
Speaker 3: been trained on. So I think that there's going to

412
00:22:49,840 --> 00:22:53,440
Speaker 3: be less opportunity for businesses to really produce that sort

413
00:22:53,480 --> 00:22:57,240
Speaker 3: of information level content that doesn't necessarily have a point

414
00:22:57,280 --> 00:23:01,479
Speaker 3: of difference, and that's where this originality piece really starts

415
00:23:01,480 --> 00:23:04,280
Speaker 3: to shine. The thing is about one three hundred percent

416
00:23:04,359 --> 00:23:08,560
Speaker 3: increase in searches for Reddit variants in the past twelve months,

417
00:23:08,560 --> 00:23:11,840
Speaker 3: and Reddit, I believe was actually the sixth most or

418
00:23:11,920 --> 00:23:15,040
Speaker 3: so visited website in the past year compared to the

419
00:23:15,160 --> 00:23:18,840
Speaker 3: sixty eighth the year before. And I think that what

420
00:23:18,880 --> 00:23:23,560
Speaker 3: that shows is a really solid display that people want information.

421
00:23:23,720 --> 00:23:25,919
Speaker 3: They want to search what do I need for the

422
00:23:25,960 --> 00:23:27,879
Speaker 3: root burn track? And they want to know I need,

423
00:23:28,119 --> 00:23:30,320
Speaker 3: you know, a range jacket and this many thermals and

424
00:23:30,359 --> 00:23:32,479
Speaker 3: that type of thing. But they also want to know

425
00:23:32,640 --> 00:23:35,040
Speaker 3: from you know, John that did the root burn track

426
00:23:35,080 --> 00:23:37,240
Speaker 3: and how we found it and if autumn was a

427
00:23:37,280 --> 00:23:39,439
Speaker 3: good time to do it, and what his experience was

428
00:23:39,480 --> 00:23:41,960
Speaker 3: and I think that there's a real harmony that can

429
00:23:42,040 --> 00:23:46,240
Speaker 3: exist in search results, in this sort of information seeking

430
00:23:47,080 --> 00:23:50,920
Speaker 3: dynamic where humans can continue to play a big part.

431
00:23:51,200 --> 00:23:55,000
Speaker 3: And so stepping back from that, there's a big sort

432
00:23:55,040 --> 00:23:59,840
Speaker 3: of semantic understanding component that Google and other search engines

433
00:23:59,840 --> 00:24:03,359
Speaker 3: have been really trying to capture the past several years.

434
00:24:03,359 --> 00:24:07,440
Speaker 3: So since about twenty thirteen and specifically in twenty fifteen,

435
00:24:08,040 --> 00:24:12,240
Speaker 3: there's pretty big updates to the Google search algorithm around

436
00:24:12,440 --> 00:24:15,520
Speaker 3: really understanding the wider meaning of what you're looking for.

437
00:24:15,960 --> 00:24:17,560
Speaker 3: It used to be the case. I mean, this is

438
00:24:17,600 --> 00:24:21,000
Speaker 3: a whole other story, but it used to be so interesting.

439
00:24:21,080 --> 00:24:23,919
Speaker 3: You could literally just write in white text on a

440
00:24:23,920 --> 00:24:26,840
Speaker 3: white background life insurance five hundred times and that would

441
00:24:26,840 --> 00:24:28,560
Speaker 3: help you get to the top of Google. And these

442
00:24:28,640 --> 00:24:30,639
Speaker 3: days you don't even have to use the exact word

443
00:24:31,119 --> 00:24:34,160
Speaker 3: life insurance to rank for the keyword life insurance. Your

444
00:24:34,200 --> 00:24:36,960
Speaker 3: page just has to be about that. And so this

445
00:24:37,119 --> 00:24:41,359
Speaker 3: sort of wider contextual semantic understanding has existed in Google's

446
00:24:41,480 --> 00:24:45,840
Speaker 3: machine learning algorithm. But what's so interesting now is really

447
00:24:45,840 --> 00:24:49,560
Speaker 3: the ability, through the patterns and data matching, to anticipate

448
00:24:49,640 --> 00:24:52,919
Speaker 3: the other things that you might want to know to

449
00:24:53,000 --> 00:24:55,840
Speaker 3: your point around what does a business need to do?

450
00:24:56,240 --> 00:25:01,320
Speaker 3: Google talks about something called EAT, which is experience, expertise, authority,

451
00:25:01,320 --> 00:25:05,000
Speaker 3: and trustworthiness, and I think that those things continue to

452
00:25:05,000 --> 00:25:07,880
Speaker 3: play a big role. Experience obviously is a big one,

453
00:25:07,920 --> 00:25:11,240
Speaker 3: as demonstrated by Reddit. Fundamentally, what we want to do

454
00:25:11,480 --> 00:25:15,680
Speaker 3: is create content that's original as a keyword or as

455
00:25:15,720 --> 00:25:19,280
Speaker 3: a key point, but also that can't be copied easily,

456
00:25:19,400 --> 00:25:22,240
Speaker 3: because that's the real point of difference here. And so

457
00:25:22,359 --> 00:25:25,239
Speaker 3: when you can leverage things like first party data. So

458
00:25:25,800 --> 00:25:27,840
Speaker 3: if you're cat man Do and you for some reason

459
00:25:27,880 --> 00:25:29,719
Speaker 3: had data on how many people are doing the rout

460
00:25:29,760 --> 00:25:32,320
Speaker 3: burn track, you could leverage that data in some way

461
00:25:32,359 --> 00:25:34,399
Speaker 3: to provide a point of difference in an article that

462
00:25:34,480 --> 00:25:37,920
Speaker 3: someone can't get elsewhere. Likewise, you could have your own

463
00:25:38,000 --> 00:25:40,679
Speaker 3: catman Do community and then break that down into the

464
00:25:40,720 --> 00:25:43,240
Speaker 3: different tracks people have done, and you have your own

465
00:25:43,240 --> 00:25:47,840
Speaker 3: sort of Reddit experience. In that sense, trustworthiness and brand

466
00:25:47,920 --> 00:25:50,280
Speaker 3: signals is a big one, and I think what we

467
00:25:50,400 --> 00:25:54,800
Speaker 3: really want to do is build around specific authors so

468
00:25:54,840 --> 00:25:58,320
Speaker 3: that they know, hey, it's Peter that's writing this article.

