1 00:00:00,120 --> 00:00:03,480 Speaker 1: It's been a big year from mergers and acquisitions. As 2 00:00:03,520 --> 00:00:07,280 Speaker 1: a tough economy, competitors to join forces and try and 3 00:00:07,320 --> 00:00:10,119 Speaker 1: achieve scale here and beyond Australasia. 4 00:00:10,200 --> 00:00:12,520 Speaker 2: But what does it mean in practical terms when two 5 00:00:12,520 --> 00:00:16,280 Speaker 2: companies merge? How do you spice everything together in an 6 00:00:16,400 --> 00:00:17,000 Speaker 2: orderly way? 7 00:00:17,200 --> 00:00:19,959 Speaker 1: Well? Two degrees is in the process of doing just 8 00:00:20,280 --> 00:00:23,800 Speaker 1: that following its merger with Vocus to create New Zealand's 9 00:00:23,840 --> 00:00:25,840 Speaker 1: third largest telecoms operator. 10 00:00:25,960 --> 00:00:28,920 Speaker 2: On the sponsored episode of the Business at Tech, Stephen Kerjier, 11 00:00:29,160 --> 00:00:32,240 Speaker 2: two degrees is chief information officer, talks to us about 12 00:00:32,280 --> 00:00:35,960 Speaker 2: the major job of bringing two large businesses together with 13 00:00:36,159 --> 00:00:39,600 Speaker 2: two million users on the platform and what he's learned 14 00:00:39,640 --> 00:00:42,600 Speaker 2: so far in the two Degrees Vocus merger. 15 00:00:42,360 --> 00:00:45,159 Speaker 1: And Stephen joins us. Now, Steven, thanks so much for 16 00:00:45,240 --> 00:00:47,600 Speaker 1: coming on the Business of Tech. How are you doing. 17 00:00:47,880 --> 00:00:49,640 Speaker 3: I'm doing great. Thanks for having me. Yeah, gat to 18 00:00:49,640 --> 00:00:50,159 Speaker 3: be here soon. 19 00:00:50,200 --> 00:00:52,280 Speaker 1: We're going to get into some really interesting things that 20 00:00:52,320 --> 00:00:57,160 Speaker 1: you've been working on around integration of these telecommunications businesses 21 00:00:57,200 --> 00:01:00,320 Speaker 1: from Vocus to two degrees. So lots of interesting stuff 22 00:01:00,360 --> 00:01:03,240 Speaker 1: to cover there. But you're also joining us for our 23 00:01:03,480 --> 00:01:06,360 Speaker 1: usual sort of run around the big tech headlines off 24 00:01:06,360 --> 00:01:09,400 Speaker 1: the week and a Talco story that's a particular interest 25 00:01:09,480 --> 00:01:13,399 Speaker 1: to you. The Commerce Commission is asking our big telcos, 26 00:01:13,440 --> 00:01:17,640 Speaker 1: including two Degrees, One End, Zed and Spark, to start 27 00:01:17,680 --> 00:01:21,200 Speaker 1: showing coverage maps. Now I thought they already did this. 28 00:01:21,280 --> 00:01:23,560 Speaker 1: You go onto your website, you'll see a coverage map there, 29 00:01:23,600 --> 00:01:26,800 Speaker 1: but they're looking for something a bit more extensive. What's 30 00:01:26,880 --> 00:01:27,440 Speaker 1: your take on this. 31 00:01:28,000 --> 00:01:30,920 Speaker 3: Yeah, I think we've all done it, but in different ways. 32 00:01:31,400 --> 00:01:33,959 Speaker 3: So we have a coverage map on our website. I 33 00:01:33,959 --> 00:01:37,080 Speaker 3: think it's actually a really good thing. It provides consistency 34 00:01:37,120 --> 00:01:40,280 Speaker 3: to our consumers on coverage, so when they sign up, 35 00:01:40,319 --> 00:01:42,720 Speaker 3: they know what they're actually getting. I think for two 36 00:01:42,720 --> 00:01:45,000 Speaker 3: Degrees in particular, how we've always been the challenger in 37 00:01:45,040 --> 00:01:48,280 Speaker 3: the market. One of our bigger issues is network perception. 38 00:01:48,520 --> 00:01:50,080 Speaker 3: You know, do you actually have the coverage or not? 39 00:01:50,160 --> 00:01:53,240 Speaker 3: Given our history, and we've invested in our network significantly 40 00:01:53,520 --> 00:01:56,200 Speaker 3: over the last three to five years and we have 41 00:01:56,320 --> 00:01:58,400 Speaker 3: parity now, so we actually think it's a great thing 42 00:01:58,440 --> 00:02:01,440 Speaker 3: for us to show. So, you know, is coverage poor, 43 00:02:01,480 --> 00:02:04,120 Speaker 3: is it good? Is it great? That transparency I think 44 00:02:04,200 --> 00:02:07,320 Speaker 3: is really important for consumers. How hard is to do 45 00:02:07,600 --> 00:02:08,520 Speaker 3: is probably another question. 46 00:02:09,160 --> 00:02:11,720 Speaker 1: That's the thing, because the maps are there, the physical 47 00:02:12,080 --> 00:02:14,280 Speaker 1: you know, your engineers go out and map this is 48 00:02:14,280 --> 00:02:17,680 Speaker 1: where the coverage is. But then putting a rating on 49 00:02:17,720 --> 00:02:20,440 Speaker 1: that is this is really good quality coverage, This is 50 00:02:20,480 --> 00:02:22,160 Speaker 1: a bit miginal. I guess that's where it will be 51 00:02:22,160 --> 00:02:23,120 Speaker 1: a little bit contentious. 52 00:02:23,160 --> 00:02:25,200 Speaker 2: And how do you actually measure it because I'm sure 53 00:02:25,200 --> 00:02:27,720 Speaker 2: there would be fluctuations and it would depend on capacity 54 00:02:27,760 --> 00:02:30,560 Speaker 2: of the towers and things like that. So how are 55 00:02:30,560 --> 00:02:32,280 Speaker 2: you thinking about that nuance? 56 00:02:32,919 --> 00:02:36,160 Speaker 3: Yeah, in one way, you've got kind of network kind 57 00:02:36,160 --> 00:02:39,120 Speaker 3: of measures, you know, signal strength, and those can be 58 00:02:39,240 --> 00:02:42,280 Speaker 3: quite well defined. But I think ultimately its customer experience 59 00:02:42,360 --> 00:02:44,680 Speaker 3: is the most important part of it. So I think 60 00:02:44,720 --> 00:02:48,480 Speaker 3: coverage maps will give you a broad view, but it 61 00:02:48,560 --> 00:02:51,800 Speaker 3: may not show kind of end to end network experience. 62 00:02:51,880 --> 00:02:54,639 Speaker 3: So I think I'll give us enough of a good 63 00:02:54,680 --> 00:02:57,320 Speaker 3: measure for consumers to make the right decisions, particularly in 64 00:02:57,360 --> 00:03:00,760 Speaker 3: the more rural areas. But our focus atally around customer 65 00:03:00,840 --> 00:03:03,120 Speaker 3: experience and how can we measure that kind of that 66 00:03:03,200 --> 00:03:06,079 Speaker 3: true end to end view, which is around collecting the 67 00:03:06,160 --> 00:03:10,880 Speaker 3: right data from the handset, either crowdsourced or you can 68 00:03:10,880 --> 00:03:13,960 Speaker 3: also get some really interesting telemetry from the network as 69 00:03:14,000 --> 00:03:17,360 Speaker 3: well on the user experience and also independent testing of 70 00:03:17,400 --> 00:03:21,840 Speaker 3: the network. So recently we've actually done really well in 71 00:03:21,880 --> 00:03:24,800 Speaker 3: some independent tests, so that's another measure. But from the 72 00:03:24,800 --> 00:03:27,919 Speaker 3: Commerce Commission, it's really just get a baseline on what 73 00:03:28,320 --> 00:03:30,880 Speaker 3: good coverage is and make it consistent, which I think 74 00:03:30,919 --> 00:03:32,520 Speaker 3: is actually a great thing to do, but it is 75 00:03:33,080 --> 00:03:34,880 Speaker 3: probably