469
00:25:58,320 --> 00:26:01,120
Speaker 3: When I follow Peter and Peter is writing for Katmandu

470
00:26:01,520 --> 00:26:03,840
Speaker 3: or Similarly, it could be if you're the CEO of

471
00:26:03,920 --> 00:26:07,919
Speaker 3: Catman do, then you build this authorship. These owned communities

472
00:26:07,920 --> 00:26:09,760
Speaker 3: are going to become more important, and I think the

473
00:26:09,920 --> 00:26:14,480
Speaker 3: placement of content across different channels, with SEO being one

474
00:26:14,520 --> 00:26:16,600
Speaker 3: of them, is really where the space is going to

475
00:26:16,640 --> 00:26:16,960
Speaker 3: move to.

476
00:26:17,560 --> 00:26:20,560
Speaker 2: Yeah, so it's easier than ever now to actually generate

477
00:26:20,640 --> 00:26:23,840
Speaker 2: content for your website with generative AI. You can get chat,

478
00:26:23,920 --> 00:26:28,760
Speaker 2: GPT or Perplexity or Gemini to do that. But the

479
00:26:28,840 --> 00:26:32,119
Speaker 2: search engine isn't going to like that because everyone is

480
00:26:32,119 --> 00:26:34,480
Speaker 2: doing that. It's not unique. What they're wanting is that

481
00:26:34,720 --> 00:26:38,639
Speaker 2: community based stuff, that experiential stuff, case studies. So what

482
00:26:38,680 --> 00:26:40,399
Speaker 2: does does that mean? Let's say, like a campman do,

483
00:26:41,040 --> 00:26:44,359
Speaker 2: should you be really pushing to get people to review

484
00:26:44,400 --> 00:26:46,720
Speaker 2: your stuff and write about your stuff on redded or

485
00:26:46,720 --> 00:26:49,440
Speaker 2: in camping forums and things like that. You're going to

486
00:26:49,480 --> 00:26:52,119
Speaker 2: have to put a lot more effort into generating that

487
00:26:53,040 --> 00:26:54,680
Speaker 2: more genuine content.

488
00:26:55,000 --> 00:26:55,879
Speaker 4: Yeah, I believe so.

489
00:26:56,040 --> 00:26:59,920
Speaker 3: And I think that there's been this concept in se

490
00:27:00,600 --> 00:27:03,320
Speaker 3: of backlinking, which Google has always described as kind of

491
00:27:03,400 --> 00:27:06,720
Speaker 3: like votes from other website about your website, and that

492
00:27:07,359 --> 00:27:10,400
Speaker 3: provides you a significant boost in your chances of ranking

493
00:27:10,480 --> 00:27:11,280
Speaker 3: for a search result.

494
00:27:11,320 --> 00:27:12,960
Speaker 4: If lots of other websites.

495
00:27:12,560 --> 00:27:16,360
Speaker 3: Are ranking a linking back to you, now what I

496
00:27:16,400 --> 00:27:19,679
Speaker 3: think the backlinks will continue to exist, But and that

497
00:27:19,800 --> 00:27:23,399
Speaker 3: component and the ranking algorithm of what we populate, whether

498
00:27:23,480 --> 00:27:27,119
Speaker 3: that's populated by an AI or it's populated in the

499
00:27:27,160 --> 00:27:32,320
Speaker 3: search results below that automated response, it's going to focus

500
00:27:32,400 --> 00:27:35,440
Speaker 3: much more on social signals, So are you on LinkedIn

501
00:27:35,520 --> 00:27:38,199
Speaker 3: and have a following and other things like that. And

502
00:27:38,280 --> 00:27:39,919
Speaker 3: I think that it's also going to focus in a

503
00:27:39,920 --> 00:27:42,320
Speaker 3: big way on the review component as well, and so

504
00:27:42,400 --> 00:27:45,600
Speaker 3: building that trustworthiness through reviews I think will become one

505
00:27:45,640 --> 00:27:48,400
Speaker 3: of the key new sort of ways that you get backlinks.

506
00:27:48,400 --> 00:27:50,280
Speaker 3: So yes, you should be on forums, you should be

507
00:27:50,320 --> 00:27:53,119
Speaker 3: on social channels, and really you just want to leverage

508
00:27:53,160 --> 00:27:58,080
Speaker 3: as much anecdotal personal reviews as you can, as opposed

509
00:27:58,119 --> 00:28:01,320
Speaker 3: to just writing a blog about here's how you can

510
00:28:01,400 --> 00:28:03,480
Speaker 3: get to the root burn track and then asking someone

511
00:28:03,560 --> 00:28:05,960
Speaker 3: else that wrote about a different track to link to

512
00:28:06,000 --> 00:28:08,320
Speaker 3: the root burn track article you've written, which is the

513
00:28:08,359 --> 00:28:09,280
Speaker 3: old way of doing it.

514
00:28:10,359 --> 00:28:11,480
Speaker 4: At the end of the day, like.

515
00:28:11,520 --> 00:28:17,840
Speaker 3: What Google rewards is providing users with a great experience.

516
00:28:17,920 --> 00:28:21,760
Speaker 3: So Google is a multi billion dollar cash machine they

517
00:28:21,800 --> 00:28:26,440
Speaker 3: invest in all sorts of stuff, from automated driving through

518
00:28:26,480 --> 00:28:28,840
Speaker 3: to all sorts of other sort of like research endeavors

519
00:28:28,880 --> 00:28:32,600
Speaker 3: that they're involved in, but really it's all funded by

520
00:28:32,880 --> 00:28:37,760
Speaker 3: Google Ads. And so for them, success looks like someone

521
00:28:37,880 --> 00:28:43,920
Speaker 3: searching in Google, getting their answer and leaving, or you know,

522
00:28:43,960 --> 00:28:47,040
Speaker 3: continuing to search for other things. And so what they

523
00:28:47,080 --> 00:28:49,720
Speaker 3: want you to do is just provide the best user experience.

524
00:28:49,760 --> 00:28:52,320
Speaker 3: And I think that the main opportunity for providing the

525
00:28:52,320 --> 00:28:55,719
Speaker 3: best user experience now comes from what can people add

526
00:28:55,760 --> 00:28:58,640
Speaker 3: to the max, not what sort of blogs can we

527
00:28:58,680 --> 00:28:59,120
Speaker 3: come up to?

528
00:28:59,200 --> 00:29:01,480
Speaker 4: Which are a GENAI could just do it?

529
00:29:02,000 --> 00:29:05,720
Speaker 2: Yeah, and we're seeing some of our B two C

530
00:29:06,520 --> 00:29:10,240
Speaker 2: brands like Charities, which is a client of yours. They're

531
00:29:10,240 --> 00:29:13,240
Speaker 2: doing great work to doing podcasts, are doing how to articles,

532
00:29:14,080 --> 00:29:17,040
Speaker 2: they're active on social media. So in the hope that

533
00:29:17,040 --> 00:29:20,280
Speaker 2: that stuff rises up because it's really good quality content.

534
00:29:20,880 --> 00:29:23,360
Speaker 2: But in terms of like maybe a company like Timely,

535
00:29:24,000 --> 00:29:26,360
Speaker 2: which is more of a B to B company, it's

536
00:29:26,400 --> 00:29:28,840
Speaker 2: the same sort of set of rules apply. How do

537
00:29:28,920 --> 00:29:33,600
Speaker 2: you introduce that experiential sort of review type experiences? Is

538
00:29:33,600 --> 00:29:37,080
Speaker 2: what using Timely is Like, I mean, there are websites

539
00:29:37,160 --> 00:29:41,120
Speaker 2: like clatching that which review businesses based on their customer

540
00:29:41,160 --> 00:29:43,520
Speaker 2: service and the quality of their products. Is there going

541
00:29:43,560 --> 00:29:44,960
Speaker 2: to be a lot more of a focus on those

542
00:29:44,960 --> 00:29:47,560
Speaker 2: sorts of review aggregator sites that are aimed at B

543
00:29:47,640 --> 00:29:49,160
Speaker 2: to B type products.

544
00:29:49,720 --> 00:29:52,240
Speaker 3: I think review aggregators are one part of it. That

545
00:29:52,320 --> 00:29:55,080
Speaker 3: market itself is actually very saturated, and if you take

546
00:29:55,120 --> 00:29:59,840
Speaker 3: a given search, like you insert your vertical plus software

547
00:30:00,080 --> 00:30:03,480
Speaker 3: type search, so say like here Salon software or something

548
00:30:03,560 --> 00:30:07,840
Speaker 3: like that. In the case of Timely, those software qualifier

549
00:30:07,880 --> 00:30:11,440
Speaker 3: searches are pretty much dominated by the likes of G

550
00:30:11,520 --> 00:30:14,400
Speaker 3: two and trust Radius, et cetera. So it's important to

551
00:30:14,440 --> 00:30:16,720
Speaker 3: be part of those, but it's also important to add

552
00:30:16,760 --> 00:30:21,160
Speaker 3: a personal touch of actual customers reviewing your product, whether

553
00:30:21,200 --> 00:30:25,480
Speaker 3: that might be other sort of business influencers in the space.

554
00:30:25,520 --> 00:30:28,760
Speaker 3: There are sort of like smaller medium business influencers around

555
00:30:28,800 --> 00:30:31,760
Speaker 3: the world that actually focus specifically on the beauty industry,

556
00:30:32,120 --> 00:30:34,960
Speaker 3: but also just talking to your specific customers and then

557
00:30:35,280 --> 00:30:38,200
Speaker 3: tying that back to the actual business benefits. Because there's

558
00:30:38,520 --> 00:30:42,480
Speaker 3: say the G two reviews that you might get, which

559
00:30:42,480 --> 00:30:45,760
Speaker 3: are more feature driven and are really about parody and

560
00:30:45,840 --> 00:30:49,240
Speaker 3: which software does what better than the other. But then

561
00:30:49,240 --> 00:30:52,120
Speaker 3: there's also this other experience of like the purpose of

562
00:30:52,240 --> 00:30:55,080
Speaker 3: using that software, which is I now have a lot

563
00:30:55,080 --> 00:30:57,480
Speaker 3: more time, and I've managed to hire two more people.

564
00:30:57,520 --> 00:30:59,680
Speaker 3: I spend that time working on the business, not in

565
00:30:59,720 --> 00:31:02,400
Speaker 3: the business, those types of things. So adding in that

566
00:31:02,480 --> 00:31:05,520
Speaker 3: first party data improvement that you've got and then putting

567
00:31:05,560 --> 00:31:07,000
Speaker 3: faces on it, because at the end of the day,

568
00:31:07,040 --> 00:31:08,800
Speaker 3: people pretty much buy from people.

569
00:31:09,080 --> 00:31:13,600
Speaker 2: Yeah, So for decades, Google has made a lot of

570
00:31:13,600 --> 00:31:16,800
Speaker 2: money by you know, those first few links, sponsored links

571
00:31:16,840 --> 00:31:19,320
Speaker 2: at the top. They've talked about that this will apply

572
00:31:19,440 --> 00:31:23,240
Speaker 2: to AI overviews at some point as well. How do

573
00:31:23,280 --> 00:31:25,760
Speaker 2: you see that playing out? So you will literally pay

574
00:31:26,040 --> 00:31:28,680
Speaker 2: to be featured in that summary overview that they give.

575
00:31:29,360 --> 00:31:33,080
Speaker 2: How do they balance that trust and interest in giving

576
00:31:33,080 --> 00:31:36,080
Speaker 2: you the really good stuff with their desire to make

577
00:31:36,120 --> 00:31:39,400
Speaker 2: money out of those those who are featured in those summaries.

578
00:31:39,520 --> 00:31:41,200
Speaker 3: Yeah, well, I think a lot of it will come

579
00:31:41,240 --> 00:31:44,680
Speaker 3: down to the actual trust for the brand for that business.

580
00:31:44,720 --> 00:31:48,160
Speaker 3: So if people still really trust Google, then the same

581
00:31:48,200 --> 00:31:51,600
Speaker 3: way that Google ads is really effective is that actually

582
00:31:51,680 --> 00:31:54,080
Speaker 3: people might just go, oh, I'll just go to this anyway,

583
00:31:54,080 --> 00:31:56,360
Speaker 3: because you know, I trust Google, I like to I'll

584
00:31:56,400 --> 00:32:02,960
Speaker 3: continue using it. Similarly, in and the AI responses. As

585
00:32:03,000 --> 00:32:06,120
Speaker 3: long as people are getting the value from those responses,

586
00:32:06,560 --> 00:32:09,600
Speaker 3: whether they're sponsored or not, people will continue to come back.