only one measure. It's not going to be the 76 00:03:34,920 --> 00:03:37,640 Speaker 3: whole story to your question. Yeah. 77 00:03:37,720 --> 00:03:39,640 Speaker 2: Yeah, And the other part of it is that the 78 00:03:39,680 --> 00:03:43,040 Speaker 2: commissioner is looking to have an exit right put into 79 00:03:43,040 --> 00:03:46,000 Speaker 2: contract so that if the customer signs up and they 80 00:03:46,000 --> 00:03:48,120 Speaker 2: find that, like I think it was thirty percent of 81 00:03:48,160 --> 00:03:51,480 Speaker 2: the small businesses that they surveyed that the coverage they're 82 00:03:51,480 --> 00:03:53,800 Speaker 2: not happy with what they're getting, that there is a 83 00:03:53,840 --> 00:03:57,600 Speaker 2: clause saying I want to exit based on this issue. Now, 84 00:03:57,600 --> 00:04:00,240 Speaker 2: two degrees is one of the tolcoes that's already that 85 00:04:00,280 --> 00:04:03,600 Speaker 2: in their contract for some time. Do you find that 86 00:04:03,720 --> 00:04:06,120 Speaker 2: is something that's triggered a lot, or is it just 87 00:04:06,400 --> 00:04:09,520 Speaker 2: gives you the confidence to bring on for customers to 88 00:04:09,640 --> 00:04:13,000 Speaker 2: join you, Like, how does that work out in the field. 89 00:04:13,480 --> 00:04:17,159 Speaker 3: Yeah, well, I say bring it on. Yeah, we've already 90 00:04:17,160 --> 00:04:21,160 Speaker 3: had it. You know, sort of money back guarantee network reliability. 91 00:04:21,279 --> 00:04:24,479 Speaker 3: More recently we've been doing more advertising on that, like 92 00:04:24,480 --> 00:04:27,159 Speaker 3: a network blind test. You can actually try the network out, 93 00:04:27,320 --> 00:04:30,760 Speaker 3: give it a go again. We're really focused on our 94 00:04:30,839 --> 00:04:35,000 Speaker 3: network's fantastic. Now it's parody across all the players, but 95 00:04:35,080 --> 00:04:37,839 Speaker 3: we have this network perception issue so that when customers 96 00:04:37,839 --> 00:04:39,520 Speaker 3: get on our network they actually love it. 97 00:04:39,600 --> 00:04:42,239 Speaker 1: So yeah, that was a big thing, getting to network 98 00:04:42,240 --> 00:04:46,480 Speaker 1: parody as a third player, considering how long it took 99 00:04:46,560 --> 00:04:50,160 Speaker 1: for both Spark and Vodafone to really flesh out coverage. 100 00:04:50,200 --> 00:04:52,760 Speaker 1: There was a lot of complaints about those networks for 101 00:04:53,160 --> 00:04:55,440 Speaker 1: a long time to even in semi rural places, even 102 00:04:55,600 --> 00:04:59,320 Speaker 1: urban areas was patchy coverage. So I guess in some ways, 103 00:04:59,320 --> 00:05:00,960 Speaker 1: coming from behind and as a third player, it was 104 00:05:00,960 --> 00:05:02,520 Speaker 1: a bit easier because you knew some of the pain 105 00:05:02,560 --> 00:05:05,600 Speaker 1: points they'd been through. But still a huge undertaking. 106 00:05:05,480 --> 00:05:09,839 Speaker 3: Significant undertaking, and it keeps going right, it never stops. 107 00:05:10,040 --> 00:05:13,280 Speaker 3: We're investing hundreds of millions dollars in our network every year. 108 00:05:13,960 --> 00:05:17,520 Speaker 3: We've got a five G modernization program that we're well 109 00:05:17,560 --> 00:05:19,840 Speaker 3: in flight. We started late for five G compared to 110 00:05:19,880 --> 00:05:22,039 Speaker 3: our competitors, but we're well on track. I think we 111 00:05:22,080 --> 00:05:23,920 Speaker 3: may be one of the ones to finish first, just 112 00:05:23,960 --> 00:05:26,760 Speaker 3: in terms of our approach. So yeah, it never stops 113 00:05:26,920 --> 00:05:29,560 Speaker 3: and it's never perfect, Like network coverage is always a 114 00:05:30,080 --> 00:05:33,560 Speaker 3: thing where you're always looking to improve. But yeah, coming 115 00:05:33,600 --> 00:05:36,920 Speaker 3: from the origins of basically starting from nothing, right, two 116 00:05:36,960 --> 00:05:39,919 Speaker 3: degrees is a challenger, effectively a startup. We're a really 117 00:05:39,920 --> 00:05:43,839 Speaker 3: big startup now and we still think like a startup, 118 00:05:44,560 --> 00:05:47,679 Speaker 3: but quite an incredible journey from what was a position 119 00:05:47,760 --> 00:05:50,040 Speaker 3: of not owning a network to where we are now, 120 00:05:50,080 --> 00:05:54,120 Speaker 3: where we own over two thousand cell towers, We've got 121 00:05:54,160 --> 00:05:57,560 Speaker 3: fiber assets that are four thousand, six hundred kilometers. The 122 00:05:57,600 --> 00:05:59,680 Speaker 3: network's out of the equation for us, so our focus 123 00:05:59,720 --> 00:06:02,440 Speaker 3: is now on customer experience and software and all that 124 00:06:02,480 --> 00:06:06,000 Speaker 3: great stuff. It takes hard work and a commitment to 125 00:06:06,040 --> 00:06:08,960 Speaker 3: just keep investing for a long term strategy. 126 00:06:09,200 --> 00:06:11,880 Speaker 1: And of course you've got the text satellite stuff comes 127 00:06:12,040 --> 00:06:13,920 Speaker 1: as well, so you've got the coverage map, but then 128 00:06:14,120 --> 00:06:16,280 Speaker 1: you're going to have where the big gray spots are 129 00:06:16,320 --> 00:06:18,960 Speaker 1: at the moment, the ability to send text messages initially 130 00:06:19,000 --> 00:06:20,960 Speaker 1: anyway via satellite. 131 00:06:20,680 --> 00:06:23,080 Speaker 3: That's right, Yeah, and that's the exciting part coming up, 132 00:06:23,320 --> 00:06:25,680 Speaker 3: which we've partnered with link as I think you know 133 00:06:26,760 --> 00:06:30,359 Speaker 3: and still trialing that. And the time for that to 134 00:06:30,360 --> 00:06:32,039 Speaker 3: be effective is just how many birds you can have 135 00:06:32,080 --> 00:06:33,919 Speaker 3: in the sky, how many satellites you can have in 136 00:06:33,960 --> 00:06:36,039 Speaker 3: the sky. So yeah, we're really excited about that to 137 00:06:36,120 --> 00:06:40,520 Speaker 3: really plug those other areas of coverage. And I'll also 138 00:06:40,560 --> 00:06:42,760 Speaker 3: say it's an industry effort as well. You've got the 139 00:06:42,839 --> 00:06:46,240 Speaker 3: Rural Connectivity Group that has been looking at how we 140 00:06:46,279 --> 00:06:49,200 Speaker 3: do things more effective, so it's not just three players 141 00:06:49,200 --> 00:06:54,640 Speaker 3: going out it themselves. We actually do collaborate significantly and 142 00:06:54,720 --> 00:06:57,600 Speaker 3: that's going really well as well. RCG went over five 143 00:06:57,680 --> 00:07:01,560 Speaker 3: hundred sites in terms of the real connectivity areas, so 144 00:07:01,920 --> 00:07:04,200 Speaker 3: that's been really good. And complement that with a low 145 00:07:04,240 --> 00:07:08,159 Speaker 3: Earth orbit for the celt satellite. So I think you 146 00:07:08,279 --> 00:07:11,200 Speaker 3: have a really great story about more ubiquitous coverage in 147 00:07:11,240 --> 00:07:11,800 Speaker 3: New Zealand. 