587
00:32:09,640 --> 00:32:13,600
Speaker 3: And in Google Ads an SEO, Google has a way

588
00:32:13,640 --> 00:32:16,120
Speaker 3: to measure the actual quality of what's happening as well,

589
00:32:16,440 --> 00:32:20,280
Speaker 3: sign a quality score to your ads, and that influences

590
00:32:20,320 --> 00:32:22,520
Speaker 3: the price that you actually pay as well. And so

591
00:32:22,840 --> 00:32:27,080
Speaker 3: I don't think necessarily that the sponsored component will make

592
00:32:27,200 --> 00:32:30,920
Speaker 3: people stop using the platform more or less. It's more

593
00:32:30,960 --> 00:32:34,000
Speaker 3: about the quality of the sponsored content if they got.

594
00:32:33,880 --> 00:32:34,840
Speaker 4: Their query answer.

595
00:32:35,000 --> 00:32:39,400
Speaker 3: Now, one bigger piece to think about is actually just

596
00:32:40,000 --> 00:32:43,320
Speaker 3: the business models that will be used, because obviously open

597
00:32:43,360 --> 00:32:46,120
Speaker 3: ai is using a subscription model at the moment, and

598
00:32:46,640 --> 00:32:49,280
Speaker 3: Google Ads is using like a pay per click model

599
00:32:49,280 --> 00:32:52,880
Speaker 3: where you click their ad, the advertiser would pay Google

600
00:32:52,920 --> 00:32:54,959
Speaker 3: a bunch of money and that's how they make these

601
00:32:55,000 --> 00:32:58,280
Speaker 3: billions of dollars. So it's interesting to think about, you know,

602
00:32:59,320 --> 00:33:01,640
Speaker 3: what the actual business models will be that will drive this.

603
00:33:01,720 --> 00:33:03,600
Speaker 3: I'm sure there will still be continue to be an

604
00:33:03,600 --> 00:33:06,400
Speaker 3: advertising component, but will that look more like a Netflix

605
00:33:06,440 --> 00:33:08,920
Speaker 3: where it's like there's sort of two tiers where you

606
00:33:08,920 --> 00:33:11,600
Speaker 3: can pay for a subscription and never get any sponsored content.

607
00:33:12,400 --> 00:33:15,000
Speaker 3: You know, essentially paying Google almost for like a type

608
00:33:15,000 --> 00:33:18,360
Speaker 3: of ad blocker, yeah, which should almost be genius from Google,

609
00:33:18,800 --> 00:33:21,520
Speaker 3: or actually just paying to see the ads, and they

610
00:33:21,560 --> 00:33:23,280
Speaker 3: get paid both ways as well.

611
00:33:23,360 --> 00:33:26,120
Speaker 4: So it'll be interesting to see how that changes.

612
00:33:25,880 --> 00:33:27,920
Speaker 2: If you do feature in an AI summary. How are

613
00:33:27,960 --> 00:33:30,320
Speaker 2: the metrics potentially going to have to change when it's

614
00:33:30,360 --> 00:33:33,680
Speaker 2: not necessarily all going to be about clickthrough and impressions.

615
00:33:34,240 --> 00:33:36,840
Speaker 3: Yeah, it is really interesting and definitely a question that

616
00:33:37,080 --> 00:33:42,440
Speaker 3: all advertisers are asking. Fundamentally though, we need to really

617
00:33:42,480 --> 00:33:46,520
Speaker 3: think about in terms of attribution, moving away from impressions

618
00:33:46,520 --> 00:33:51,440
Speaker 3: and clicks anyway, in my opinion, I fund you know,

619
00:33:51,680 --> 00:33:55,560
Speaker 3: Atlas Digital obviously measures to a very great degree right

620
00:33:55,600 --> 00:33:58,320
Speaker 3: through to close one business, how much revenue we're actually

621
00:33:58,400 --> 00:34:01,800
Speaker 3: generating for a customer. Somations useful for us. But what

622
00:34:01,840 --> 00:34:05,320
Speaker 3: we do know is that when you focus on top

623
00:34:05,440 --> 00:34:09,840
Speaker 3: level brand awareness and make people feel something about your brand,

624
00:34:09,920 --> 00:34:13,399
Speaker 3: that's generally what tends to drive the majority of new

625
00:34:13,480 --> 00:34:17,520
Speaker 3: users into your company, especially as you grow. And so

626
00:34:17,600 --> 00:34:20,240
Speaker 3: I think that brand advertising is going to be really

627
00:34:20,280 --> 00:34:22,799
Speaker 3: important and that you just need to really focus on.

628
00:34:23,400 --> 00:34:25,400
Speaker 3: There needs to be almost some sort of metric to

629
00:34:25,520 --> 00:34:29,200
Speaker 3: understand the quality of the experience someone had, and maybe

630
00:34:29,200 --> 00:34:31,799
Speaker 3: that will be solved by things like, you know, you've

631
00:34:31,840 --> 00:34:35,160
Speaker 3: got your laptop camera and can recognize the emotions in

632
00:34:35,200 --> 00:34:38,080
Speaker 3: your face when you're looking at something. Obviously there's hover

633
00:34:38,160 --> 00:34:40,200
Speaker 3: over rates and other things like that, but I think

634
00:34:40,200 --> 00:34:43,359
Speaker 3: it's these kind of micro engagements that will start being

635
00:34:43,360 --> 00:34:45,799
Speaker 3: measured to a greater degree so that we can know

636
00:34:45,880 --> 00:34:49,080
Speaker 3: if we're eliciting positive responses around what the person's seeing

637
00:34:49,239 --> 00:34:49,920
Speaker 3: in our brand.

638
00:34:50,080 --> 00:34:52,560
Speaker 2: We have seen sort of the dark side of AI

639
00:34:52,640 --> 00:34:58,040
Speaker 2: generated search already. You know, Gemini and Google's oversights. AI

640
00:34:58,080 --> 00:35:02,799
Speaker 2: insights had a rough start putting glue on pizza that thing.

641
00:35:03,160 --> 00:35:07,960
Speaker 2: So there's that aspect, the hallucinations, the misinformation that could

642
00:35:08,000 --> 00:35:11,600
Speaker 2: potentially be built into those If they see content being

643
00:35:11,600 --> 00:35:15,600
Speaker 2: engaged with on Reddit in a huge way, it might

644
00:35:15,680 --> 00:35:18,920
Speaker 2: be featured on Google, but it might be wrong. So

645
00:35:18,960 --> 00:35:21,200
Speaker 2: there's that problem. Then, is I guess the ethical issue

646
00:35:21,239 --> 00:35:25,200
Speaker 2: off where they're scraping content from and who gets rewarded

647
00:35:25,200 --> 00:35:28,600
Speaker 2: for producing that content If it doesn't have a product

648
00:35:28,640 --> 00:35:30,680
Speaker 2: you can buy at the end of it, someone's just

649
00:35:30,719 --> 00:35:34,799
Speaker 2: creating great quality content which Google is scooping up.