148 00:07:11,880 --> 00:07:15,680 Speaker 2: Yeah, you said you were thinking you'll possibly be completed 149 00:07:15,680 --> 00:07:18,160 Speaker 2: your five G roll out ahead maybe some of the others. 150 00:07:18,160 --> 00:07:20,600 Speaker 2: What does that actually look like. Would it be would 151 00:07:20,720 --> 00:07:24,880 Speaker 2: have equitable coverage to for G or is it separate 152 00:07:25,200 --> 00:07:27,160 Speaker 2: like more specified area rollout. 153 00:07:27,600 --> 00:07:31,240 Speaker 3: Yeah, it's a great question. I think everyone starts with 154 00:07:31,360 --> 00:07:33,840 Speaker 3: the higher population areas, right, that's where you start with 155 00:07:34,320 --> 00:07:37,440 Speaker 3: terms of just needing more capacity, but there's also being 156 00:07:37,520 --> 00:07:40,320 Speaker 3: a program of work to get five G in rural towns, 157 00:07:40,360 --> 00:07:43,560 Speaker 3: working again with the government, so that's been really successful. 158 00:07:43,560 --> 00:07:47,600 Speaker 3: Again that's part of investment strategy to get better coverage 159 00:07:47,600 --> 00:07:51,840 Speaker 3: into rural towns. But yeah, our approach has been basically 160 00:07:51,840 --> 00:07:55,080 Speaker 3: to modernize everything on that cell tart once. So not 161 00:07:55,160 --> 00:07:57,920 Speaker 3: deploying just five G but modernizing four G and five 162 00:07:58,000 --> 00:08:02,600 Speaker 3: G simultaneously means to actually get better four G experience 163 00:08:02,640 --> 00:08:06,480 Speaker 3: and a five G experience at the same time, which 164 00:08:06,520 --> 00:08:08,320 Speaker 3: I think is a really good strategy for us. It 165 00:08:08,360 --> 00:08:10,920 Speaker 3: does mean probably it's a little bit slower initially because 166 00:08:10,920 --> 00:08:12,640 Speaker 3: you're having to do a bit more work, but in 167 00:08:12,640 --> 00:08:15,520 Speaker 3: the long run it's a great strategy for us to 168 00:08:15,560 --> 00:08:16,200 Speaker 3: move forward on. 169 00:08:16,880 --> 00:08:18,640 Speaker 1: So we'll keep an eye on those coverage maps to 170 00:08:18,680 --> 00:08:19,960 Speaker 1: ComCom wants them within a year. 171 00:08:20,240 --> 00:08:24,040 Speaker 3: Yes, yeah, the team are busy already working through it timeframes. 172 00:08:24,120 --> 00:08:26,440 Speaker 1: Yeah, so we'll keep an eye on out. Another big 173 00:08:26,480 --> 00:08:29,280 Speaker 1: story making headlines this week, which is sort of relevance 174 00:08:29,320 --> 00:08:33,800 Speaker 1: to us, the Australian government has proposed mandatory guide rails 175 00:08:33,840 --> 00:08:39,520 Speaker 1: for use of artificial intelligence. They want require testing, transparency 176 00:08:39,600 --> 00:08:43,000 Speaker 1: about how AI is being used, and accountability. There's a 177 00:08:43,040 --> 00:08:47,040 Speaker 1: ten point checklist that they have floated. This is out 178 00:08:47,080 --> 00:08:50,160 Speaker 1: for public consultation for the next month or so. We 179 00:08:50,200 --> 00:08:53,360 Speaker 1: haven't moved to this step yet, and I think the 180 00:08:53,400 --> 00:08:56,920 Speaker 1: government will look quite closely at this. NBA has laid 181 00:08:56,920 --> 00:09:00,720 Speaker 1: out sort of the roadmap that it's following, and Collins 182 00:09:00,880 --> 00:09:04,880 Speaker 1: has said that she would like a risk based, proportional 183 00:09:04,920 --> 00:09:09,520 Speaker 1: and light touch regulatory regime around AI. Here they'll only 184 00:09:09,559 --> 00:09:12,080 Speaker 1: create AI legislation if there's a real need to, So 185 00:09:12,120 --> 00:09:14,720 Speaker 1: there's a lot of parity sort of going on with 186 00:09:14,880 --> 00:09:18,920 Speaker 1: the Australians. What's your take having looked at this already 187 00:09:19,000 --> 00:09:22,560 Speaker 1: using AI at two degrees and the practicalities of trying 188 00:09:22,559 --> 00:09:23,960 Speaker 1: to conform to something like this. 189 00:09:24,360 --> 00:09:26,240 Speaker 3: I think what I've seen was really doing is actually 190 00:09:26,320 --> 00:09:28,640 Speaker 3: really good because it is a risk based approach and 191 00:09:28,679 --> 00:09:33,640 Speaker 3: actually focused on the higher risk areas, but also holistically, 192 00:09:33,720 --> 00:09:36,320 Speaker 3: they're looking to invest in AI and actually accelerate it, 193 00:09:36,320 --> 00:09:39,840 Speaker 3: particularly in government, So I think that's a really great thing. Yeah, 194 00:09:39,840 --> 00:09:42,440 Speaker 3: for us, we've already been doing a lot in this space. 195 00:09:42,520 --> 00:09:45,960 Speaker 3: So when we compare our AI policy and responsible use 196 00:09:46,160 --> 00:09:49,480 Speaker 3: and all the great things around governance for AI. It 197 00:09:49,600 --> 00:09:51,440 Speaker 3: very much aligns to what we're already doing at two 198 00:09:51,480 --> 00:09:57,200 Speaker 3: degrees trust and transparency, having a human loop. Yeah, the 199 00:09:57,360 --> 00:10:00,719 Speaker 3: explainability aspect, which is hard for a black box. But 200 00:10:00,720 --> 00:10:04,360 Speaker 3: I think it's also important to recognize AI of the 201 00:10:04,440 --> 00:10:07,800 Speaker 3: past is very different to AI now with large language models. 202 00:10:07,800 --> 00:10:11,720 Speaker 3: So regulation previously was effectively based on like machine learning, 203 00:10:12,200 --> 00:10:16,680 Speaker 3: are you applying decisions to HR process or to a 204 00:10:16,679 --> 00:10:21,520 Speaker 3: criminal activity or churn predictions and how are you using it? 205 00:10:21,559 --> 00:10:24,280 Speaker 3: But large language model is actually very different to that, 206 00:10:24,600 --> 00:10:28,720 Speaker 3: although I wouldn't use them for actually making autonomous decisions yet, 207 00:10:28,720 --> 00:10:32,320 Speaker 3: but decision support is really great. So I think a 208 00:10:32,320 --> 00:10:34,600 Speaker 3: lot of the regulation is effectively the AI of the 209 00:10:34,640 --> 00:10:37,600 Speaker 3: past that I've actually seen trying to control that. But 210 00:10:37,640 --> 00:10:40,280 Speaker 3: the AI of the future is very different, And what 211 00:10:40,360 --> 00:10:43,600 Speaker 3: the Australian proposal is is actually taking more of a 212 00:10:43,600 --> 00:10:46,240 Speaker 3: holistic view of that, kind of breaking down what is 213 00:10:46,240 --> 00:10:49,199 Speaker 3: a developer of a model, you're training a pre trained model, 214 00:10:49,240 --> 00:10:50,960 Speaker 3: and what you need to do because that's a higher 215 00:10:51,040 --> 00:10:53,000 Speaker 3: risk area. You know, if you don't have the guardrails on, 216 00:10:53,080 --> 00:10:56,880 Speaker 3: you could create an unmitigated AI l em that could 217 00:10:56,880 --> 00:11:01,280 Speaker 3: do anything versus someone who's deploying LM into an organization, 218 00:11:01,320 --> 00:11:04,760 Speaker 3: which is effectively where we're at. And then also just 219 00:11:04,880 --> 00:11:07,240 Speaker 3: users of AI. You know, we're all using it every day. 220 00:11:07,280 --> 00:11:10,440 Speaker 3: I think nearly everyone is, So I think what they've 221 00:11:10,440 --> 00:11:13,120 Speaker 3: done is actually broken it down quite pragmatically into those 222 00:11:13,120 --> 00:11:16,719 Speaker 3: different areas and focusing on the higher risk components. So yeah, 223 00:11:16,720 --> 00:11:18,800 Speaker 3: I think it's actually really good. It aligns with what 224 00:11:18,920 --> 00:11:21,920 Speaker 3: we're doing. I don't like more audits, so I think 225 00:11:21,920 --> 00:11:24,240 Speaker 3: there was a mention in the proposal around actually auditing 226 00:11:25,559 --> 00:11:29,200 Speaker 3: entities around it. So having a more integrated approach where 227 00:11:29,360 --> 00:11:32,120 Speaker 3: it's actually part of existing audits and compliance, I think 228 00:11:32,200 --> 00:11:33,560 Speaker 3: is a better way to go. And I think they 229 00:11:33,600 --> 00:11:36,040 Speaker 3: actually proposed a few options have it more integrated as 230 00:11:36,080 --> 00:11:37,240 Speaker 3: well versus separate. 231 00:11:37,920 --> 00:11:40,680 Speaker 2: It seems like when you look at AI risk, it 232 00:11:40,720 --> 00:11:43,840 Speaker 2: doesn't immediately jump out that Talco would particularly be one 233 00:11:43,840 --> 00:11:46,200 Speaker 2: of the more risky areas we are using AI. But 234 00:11:46,280 --> 00:11:50,200 Speaker 2: then when you add that large language model factor into 235 00:11:50,240 --> 00:11:54,320 Speaker 2: it for things like customer service or whatever else you 236 00:11:54,400 --> 00:11:57,280 Speaker 2: might use for decision making support, that does kind of 237 00:11:57,760 --> 00:12:01,200 Speaker 2: enhance the risk a step up by default of using 238 00:12:01,480 --> 00:12:04,480 Speaker 2: those models. So how are you thinking about that? At 239 00:12:04,520 --> 00:12:09,240 Speaker 2: two degrees in terms of maintaining your level of I 240 00:12:09,240 --> 00:12:12,000 Speaker 2: guess soft compliance because we don't have that regulation, but 241 00:12:12,320 --> 00:12:15,400 Speaker 2: compliance with your own internal guidelines. 242 00:12:16,360 --> 00:12:20,080 Speaker 3: So we did deploy straight away quite early on, obviously 243 00:12:20,120 --> 00:12:22,360 Speaker 3: the explosion of chat gipt and everyone got really excited 244 00:12:22,360 --> 00:12:26,880 Speaker 3: about it, an AI responsible use policy with the intent 245 00:12:27,000 --> 00:12:29,920 Speaker 3: to actually embrace the technology, not to kind of stop 246 00:12:29,960 --> 00:12:33,360 Speaker 3: people using it, but put some guardrails around it so 247 00:12:33,400 --> 00:12:36,520 Speaker 3: people can use it safely. So that was kind of 248 00:12:36,520 --> 00:12:42,160 Speaker 3: immediately done and create more awareness around use of AI. 249 00:12:42,520 --> 00:12:44,760 Speaker 3: So a lot of our work has been just creating, 250 00:12:45,320 --> 00:12:50,240 Speaker 3: just doing lunch and learns and awareness training. So we 251 00:12:50,320 --> 00:12:52,160 Speaker 3: do want to invite people to use it for their 252 00:12:52,240 --> 00:12:53,800 Speaker 3: daily work. I think that's the best way to test 253 00:12:53,840 --> 00:12:56,520 Speaker 3: and learn get ready for what I think is actually 254 00:12:56,520 --> 00:12:59,680 Speaker 3: a huge you know, we're in a new revolution of technology. 255 00:13:00,240 --> 00:13:02,439 Speaker 3: We're just we're trying to get ready and taking many 256 00:13:02,559 --> 00:13:05,280 Speaker 3: different stakeholders on that journey, your board who are probably 257 00:13:05,400 --> 00:13:09,120 Speaker 3: very risk adverse, you know, the use of AI just 258 00:13:09,160 --> 00:13:11,960 Speaker 3: in simple things like if you're putting Microsoft Copilot on 259 00:13:12,000 --> 00:13:14,640 Speaker 3: the teams meeting and the people that you invite into 260 00:13:14,679 --> 00:13:18,520 Speaker 3: it will automatically get all the summarization notes and you 261 00:13:18,559 --> 00:13:21,520 Speaker 3: may not want them all to so just being aware of, 262 00:13:21,760 --> 00:13:23,480 Speaker 3: you know, the technology and what it could do in 263 00:13:23,559 --> 00:13:26,120 Speaker 3: terms of transparency of information, we may not want it, 264 00:13:26,760 --> 00:13:30,840 Speaker 3: and I think ethically we've just our main purpose is 265 00:13:30,880 --> 00:13:34,760 Speaker 3: to put the customer first. So when you think about 266 00:13:34,760 --> 00:13:37,199 Speaker 3: your customer experience as your north star, that's your guiding 267 00:13:37,280 --> 00:13:40,200 Speaker 3: light for anything. It doesn't necessarily have to be AI, right, 268 00:13:40,200 --> 00:13:43,280 Speaker 3: it could be any technology. When you put the customer first, 269 00:13:43,280 --> 00:13:45,840 Speaker 3: that's the most important thing for us. So that's kind 270 00:13:45,880 --> 00:13:50,000 Speaker 3: of our main principle under the use of AIS. You know, 271 00:13:50,120 --> 00:13:52,320 Speaker 3: is it going to actually make a better experience for 272 00:13:52,360 --> 00:13:53,000 Speaker 3: our customers. 273 00:13:54,280 --> 00:13:56,760 Speaker 1: You've got to foster that trust, ye You've it's got 274 00:13:56,760 --> 00:13:59,080 Speaker 1: to be a high trust system and part of that, 275 00:13:59,120 --> 00:14:01,400 Speaker 1: you know, this is an interesting point that they've floated 276 00:14:01,720 --> 00:14:05,920 Speaker 1: across the Tasman be transparent with other organizations across the 277 00:14:06,080 --> 00:14:09,240 Speaker 1: entire AI supply chain. So being able to explain to 278 00:14:09,280 --> 00:14:13,400 Speaker 1: your customers, Okay, we might be using something of AWS 279 00:14:13,559 --> 00:14:16,760 Speaker 1: or Azure, which is underpinned by open AI. We need 280 00:14:16,800 --> 00:14:18,160 Speaker 1: to be able to tell you a little bit and 281 00:14:18,240 --> 00:14:23,080 Speaker 1: have confidence that our suppliers are giving uugh enough information 282 00:14:23,600 --> 00:14:25,800 Speaker 1: to be confident that we're not going to breach privacy 283 00:14:26,400 --> 00:14:29,320 Speaker 1: or that. So I guess that'll be something that increasingly 284 00:14:29,320 --> 00:14:32,400 Speaker 1: you'll be looking at through the entire supply chain. What 285 00:14:32,440 --> 00:14:34,640 Speaker 1: AI are we using and do we know enough about it? 286 00:14:35,160 --> 00:14:38,560 Speaker 3: Yeah, definitely, And how using our data? We use a 287 00:14:38,560 --> 00:14:41,960 Speaker 3: lot of third party systems that probably hasn't changed in 288 00:14:42,040 --> 00:14:45,600 Speaker 3: risk profile. You've got to effectively trust. If you're using 289 00:14:45,600 --> 00:14:48,400 Speaker 3: a cloud provider like Microsoft or Amazon, you have to 290 00:14:48,440 --> 00:14:53,400 Speaker 3: trust what they're saying around their data use even something 291 00:14:53,480 --> 00:14:57,440 Speaker 3: like open ai GPT four, which may be a misconceptionist 292 00:14:57,480 --> 00:14:59,800 Speaker 