650
00:35:35,280 --> 00:35:38,920
Speaker 3: Yeah, I definitely think that the hallucinations component will be

651
00:35:39,000 --> 00:35:41,799
Speaker 3: solved in time or at least obviously reduce and the

652
00:35:42,000 --> 00:35:46,200
Speaker 3: frequency that it happens over time. The licensing component is

653
00:35:46,280 --> 00:35:49,120
Speaker 3: quite interesting to think about, and I'm not too sure

654
00:35:49,360 --> 00:35:53,000
Speaker 3: how Google is going to really deal with scooping that content,

655
00:35:53,040 --> 00:35:56,160
Speaker 3: as you put it, from different websites in the years

656
00:35:56,200 --> 00:35:59,279
Speaker 3: to come. If we think about the current information and

657
00:35:59,360 --> 00:36:03,000
Speaker 3: the web, then we've got that's what all these llms

658
00:36:03,040 --> 00:36:04,759
Speaker 3: have been trained on. And if we think of that

659
00:36:04,800 --> 00:36:09,799
Speaker 3: from a static picture, producing new content right now potentially

660
00:36:09,920 --> 00:36:13,799
Speaker 3: has become less advantageous. If you're a business, you might

661
00:36:13,840 --> 00:36:17,720
Speaker 3: want to produce fewer pieces of content that's more original experience,

662
00:36:17,880 --> 00:36:20,600
Speaker 3: as we've been discussing, or you can simply use an

663
00:36:20,680 --> 00:36:23,799
Speaker 3: LLM to mass produce content that effectively exists already and

664
00:36:23,840 --> 00:36:26,200
Speaker 3: doesn't really add anything to the ecosystem.

665
00:36:26,640 --> 00:36:27,720
Speaker 4: So, if we think.

666
00:36:27,520 --> 00:36:31,000
Speaker 3: About in two decades time, what's all the new content

667
00:36:31,080 --> 00:36:33,279
Speaker 3: going to be, people are going to have to produce it,

668
00:36:33,320 --> 00:36:35,920
Speaker 3: and Google or other searchings that are going to have

669
00:36:36,000 --> 00:36:39,560
Speaker 3: to provide some sort of way for people to get

670
00:36:39,600 --> 00:36:42,600
Speaker 3: some sort of commercial benefit from doing it. I actually

671
00:36:42,640 --> 00:36:45,600
Speaker 3: think that, you know, owning your own audience is the

672
00:36:45,640 --> 00:36:47,840
Speaker 3: best thing that you can do at the moment, whether

673
00:36:47,880 --> 00:36:50,800
Speaker 3: that's all your followers on LinkedIn, and you know YouTube

674
00:36:50,840 --> 00:36:53,400
Speaker 3: does it so well with all their creators monetizing all

675
00:36:53,440 --> 00:36:56,319
Speaker 3: of the audiences that they've built. And then over time,

676
00:36:56,360 --> 00:36:59,280
Speaker 3: I think that there will be opportunities to show and Google,

677
00:36:59,280 --> 00:37:02,200
Speaker 3: whether it's through life licensing or advertising or whatever it

678
00:37:02,200 --> 00:37:02,520
Speaker 3: may be.

679
00:37:03,200 --> 00:37:06,359
Speaker 2: So then if you do own your own customer base,

680
00:37:06,520 --> 00:37:11,480
Speaker 2: essentially we're talking about things like email distribution lists, maybe

681
00:37:11,640 --> 00:37:15,040
Speaker 2: setting up a forum for them. And if you can

682
00:37:15,040 --> 00:37:17,880
Speaker 2: do that on your own platform as opposed to just

683
00:37:17,920 --> 00:37:21,680
Speaker 2: being in a Facebook group or on LinkedIn, the insights

684
00:37:21,680 --> 00:37:23,600
Speaker 2: you're going to get about their behavior and what they

685
00:37:23,640 --> 00:37:24,840
Speaker 2: want are going to be that much greater.

686
00:37:25,280 --> 00:37:28,160
Speaker 3: Yeah, exactly, And I think that there's something that exists

687
00:37:28,160 --> 00:37:31,080
Speaker 3: for all of them, right, Like larger enterprises will be

688
00:37:31,120 --> 00:37:33,200
Speaker 3: able to build their own communities. I know law View

689
00:37:33,239 --> 00:37:37,319
Speaker 3: has done a fantastic job of building a community. I

690
00:37:37,360 --> 00:37:40,440
Speaker 3: think total on a B to B SaaS company. Whereas

691
00:37:40,480 --> 00:37:42,319
Speaker 3: if you're a smaller business, it might be that you

692
00:37:42,440 --> 00:37:45,360
Speaker 3: just have a subreddit and you're just constantly in that subreddit.

693
00:37:45,360 --> 00:37:47,960
Speaker 3: You might have a Facebook group, and certainly there's email

694
00:37:48,000 --> 00:37:50,600
Speaker 3: distribution lists and LinkedIn groups that you can.

695
00:37:50,480 --> 00:37:51,080
Speaker 4: Build as well.

696
00:37:51,120 --> 00:37:53,680
Speaker 3: And so I think it's just not really putting all

697
00:37:53,680 --> 00:37:57,279
Speaker 3: of your eggs in one basket and then seeing how

698
00:37:57,320 --> 00:37:58,400
Speaker 3: things play out.

699
00:37:58,680 --> 00:38:03,600
Speaker 2: Yeah, what's your sort of vision for where search is

700
00:38:03,640 --> 00:38:05,759
Speaker 2: going to go in sort of the next decade? If

701
00:38:05,760 --> 00:38:08,959
Speaker 2: we're looking big picture, it's the game has been sewn

702
00:38:09,040 --> 00:38:11,840
Speaker 2: up by Google for so long? Is this really the

703
00:38:11,880 --> 00:38:16,239
Speaker 2: starting point of a proliferation? We've got perplexity, We've got

704
00:38:16,280 --> 00:38:20,359
Speaker 2: all these stability, we've got all these AI companies. Are

705
00:38:20,360 --> 00:38:22,920
Speaker 2: they going to be able to hop over that wall

706
00:38:22,960 --> 00:38:27,000
Speaker 2: that was insurmountable before by building sort of search related

707
00:38:27,040 --> 00:38:29,960
Speaker 2: functions into their own AI products to compete with Google?