3: because I think people might think of AI or as 293 00:14:59,840 --> 00:15:02,480 Speaker 3: a entity. So when you put data in there, it's 294 00:15:02,520 --> 00:15:05,960 Speaker 3: just learning everything, but it's still just a computer program 295 00:15:05,960 --> 00:15:09,160 Speaker 3: in some respects, or a data program. And there are 296 00:15:09,240 --> 00:15:12,120 Speaker 3: features in there where you can specifically say don't train 297 00:15:12,200 --> 00:15:14,880 Speaker 3: on my data or use it privacy features like you 298 00:15:14,880 --> 00:15:18,760 Speaker 3: would get with anything like Dropbox or anything. How using 299 00:15:18,960 --> 00:15:21,840 Speaker 3: your email Gmail, for example, you may not trust Google 300 00:15:21,880 --> 00:15:23,720 Speaker 3: to be training on your data either, but there are 301 00:15:23,760 --> 00:15:25,960 Speaker 3: ways to actually opt in or out. So I think 302 00:15:26,040 --> 00:15:27,920 Speaker 3: that's you do have an element of trust in the 303 00:15:27,920 --> 00:15:31,400 Speaker 3: supply chain definitely, which we've already had I think in 304 00:15:31,440 --> 00:15:36,520 Speaker 3: the past. So whether organizations trust these big tech companies 305 00:15:36,520 --> 00:15:39,680 Speaker 3: to do that is another question. You've got themes around 306 00:15:39,720 --> 00:15:43,720 Speaker 3: more sovereign AI and private AI as definitely building momentum 307 00:15:43,760 --> 00:15:46,800 Speaker 3: as well. You're seeing that globally, seeing what France is 308 00:15:46,800 --> 00:15:49,160 Speaker 3: doing that EU are doing, So I could see a 309 00:15:49,200 --> 00:15:52,560 Speaker 3: future actually where we do have more sovereign AI use, 310 00:15:53,360 --> 00:15:55,560 Speaker 3: which I think is actually not a bad thing, because 311 00:15:55,600 --> 00:15:57,640 Speaker 3: the future of AIS, I think will be actually a 312 00:15:57,760 --> 00:16:01,480 Speaker 3: zoo of them. There'll be LMS for various things. You've 313 00:16:01,480 --> 00:16:04,800 Speaker 3: got your obviously your big foundational frontier models, but you've 314 00:16:04,840 --> 00:16:07,000 Speaker 3: obviously got a huge amount of open source models. So 315 00:16:07,240 --> 00:16:09,080 Speaker 3: I think it's going to be more of an ecosystem 316 00:16:09,360 --> 00:16:12,880 Speaker 3: and more of a hybrid approach going forward, which I 317 00:16:12,880 --> 00:16:14,920 Speaker 3: think we have to be ready for and regulation needs 318 00:16:14,960 --> 00:16:18,360 Speaker 3: to make sure that is safe. 319 00:16:17,720 --> 00:16:22,120 Speaker 1: And small language models for very specific purposes, and of 320 00:16:22,160 --> 00:16:25,000 Speaker 1: course we've got the hyperscale is setting up here, so 321 00:16:25,280 --> 00:16:28,960 Speaker 1: the option of keeping all your data within New Zealand's 322 00:16:29,440 --> 00:16:32,000 Speaker 1: borders will be attractive as well from a privacy point 323 00:16:32,000 --> 00:16:32,320 Speaker 1: of view. 324 00:16:32,520 --> 00:16:36,320 Speaker 3: Absolutely, yeah. Are we using Amazon bedrock for a lot 325 00:16:36,360 --> 00:16:39,000 Speaker 3: of our kind of experimentation and what we're doing today 326 00:16:39,720 --> 00:16:44,600 Speaker 3: which has the LMS based in Australia, but you also 327 00:16:44,600 --> 00:16:47,320 Speaker 3: don't get the latest models that often either. That's also 328 00:16:47,360 --> 00:16:52,040 Speaker 3: a challenge with these entities. They do deploy them in 329 00:16:52,080 --> 00:16:54,440 Speaker 3: their main regions first and they come down to our region. 330 00:16:55,760 --> 00:16:57,280 Speaker 3: So we still want to be on the bleeding edge 331 00:16:57,280 --> 00:17:00,520 Speaker 3: as well and experimenting. So yeah, we're also are minting 332 00:17:00,640 --> 00:17:04,040 Speaker 3: with open source models, creating kind of those foundations to 333 00:17:04,080 --> 00:17:07,879 Speaker 3: be able to have optionality. Earlier in the year, we 334 00:17:07,920 --> 00:17:11,879 Speaker 3: released a chatbot, which was a generative AI chatbot to 335 00:17:12,000 --> 00:17:14,879 Speaker 3: our customers in a test and learn approach. We actually 336 00:17:14,880 --> 00:17:17,080 Speaker 3: deployed it into Slingshot, which is one of our brands 337 00:17:17,160 --> 00:17:20,560 Speaker 3: still rather than our bigger two Degree space, just to 338 00:17:20,600 --> 00:17:23,359 Speaker 3: really understand it's actually quite a dangerous thing to do. 339 00:17:23,440 --> 00:17:26,400 Speaker 3: In some respects it can hallucinate. We put a lot 340 00:17:26,400 --> 00:17:29,280 Speaker 3: of effort in testing guardrails and making sure it was 341 00:17:29,320 --> 00:17:32,160 Speaker 3: defined in its own context so it couldn't go out 342 00:17:32,160 --> 00:17:35,200 Speaker 3: of the context. My own team we're testing it. You know, 343 00:17:35,880 --> 00:17:38,480 Speaker 3: you know, what would Steve buy from two Degrees or 344 00:17:38,520 --> 00:17:40,840 Speaker 3: from Slingshot? Would you like a pair of shoes with 345 00:17:40,920 --> 00:17:44,439 Speaker 3: that or something? And we had a lot of learnings 346 00:17:44,440 --> 00:17:47,960 Speaker 3: from that, so we've built kind of an orchestration guardrail 347 00:17:48,960 --> 00:17:52,399 Speaker 3: ecosystem around it to ensure that the chatbot doesn't stray 348 00:17:52,440 --> 00:17:54,960 Speaker 3: off too far, which is a great learning for us, 349 00:17:55,160 --> 00:17:59,160 Speaker 3: and it also taught us many other things like our 350 00:17:59,200 --> 00:18:02,960 Speaker 3: knowledge bases need improving data is a huge aspect to 351 00:18:03,040 --> 00:18:07,200 Speaker 3: get value out of these things. So it's accelerated our 352 00:18:07,280 --> 00:18:10,480 Speaker 3: view on data modernization and knowledge management along the way. 353 00:18:10,520 --> 00:18:13,800 Speaker 3: So it's I think every organization should go on this 354 00:18:13,960 --> 00:18:17,200 Speaker 3: journey and actually test learn scale, just try it out, 355 00:18:17,320 --> 00:18:19,159 Speaker 3: take a few risks, but make sure you've got the 356 00:18:19,200 --> 00:18:22,240 Speaker 3: guardrails in place. That one is an example where it's 357 00:18:22,280 --> 00:18:25,239 Speaker 3: actually customer facing. I think a lot of organizations are 358 00:18:25,240 --> 00:18:27,920 Speaker 3: doing internal based chatbots as well, which is what we're doing, 359 00:18:28,320 --> 00:18:29,960 Speaker 3: which is more is a safe way to do it 360 00:18:29,960 --> 00:18:30,359 Speaker 3: as well. 361 00:18:30,560 --> 00:18:33,120 Speaker 2: Did you have any moments of like real shock when 362 00:18:33,119 --> 00:18:35,960 Speaker 2: a customer tried to do something or found a way 363 00:18:35,960 --> 00:18:37,879 Speaker 2: to break the bot in any kind of way that 364 00:18:37,960 --> 00:18:40,719 Speaker 2: you suddenly you had to deal with. 365 00:18:40,920 --> 00:18:43,879 Speaker 3: I think we got most of the guardrails pretty well 366 00:18:44,160 --> 00:18:48,560 Speaker 3: sorted what it did highlights because we do have the 367 00:18:48,640 --> 00:18:51,600 Speaker 3: human and the loop still and looking at the actual 368 00:18:51,640 --> 00:18:55,080 Speaker 3: responses and actually measuring how effective it is. It just 369 00:18:55,119 --> 00:18:58,040 Speaker 3: really highlighted our knowledge management wasn't as good as it 370 00:18:58,080 --> 00:19:01,400 Speaker 3: should be. So our knowledge base is our help desk articles, 371 00:19:01,440 --> 00:19:04,680 Speaker 3: which is what the model is using to refine its answers. 