708
00:38:30,520 --> 00:38:33,560
Speaker 3: Yeah, I think that the search market definitely has been

709
00:38:33,560 --> 00:38:35,880
Speaker 3: a win it takes all scenario. I think that the

710
00:38:35,960 --> 00:38:38,200
Speaker 3: sad reality is that up until and I don't know

711
00:38:38,200 --> 00:38:40,200
Speaker 3: if this has changed, but up until about a year ago,

712
00:38:40,360 --> 00:38:43,200
Speaker 3: the most search thing in being was Google, or at

713
00:38:43,280 --> 00:38:48,440
Speaker 3: least the top five, which is really demonstrative of Google's

714
00:38:48,480 --> 00:38:52,920
Speaker 3: dominance in the market. It's possible that new entrants do

715
00:38:53,040 --> 00:38:55,920
Speaker 3: come and shake that up quite significantly, but we also

716
00:38:55,960 --> 00:38:58,760
Speaker 3: need to take into account when thinking about the future

717
00:38:58,800 --> 00:39:01,719
Speaker 3: of search that whole adoption cycle as well.

718
00:39:01,800 --> 00:39:03,200
Speaker 4: It doesn't.

719
00:39:03,520 --> 00:39:06,120
Speaker 3: While it seems like ais everywhere, a lot of people

720
00:39:06,160 --> 00:39:08,640
Speaker 3: aren't using it outside of the bubbles that we might

721
00:39:08,680 --> 00:39:11,239
Speaker 3: exist in, a lot of people have never even used it,

722
00:39:11,280 --> 00:39:13,200
Speaker 3: and they're still using Google in a big way.

723
00:39:13,560 --> 00:39:14,520
Speaker 4: And so I think that the.

724
00:39:14,520 --> 00:39:17,880
Speaker 3: Actual time it will take before people have migrated to

725
00:39:17,920 --> 00:39:20,239
Speaker 3: a totally different system will be a very long time,

726
00:39:20,280 --> 00:39:23,680
Speaker 3: and certainly the younger generation might adopt it earlier.

727
00:39:24,600 --> 00:39:25,840
Speaker 4: Could be the case.

728
00:39:25,880 --> 00:39:30,320
Speaker 3: But likewise, Google has a massive war chest of capital

729
00:39:30,360 --> 00:39:33,200
Speaker 3: that they can invest and they've got Gemini rolled out. Obviously,

730
00:39:33,520 --> 00:39:35,359
Speaker 3: eating a rock a day isn't the type of thing

731
00:39:35,400 --> 00:39:37,560
Speaker 3: that's great for their brand, but they do have a

732
00:39:37,600 --> 00:39:40,640
Speaker 3: fantastic brand that people do love. Like Google, it is

733
00:39:40,719 --> 00:39:43,120
Speaker 3: literally a verb you know, you don't bring it, And

734
00:39:43,160 --> 00:39:48,040
Speaker 3: so I think to unshake that solid position they've got

735
00:39:48,120 --> 00:39:51,600
Speaker 3: will actually be quite difficult. Where I think commercially, in

736
00:39:51,719 --> 00:39:56,600
Speaker 3: terms of digital marketing, the search market is going, I'm

737
00:39:56,600 --> 00:39:58,759
Speaker 3: not too sure. I think that I've got a couple

738
00:39:58,840 --> 00:40:01,760
Speaker 3: of views on it. I think think that advertising models

739
00:40:01,800 --> 00:40:04,600
Speaker 3: will change quite a bit away from pay per click.

740
00:40:05,520 --> 00:40:06,719
Speaker 4: You know, if you're going to.

741
00:40:06,719 --> 00:40:09,799
Speaker 3: Reduce the attribution that you can show in terms of

742
00:40:09,920 --> 00:40:13,200
Speaker 3: how the efficacy of using those channels, then it makes

743
00:40:13,239 --> 00:40:16,040
Speaker 3: sense that they will start to stack up less versus

744
00:40:16,080 --> 00:40:19,440
Speaker 3: other channels because you simply just don't know the roy

745
00:40:20,280 --> 00:40:21,799
Speaker 3: So I think that there might be more of a

746
00:40:21,840 --> 00:40:27,480
Speaker 3: shift maybe to more subscription driven businesses versus just pure

747
00:40:27,600 --> 00:40:32,600
Speaker 3: like pay per click models. That's probably my main view.

748
00:40:32,880 --> 00:40:36,600
Speaker 3: I also just think that it's important not to shoot

749
00:40:36,600 --> 00:40:40,759
Speaker 3: your shot too early, to iteratively change your business as

750
00:40:40,760 --> 00:40:43,839
Speaker 3: you go and adjust to the times, but don't go

751
00:40:44,000 --> 00:40:46,560
Speaker 3: all in, because I think this is really just scratching

752
00:40:46,600 --> 00:40:49,520
Speaker 3: the surface of, you know, where everything's going to go.

753
00:40:56,200 --> 00:40:58,680
Speaker 1: In some ways, I guess it seems like Ryan is

754
00:40:58,719 --> 00:41:01,799
Speaker 1: saying that things won't actually change that much. There's still

755
00:41:01,840 --> 00:41:04,080
Speaker 1: going to be a lot of energy and effort put

756
00:41:04,120 --> 00:41:07,160
Speaker 1: into figuring out how to be ahead of the game

757
00:41:07,160 --> 00:41:11,200
Speaker 1: when it comes to search compared to others, and it's

758
00:41:11,239 --> 00:41:13,480
Speaker 1: going to take specialists. It's going to take time and

759
00:41:13,560 --> 00:41:15,439
Speaker 1: energy to actually learn how to do that. But it's

760
00:41:15,520 --> 00:41:17,800
Speaker 1: just the way that you do it might be changing

761
00:41:17,840 --> 00:41:18,600
Speaker 1: a little bit.