372 00:19:05,400 --> 00:19:08,320 Speaker 3: So when we had some results that we thought weren't 373 00:19:08,400 --> 00:19:11,879 Speaker 3: quite right, it's actually we're already saying that to our customers. 374 00:19:12,119 --> 00:19:16,200 Speaker 3: So it's actually helped our feedback loop. So not necessarily 375 00:19:16,280 --> 00:19:19,760 Speaker 3: anything too wrong. Yeah, I think there are some cases 376 00:19:19,800 --> 00:19:21,919 Speaker 3: where people tried to say, you know what I like 377 00:19:21,960 --> 00:19:23,919 Speaker 3: some would you like frise with that or you know 378 00:19:24,280 --> 00:19:26,760 Speaker 3: a few extra things, and the bot did actually go 379 00:19:26,880 --> 00:19:29,919 Speaker 3: off and try to help. It really wants to please people, 380 00:19:30,119 --> 00:19:33,760 Speaker 3: you know, that's the thing. So we had a few 381 00:19:33,760 --> 00:19:36,840 Speaker 3: instances where it did stray off a little bit to 382 00:19:36,920 --> 00:19:39,600 Speaker 3: try and please our customers because that was you know, 383 00:19:39,640 --> 00:19:43,120 Speaker 3: it was kind of told to do that. So yeah, 384 00:19:43,160 --> 00:19:45,359 Speaker 3: it's never perfect. I think. I think these things do 385 00:19:45,480 --> 00:19:48,600 Speaker 3: hallucinate a little bit, so that's why it's effect. You'd 386 00:19:48,600 --> 00:19:50,400 Speaker 3: still need the human and the loop, which is part 387 00:19:50,440 --> 00:19:55,399 Speaker 3: of the Australian proposal and human oversight. Do you need 388 00:19:55,440 --> 00:19:58,680 Speaker 3: to pull the trigger as well. We've got a within 389 00:19:58,720 --> 00:20:01,680 Speaker 3: our own policy like a something does terribly go wrong 390 00:20:01,720 --> 00:20:03,879 Speaker 3: with it we can just turn it off immediately. So 391 00:20:03,960 --> 00:20:06,840 Speaker 3: we've got, yeah, the big red button, the control loop. 392 00:20:07,560 --> 00:20:11,160 Speaker 1: Yeah, So that's what the Aussies will decide on. They've 393 00:20:11,160 --> 00:20:14,199 Speaker 1: got three options in terms of how they do this 394 00:20:14,280 --> 00:20:16,280 Speaker 1: from a framework point of view. They might have an 395 00:20:16,320 --> 00:20:18,919 Speaker 1: AI act like the EU, or they might just tweak 396 00:20:19,480 --> 00:20:24,320 Speaker 1: existing legislation. Does it sort of fill you with trepidation 397 00:20:24,440 --> 00:20:28,080 Speaker 1: the idea of dedicated legislation being introduced. Would you prefer 398 00:20:28,160 --> 00:20:30,840 Speaker 1: to see using existing legislation or is it much of 399 00:20:30,880 --> 00:20:31,639 Speaker 1: a muchness to you? 400 00:20:32,400 --> 00:20:34,160 Speaker 3: I don't think there's any right or wrong answer here 401 00:20:34,240 --> 00:20:38,639 Speaker 3: on this one. I generally would prefer it integrated into legislation, 402 00:20:38,680 --> 00:20:41,680 Speaker 3: which I think will need to happen anyway. You've got 403 00:20:41,680 --> 00:20:45,160 Speaker 3: the privacy Acts that we comply with today, so there's 404 00:20:45,240 --> 00:20:48,960 Speaker 3: maybe amendments to that. But it is such a transformational 405 00:20:49,000 --> 00:20:54,119 Speaker 3: technology that may need some different aspects to it. But 406 00:20:54,160 --> 00:20:57,639 Speaker 3: I'd like to see, you know, this kind of happening 407 00:20:57,640 --> 00:21:01,120 Speaker 3: in parallel to the kind of scientific development improvements, rather 408 00:21:01,160 --> 00:21:03,040 Speaker 3: than being kind of at the end of the process 409 00:21:03,080 --> 00:21:06,280 Speaker 3: and being preemptive and blocking technology actually going together on 410 00:21:06,280 --> 00:21:11,000 Speaker 3: the journey, which would be a completely different way of regulation. 411 00:21:11,200 --> 00:21:14,680 Speaker 3: I think technology is going to move so fast anyway, 412 00:21:14,840 --> 00:21:18,879 Speaker 3: and regulation is going to catch up. So yeah, I 413 00:21:18,920 --> 00:21:21,359 Speaker 3: do fundamentally believe that would probably need a different way 414 00:21:21,560 --> 00:21:25,280 Speaker 3: of regulating in the future given the speed of this. 415 00:21:27,280 --> 00:21:29,680 Speaker 3: But yeah, I think getting a few smart people together 416 00:21:29,720 --> 00:21:31,520 Speaker 3: to work it out would be really important in New 417 00:21:31,600 --> 00:21:34,840 Speaker 3: Zealand context. I think jud To Collins is already kind 418 00:21:34,880 --> 00:21:38,159 Speaker 3: of working on that, bringing the experts together. So I 419 00:21:38,200 --> 00:21:39,800 Speaker 3: don't think there's a right or wrong there, but I'd 420 00:21:39,880 --> 00:21:41,760 Speaker 3: prefer integrated. Yeah, And if we. 421 00:21:41,720 --> 00:21:44,720 Speaker 1: Can harmonize where possible with the Aussies, that would be 422 00:21:44,840 --> 00:21:47,119 Speaker 1: ideal as well. We have a lot of businesses operating 423 00:21:47,200 --> 00:21:49,960 Speaker 1: both markets, so it's good. Thanks for your views on 424 00:21:50,320 --> 00:21:53,959 Speaker 1: those two topical stories. And this really speaks to what 425 00:21:54,000 --> 00:21:56,119 Speaker 1: we wanted to talk to you about today, which is, 426 00:21:56,200 --> 00:21:58,760 Speaker 1: you know, there's a lot of mergers and acquisitions happening 427 00:21:58,800 --> 00:22:02,359 Speaker 1: and tough economic times. We've covered at Business Desk a 428 00:22:02,359 --> 00:22:05,760 Speaker 1: lot of takeover attempts of New Zealand companies. Private equity 429 00:22:05,760 --> 00:22:07,200 Speaker 1: out there is looking at New Zealand going Some of 430 00:22:07,240 --> 00:22:11,280 Speaker 1: these businesses are relatively cheap at the moment out of 431 00:22:11,320 --> 00:22:15,000 Speaker 1: economic necessity, Competitors are looking at each other going would 432 00:22:15,040 --> 00:22:18,640 Speaker 1: we be better off have more critical mass merging. You've 433 00:22:18,640 --> 00:22:21,520 Speaker 1: been through this process over the last couple of years 434 00:22:21,800 --> 00:22:27,160 Speaker 1: with Vocus merging with two degrees and really fascinated. First 435 00:22:27,160 --> 00:22:28,960 Speaker 1: of all, if you can tell us where is the 436 00:22:29,280 --> 00:22:32,119 Speaker 1: merger at at the moment thousands of people have been 437 00:22:32,160 --> 00:22:36,520 Speaker 1: brought together, but also all the systems underpinning these customer 438 00:22:36,600 --> 00:22:39,320 Speaker 1: bases coming together. Where are you at in that transformation. 439 00:22:40,680 --> 00:22:43,440 Speaker 3: Yeah, when you ask me that, I reflect back. It's 440 00:22:43,440 --> 00:22:47,119 Speaker 3: been a wild journey. So it's been two years into 441 00:22:47,200 --> 00:22:49,879 Speaker 3: effectively what is a three year strategy for us to integrate. 442 00:22:51,040 --> 00:22:54,640 Speaker 3: Merger happened first June twenty twenty two, and we've made 443 00:22:54,680 --> 00:22:58,680 Speaker 3: a huge amount of progress, and I think it's important 444 00:22:58,720 --> 00:23:02,960 Speaker 3: to kind of reflect. You know, both organizations have been 445 00:23:03,240 --> 00:23:06,240 Speaker 3: built up in different ways from different mergers and acquisitions. 446 00:23:06,280 --> 00:23:09,320 Speaker 3: Like you mentioned, if you look at the history of 447 00:23:09,400 --> 00:23:13,600 Speaker 3: the organizations, we're actually built up over sixty different service 448 00:23:13,640 --> 00:23:16,919 Speaker 3: providers and entities. So in a way, mergers and acquisitions 449 00:23:16,960 --> 00:23:19,520 Speaker 3: are in our DNA, but we're also there's a common 450 00:23:19,600 --> 00:23:22,360 Speaker 3: thread through all of it, which is that we're a challenger. 451 00:23:23,800 --> 00:23:26,520 Speaker 3: And Yeah, if you look back through all the different entities, 452 00:23:26,520 --> 00:23:30,160 Speaker 3: they've had their own stories around challenging the market, disrupting 453 00:23:30,160 --> 00:23:32,240 Speaker 3: the market. You look at two Degrees and the two 454 00:23:32,280 --> 00:23:36,320 Speaker 3: Degrees effect that occurred in two thousand and nine on 455 00:23:36,440 --> 00:23:40,320 Speaker 3: the Vocus side, there are many similar stories. So I 456 00:23:40,359 --> 00:23:42,800 Speaker 3: think when we came into the merger, we had a 457 00:23:42,800 --> 00:23:45,520 Speaker 3: lot of knowledge and how to do it, and we 458 00:23:45,560 --> 00:23:48,159 Speaker 3: had a blueprints and what we needed to achieve. But 459 00:23:48,200 --> 00:23:51,879 Speaker 3: every time it's different. Fundamentally, you can have a great vision, 460 00:23:52,520 --> 00:23:56,040 Speaker 3: but it's how you achieve it that's really important. So, yeah, 461 00:23:56,040 --> 00:23:59,760 Speaker 3: we're two years in and yeah, effectively nearly at the 462 00:23:59,840 --> 00:24:03,760 Speaker 3: end end of our three year integration strategy. It's going 463 00:24:03,800 --> 00:24:07,920 Speaker 3: really well. The main things we've got left kind of 464 00:24:07,920 --> 00:24:10,840 Speaker 3: at a high level from a system's perspective, we had 465 00:24:10,840 --> 00:24:17,240 Speaker 3: these two probably a large business support system which is 466 00:24:17,440 --> 00:24:20,800 Speaker 3: the two Degrees of Origin that's got most of our 467 00:24:20,800 --> 00:24:23,400 Speaker 3: mobile customers in it. We've got this other one called Tahi, 468 00:24:23,760 --> 00:24:27,240 Speaker 3: which is our target state. We rebranded it as part 469 00:24:27,280 --> 00:24:29,680 Speaker 3: of the merger as well, which is part of kind 470 00:24:29,680 --> 00:24:33,280 Speaker 3: of getting people on the journey. And we've been building 471 00:24:33,320 --> 00:24:36,680 Speaker 3: all the mobile journeys in that system, and we've done 472 00:24:36,680 --> 00:24:40,240 Speaker 3: a few trials in the last actually three weeks through 473 00:24:40,280 --> 00:24:42,680 Speaker 3: those mobile journeys and it's night and day. The difference. 474 00:24:43,320 --> 00:24:45,440 Speaker 3: It's not to say the other one was bad necessarily, 475 00:24:45,760 --> 00:24:49,200 Speaker 3: but we've invested so much in the customer experience aspect. 476 00:24:49,760 --> 00:24:52,040 Speaker 3: So while we're integrating trying to consolidate, we're trying to 477 00:24:52,080 --> 00:24:55,440 Speaker 3: improve customer experience. And to give an example of the difference, 478 00:24:56,320 --> 00:24:58,399 Speaker 3: kind of signing up a customer in the retail shop 479 00:24:58,480 --> 00:25:01,639 Speaker 3: that used to take probably around ten minutes on the 480 00:25:01,680 --> 00:25:04,439 Speaker 3: old system is now under a minute, so that's a 481 00:25:04,720 --> 00:25:08,879 Speaker 3: significant change. And the digital experience as well for our 482 00:25:08,920 --> 00:25:12,440 Speaker 3: customers to sign up online and also our digital experience 483 00:25:12,440 --> 00:25:15,240 Speaker 3: for our colleagues to actually interact with our customers is 484 00:25:15,320 --> 00:25:17,880 Speaker 3: night and day. So it's very exciting. So now we're 485 00:25:17,880 --> 00:25:21,480 Speaker 3: focusing on our migration aspect to actually migrate customers without 486 00:25:21,520 --> 00:25:26,920 Speaker 3: impacting the customer experience. So last year we've actually been 487 00:25:27,080 --> 00:25:30,320 Speaker 3: working on a range of techniques. This is more of 488 00:25:30,359 --> 00:25:34,480 Speaker 3: the geeky technology side of it in terms of transitional 489 00:25:34,640 --> 00:25:38,600 Speaker 3: architectures so that we can actually support customers through both 490 00:25:38,640 --> 00:25:42,639 Speaker 3: stacks at the same time, so building effectively transitional overlays, 491 00:25:42,680 --> 00:25:46,040 Speaker 3: so it's cross tax support. So from a customer experience, 492 00:25:46,240 --> 00:25:50,040 Speaker 3: if you've got your mobile app or your web experience. 493 00:25:50,480 --> 00:25:53,560 Speaker 3: It's seamless because it actually is supported across both systems. 494 00:25:53,560 --> 00:25:56,120 Speaker 3: So there's a range of little techniques that we use 495 00:25:56,200 --> 00:25:58,920 Speaker 3: to try and make it as seamless and effective as possible. 496 00:25:59,200 --> 00:26:03,000 Speaker 3: Our main focus, which is, don't treat it like a 497 00:26:03,000 --> 00:26:06,560 Speaker 3: big band kind of big project. Get incremental delivery time 498 00:26:06,600 --> 00:26:10,359 Speaker 3: to value is one of our key principles. So, you know, 499 00:26:10,440 --> 00:26:13,439 Speaker 3: the traditional digital transformation efforts that we go two years in, 500 00:26:14,000 --> 00:26:16,200 Speaker 3: I'm sitting here quite happily. I'd normally, you know, I'll 501 00:26:16,400 --> 00:26:18,800 Speaker 3: be gray by now as a CIO or CTO, and 502 00:26:18,840 --> 00:26:20,840 Speaker 3: normally the half life of a CIO is around that 503 00:26:20,840 --> 00:26:23,600 Speaker 3: two to three year mark. I'm sitting here reasonably comfortably 504 00:26:23,640 --> 00:26:26,760 Speaker 3: touch would. But it's because we've been so focused on 505 00:26:26,800 --> 00:26:30,919 Speaker 3: the incremental delivery. We've got that value and learning often 506 00:26:31,040 --> 00:26:34,480 Speaker 3: we don't want to do big bang approaches, and we've 507 00:26:34,480 --> 00:26:37,880 Speaker 3: seen that over the last two years we've actually had 508 00:26:37,960 --> 00:26:41,360 Speaker 3: value delivered. We are brought in the fixed broadband from 509 00:26:41,359 --> 00:26:44,280 Speaker 3: one system to in to TAHI quite early on, within 510 00:26:44,359 --> 00:26:47,200 Speaker 3: four months, and now the focus on the mobile journey. 