762
00:41:19,040 --> 00:41:22,200
Speaker 2: Yeah, and where you put your effort not necessarily just

763
00:41:22,760 --> 00:41:24,960
Speaker 2: you know, in terms of where you spend money on

764
00:41:25,960 --> 00:41:30,440
Speaker 2: paid search and social media, but in terms of building

765
00:41:31,120 --> 00:41:33,760
Speaker 2: that sort of content base that is going to appeal

766
00:41:33,920 --> 00:41:37,800
Speaker 2: to search engines, not just Googles, but any search engine

767
00:41:37,800 --> 00:41:40,640
Speaker 2: that it has AI at the heart of it, it's

768
00:41:40,680 --> 00:41:46,040
Speaker 2: all about experiential content. It's about how to This was

769
00:41:46,080 --> 00:41:48,919
Speaker 2: what I did with this product, and this is why

770
00:41:48,960 --> 00:41:51,560
Speaker 2: I liked it or didn't like it, which is why

771
00:41:51,600 --> 00:41:53,800
Speaker 2: I read it and that I have suddenly seen a

772
00:41:53,880 --> 00:41:58,359
Speaker 2: surge in traffic. So, thinking about any company, whether you're

773
00:41:58,360 --> 00:42:00,160
Speaker 2: a B to B or a B two C company,

774
00:42:00,280 --> 00:42:04,120
Speaker 2: how do I build a platform or a community or

775
00:42:04,160 --> 00:42:06,560
Speaker 2: a forum that creates that sort of content that is

776
00:42:06,600 --> 00:42:08,760
Speaker 2: going to be the real rich stuff that when Google

777
00:42:08,840 --> 00:42:13,560
Speaker 2: indexes my site, the algorithm goes, ooh, this is really

778
00:42:13,680 --> 00:42:18,840
Speaker 2: useful storytelling or experiential evidence. I want to use that

779
00:42:19,280 --> 00:42:22,080
Speaker 2: to feature in an AI overview of some sort of summary.

780
00:42:22,600 --> 00:42:24,680
Speaker 2: That's what people are going to have to be thinking about,

781
00:42:24,719 --> 00:42:29,160
Speaker 2: and that is quite tricky because maintaining a community takes

782
00:42:29,160 --> 00:42:30,960
Speaker 2: a lot of work. You have to sort of police

783
00:42:30,960 --> 00:42:32,880
Speaker 2: it and make sure that it's a nice place to be.

784
00:42:33,640 --> 00:42:36,160
Speaker 2: We saw that in the nineties early two thousands, and

785
00:42:36,280 --> 00:42:38,879
Speaker 2: with the rise of blogging, it was a great time

786
00:42:38,920 --> 00:42:42,160
Speaker 2: for the Internet because there were genuine communities that were moderated,

787
00:42:43,400 --> 00:42:47,000
Speaker 2: usually with a light touch, and it was great commentary

788
00:42:47,000 --> 00:42:49,040
Speaker 2: going on in those places. We're sort of seeing return

789
00:42:49,120 --> 00:42:52,160
Speaker 2: to that sort of goal that comes out of those communities.

790
00:42:52,160 --> 00:42:55,600
Speaker 2: But it's quite a big investment to sort of especially

791
00:42:55,600 --> 00:42:57,480
Speaker 2: if you're doing it off your own platforms, to sort

792
00:42:57,480 --> 00:42:58,600
Speaker 2: of build that and maintain it.

793
00:42:59,120 --> 00:43:02,520
Speaker 1: Yeah, I think we may, you know, unintentionally, see a

794
00:43:02,560 --> 00:43:05,279
Speaker 1: new arms race in terms of that content creation. You

795
00:43:05,320 --> 00:43:08,479
Speaker 1: already have probably seen those gifts of videos going around

796
00:43:08,480 --> 00:43:12,320
Speaker 1: the Internet of people with a literal wall of mobile

797
00:43:12,400 --> 00:43:15,440
Speaker 1: phones that are all plugged into a central computer and

798
00:43:15,440 --> 00:43:18,360
Speaker 1: they're just going through and putting fake reviews up and

799
00:43:18,400 --> 00:43:20,359
Speaker 1: doing all this kind of stuff. So no doubt we'll

800
00:43:20,360 --> 00:43:22,400
Speaker 1: start to see people trying to game the system with

801
00:43:23,960 --> 00:43:28,640
Speaker 1: some fakery as well. I don't know how that can

802
00:43:29,120 --> 00:43:33,439
Speaker 1: be done. In terms of creating your own community on

803
00:43:33,480 --> 00:43:36,600
Speaker 1: your own website, I don't think that's something that necessarily

804
00:43:37,320 --> 00:43:39,360
Speaker 1: will be able to be done. It's going to be

805
00:43:39,360 --> 00:43:43,319
Speaker 1: about creating communities on existing forums. Like you mentioned Reddit, right,

806
00:43:43,360 --> 00:43:45,760
Speaker 1: and you have your subreddits, but you also have people

807
00:43:45,800 --> 00:43:49,040
Speaker 1: mentioning things in various reddits, so you might have an

808
00:43:49,120 --> 00:43:51,360
Speaker 1: art reddit and people will ask what are the best

809
00:43:51,719 --> 00:43:56,600
Speaker 1: paint brushes? That subreddit isn't controlled by I don't know

810
00:43:57,080 --> 00:43:59,359
Speaker 1: what's an art brand. I really should have picked something

811
00:43:59,400 --> 00:44:05,000
Speaker 1: I know anything about that isn't necessarily controlled by a

812
00:44:05,040 --> 00:44:09,960
Speaker 1: particular art brand. So it's about getting your valued customers

813
00:44:10,040 --> 00:44:14,040
Speaker 1: to engage with you in those common spaces online.

814
00:44:14,440 --> 00:44:17,640
Speaker 2: Yeah, and the other thing that really struck me that

815
00:44:17,719 --> 00:44:21,600
Speaker 2: Ryan said towards the end, there is the rise of

816
00:44:21,719 --> 00:44:25,839
Speaker 2: gen Ai built into personal assistance and voice assistance, and

817
00:44:26,080 --> 00:44:30,360
Speaker 2: the changing nature off searching for things as a result

818
00:44:30,440 --> 00:44:34,840
Speaker 2: of that, where we sort of outsource these search queries

819
00:44:34,920 --> 00:44:38,240
Speaker 2: to a voice assistant. To date, that's been a pretty

820
00:44:38,239 --> 00:44:43,200
Speaker 2: crappy experience. It's very clunky. But what Amazon is going

821
00:44:43,239 --> 00:44:46,719
Speaker 2: to try and do with Alexa and Apple with Siri,

822
00:44:46,800 --> 00:44:49,480
Speaker 2: with Apple Intelligence and all of that will elevate that.

823
00:44:49,640 --> 00:44:53,520
Speaker 2: So that may mean that impressions and clicks are less

824
00:44:53,560 --> 00:44:57,200
Speaker 2: important in future, and that will be a great relief

825
00:44:57,239 --> 00:45:00,160
Speaker 2: to a lot of people whose lives are dictated by

826
00:45:00,200 --> 00:45:01,960
Speaker 2: those sort of metrics at the moment. And I think

827
00:45:02,000 --> 00:45:04,000
Speaker 2: at Hired, a lot of people know that those are

828
00:45:04,040 --> 00:45:07,880
Speaker 2: hollow measurements. But what do you replaced them with. So

829
00:45:08,160 --> 00:45:12,240
Speaker 2: if there is some other ways of measuring success through

830
00:45:12,560 --> 00:45:15,000
Speaker 2: searching for things in a different way and being stepped

831
00:45:15,040 --> 00:45:18,839
Speaker 2: through processes how to make a recipe, a dish based

832
00:45:18,880 --> 00:45:23,120
Speaker 2: on a recipe, how to go travel, and companies are

833
00:45:23,160 --> 00:45:25,440
Speaker 2: rewarded along the way, I think a lot of people

834
00:45:25,440 --> 00:45:26,160
Speaker 2: will welcome that.

835
00:45:26,920 --> 00:45:31,360
Speaker 1: Yeah, you know, it's a nice idea of the future

836
00:45:31,400 --> 00:45:33,319
Speaker 1: of search, isn't it that we would be getting our

837
00:45:33,360 --> 00:45:37,160
Speaker 1: information from these people who are experts and experienced and

838
00:45:37,680 --> 00:45:40,080
Speaker 1: have done it and can talk about it with that authority.

839
00:45:40,600 --> 00:45:43,799
Speaker 1: I think traditionally we haven't really seen that shakeout in

840
00:45:43,880 --> 00:45:47,520
Speaker 1: the Internet when it comes to corporate interests and advertising.

841
00:45:47,680 --> 00:45:52,120
Speaker 1: So maybe this will be the golden time. Maybe this

842
00:45:52,160 --> 00:45:54,080
Speaker 1: will be the golden age of the Internet in terms

843
00:45:54,120 --> 00:45:59,200
Speaker 1: of how we get information from other people and authoritative sources.

844
00:45:59,239 --> 00:46:00,000
Speaker 1: That would be nice.

845
00:46:00,400 --> 00:46:04,719
Speaker 2: And literally the day after we spoke to Ryan, the

846
00:46:04,800 --> 00:46:08,800
Speaker 2: news broke about this Department of Justice case against Google,

847
00:46:09,239 --> 00:46:13,120
Speaker 2: basically saying confirming in court Google is a monopoly and

848
00:46:13,160 --> 00:46:15,360
Speaker 2: we are going to now take some time to decide

849
00:46:15,400 --> 00:46:18,920
Speaker 2: what to do about it, and that could include everything

850
00:46:19,000 --> 00:46:24,320
Speaker 2: from telling them they can't pay Apple and Mozilla billions

851
00:46:24,320 --> 00:46:27,160
Speaker 2: of dollars to favor the Google search engine anymore, all

852
00:46:27,200 --> 00:46:28,719
Speaker 2: the way through to We're going to break you up

853
00:46:29,360 --> 00:46:32,279
Speaker 2: and how that might work. That's got years to play out,

854
00:46:33,040 --> 00:46:36,400
Speaker 2: so we won't see any immediate results from that. But

855
00:46:36,600 --> 00:46:39,600
Speaker 2: another fact there, and maybe something to your sort of

856
00:46:39,880 --> 00:46:43,120
Speaker 2: optimistic view about the era that we may be entering.

857
00:46:43,200 --> 00:46:46,400
Speaker 2: If we do have three or four strong players in search,

858
00:46:47,000 --> 00:46:49,399
Speaker 2: what that would mean for the average Kiwi company that's

859
00:46:49,440 --> 00:46:51,840
Speaker 2: going out to try and buy some Adam ventry on

860
00:46:51,960 --> 00:46:54,960
Speaker 2: a search engine. Maybe it's going to lead to them

861
00:46:55,000 --> 00:46:57,120
Speaker 2: paying a lot less and therefo we're able to invest

862
00:46:57,160 --> 00:46:59,640
Speaker 2: more to find an audience. That would be a great thing.

863
00:47:00,040 --> 00:47:03,319
Speaker 1: Yeah, he is hoping anyway. I feel like I say

864
00:47:03,320 --> 00:47:06,120
Speaker 1: that a lot at the end of our episodes, a

865
00:47:06,120 --> 00:47:08,680
Speaker 1: lot of hope. Hope is good. We finished on an

866
00:47:08,719 --> 00:47:09,840
Speaker 1: optimistic note.

867
00:47:11,000 --> 00:47:14,120
Speaker 2: Thanks to Ryan McMillan from Atlas Digital for joining us

868
00:47:14,160 --> 00:47:16,480
Speaker 2: this week. I learned a lot and I hope you

869
00:47:16,520 --> 00:47:16,960
Speaker 2: did too.

870
00:47:17,239 --> 00:47:19,319
Speaker 1: And don't forget. The Business of Tech is on all

871
00:47:19,400 --> 00:47:22,960
Speaker 1: major podcasts platforms as well as iHeartRadio, where you can

872
00:47:22,960 --> 00:47:25,560
Speaker 1: stream every episode. Show notes are in the Tech section

873
00:47:25,680 --> 00:47:28,360
Speaker 1: on the Business Desk website. Please leave us a review

874
00:47:28,640 --> 00:47:30,239
Speaker 1: and share the podcast with your.

875
00:47:30,160 --> 00:47:33,280
Speaker 2: Friends and colleagues, and get in touch with your feedback, ideas,

876
00:47:33,360 --> 00:47:38,000
Speaker 2: topics and guest suggestions. Email Ben on Ben at businessdesk

877
00:47:38,040 --> 00:47:40,480
Speaker 2: dot co dot nz. You can find both of us

878
00:47:40,960 --> 00:47:42,640
Speaker 2: looking on LinkedIn and x as well.

879
00:47:43,000 --> 00:47:45,319
Speaker 1: Stay tuned for another dose of the business of tech

880
00:47:45,400 --> 00:47:47,320
Speaker 1: coming your way next Thursday morning.

881
00:47:47,800 --> 00:47:48,359
Speaker 2: Catch you then,