511 00:26:47,960 --> 00:26:50,359 Speaker 3: But underpinning that, you know, you've got this business support 512 00:26:50,400 --> 00:26:54,959 Speaker 3: system you've also got operational support systems, you've got ERP systems, 513 00:26:55,160 --> 00:26:59,680 Speaker 3: you've got the billing charging. Yeah, all of that, TAHI 514 00:26:59,720 --> 00:27:01,919 Speaker 3: if it does a lot of that for us, But 515 00:27:02,040 --> 00:27:05,600 Speaker 3: there are hundreds of different systems. Got your data systems 516 00:27:05,640 --> 00:27:09,320 Speaker 3: started warehousing, range of programs. So it is quite a 517 00:27:09,320 --> 00:27:14,080 Speaker 3: diverse thing and it's very very hard to do and 518 00:27:14,119 --> 00:27:16,120 Speaker 3: you have to empower your teams to do it. So 519 00:27:16,480 --> 00:27:19,640 Speaker 3: I guess that's one of our key principles is actually 520 00:27:20,119 --> 00:27:22,760 Speaker 3: empowering our teams. Focus on the outcomes, focus on the why, 521 00:27:22,920 --> 00:27:27,199 Speaker 3: but empower the teams to deliver it. Yeah. 522 00:27:27,520 --> 00:27:31,080 Speaker 2: How do you start to make decisions about which parts 523 00:27:31,200 --> 00:27:33,920 Speaker 2: of your tech stack you've got? So you've got two 524 00:27:34,520 --> 00:27:39,280 Speaker 2: CRMs from different different entities that have come together. How 525 00:27:39,320 --> 00:27:40,960 Speaker 2: do you start to think about which one you're going 526 00:27:40,960 --> 00:27:43,080 Speaker 2: to wind down, which one you're going to build up. 527 00:27:43,280 --> 00:27:45,919 Speaker 2: What is the strategy that you think through there? Do 528 00:27:45,960 --> 00:27:48,640 Speaker 2: you go to your customers and ask them, do you 529 00:27:48,720 --> 00:27:51,760 Speaker 2: just take other metrics to see which are delivering better outcomes? 530 00:27:51,960 --> 00:27:55,040 Speaker 3: What does that look like? Yeah, a really good question, 531 00:27:56,200 --> 00:27:58,440 Speaker 3: and again, yeah, you've got to face it on its 532 00:27:58,520 --> 00:28:03,679 Speaker 3: merits at the time actually making these decisions and when 533 00:28:03,720 --> 00:28:06,639 Speaker 3: you look back at the different entities, each one has 534 00:28:06,680 --> 00:28:10,199 Speaker 3: their own legacy as well. So fundamentally, our approach has 535 00:28:10,200 --> 00:28:13,880 Speaker 3: always been probably to choose the best based on certain metrics, 536 00:28:13,920 --> 00:28:16,399 Speaker 3: but not build something new on top of that. So 537 00:28:16,520 --> 00:28:20,000 Speaker 3: not a green fields approach, actually an evolutionary approach. So 538 00:28:20,000 --> 00:28:24,120 Speaker 3: you're looking at your ecosystem of systems on its merits. 539 00:28:24,200 --> 00:28:27,000 Speaker 3: You do obviously have to factor in commercial constraints as well, 540 00:28:27,640 --> 00:28:30,600 Speaker 3: but basically capability and the people around it that can 541 00:28:30,640 --> 00:28:33,480 Speaker 3: support it. It may not be the system that has 542 00:28:33,520 --> 00:28:38,680 Speaker 3: the most amount of customers or usage or adoption, it's 543 00:28:38,760 --> 00:28:40,960 Speaker 3: really the system that is going to be future proof 544 00:28:41,000 --> 00:28:43,720 Speaker 3: for us going forward. So when I mentioned TAHI, it's 545 00:28:43,720 --> 00:28:47,640 Speaker 3: actually in house built system, which is quite key for 546 00:28:47,760 --> 00:28:51,480 Speaker 3: us because it's basically our intellectual property, our unique proposition. 547 00:28:52,400 --> 00:28:54,720 Speaker 3: I've got a team around four hundred and fifty staff 548 00:28:54,880 --> 00:29:01,240 Speaker 3: in the technology team that crosses digital, cybersecurity, data, network engineering, 549 00:29:01,520 --> 00:29:03,840 Speaker 3: the radio access engineers is quite a big team, but 550 00:29:03,880 --> 00:29:07,400 Speaker 3: we have over one hundred software developers. I consider ourselves 551 00:29:07,440 --> 00:29:11,600 Speaker 3: actually a software company with great Taco assets. So one 552 00:29:11,640 --> 00:29:14,520 Speaker 3: of the decisions that rarely is you know, for TAHI, 553 00:29:14,560 --> 00:29:16,800 Speaker 3: it was a decision around we have people around it, 554 00:29:16,840 --> 00:29:19,400 Speaker 3: we support it, We can innovate every day. You know, 555 00:29:19,440 --> 00:29:24,800 Speaker 3: we deploy features tens of times a day, launching darkly 556 00:29:25,000 --> 00:29:28,600 Speaker 3: into our system. So we have an ability to iterate 557 00:29:28,720 --> 00:29:30,960 Speaker 3: and have really fast feedback on that system. So that 558 00:29:31,000 --> 00:29:34,040 Speaker 3: was one of the key decision points, was actually time 559 00:29:34,080 --> 00:29:37,360 Speaker 3: to value, time to learning versus other systems we might 560 00:29:37,440 --> 00:29:41,840 Speaker 3: be dependent on third parties or vendors, or even the 561 00:29:41,880 --> 00:29:45,640 Speaker 3: system itself might just have constraints. So one of our 562 00:29:45,760 --> 00:29:48,720 Speaker 3: main factors was again back to that time to value. 563 00:29:49,360 --> 00:29:51,600 Speaker 3: May not be the best system in terms of capability, 564 00:29:51,600 --> 00:29:54,800 Speaker 3: but we could back ourselves to build certain capabilities on 565 00:29:54,840 --> 00:29:58,920 Speaker 3: that system due to our ability to deliver quickly. So 566 00:29:58,960 --> 00:30:01,400 Speaker 3: I think that's actually unique for us. Is nothing I 567 00:30:01,400 --> 00:30:03,520 Speaker 3: don't think in the New Zealand markets or even when 568 00:30:03,560 --> 00:30:08,280 Speaker 3: I look globally in terms of that software approach, and 569 00:30:08,360 --> 00:30:10,080 Speaker 3: that means it turns up to customers in terms of 570 00:30:10,120 --> 00:30:11,480 Speaker 3: customer experience. 571 00:30:12,120 --> 00:30:14,400 Speaker 1: That is yeah, I think quite quite unique because we've 572 00:30:14,440 --> 00:30:17,000 Speaker 1: heard so much. A lot of the big software vendors 573 00:30:17,000 --> 00:30:20,240 Speaker 1: have tailored a version of their software for the Talco market. 574 00:30:20,280 --> 00:30:22,800 Speaker 1: They see that as an important market to serve, whether 575 00:30:22,840 --> 00:30:27,200 Speaker 1: it's SAP or salesforce, for instance, you've gone to bespoke 576 00:30:27,560 --> 00:30:30,560 Speaker 1: route is really part of that your legacy as a 577 00:30:30,640 --> 00:30:33,520 Speaker 1: challenger brand. You're one of the originals way back in 578 00:30:33,640 --> 00:30:36,160 Speaker 1: the in the early days of Slingshot and all that, 579 00:30:36,240 --> 00:30:39,800 Speaker 1: with Malcolm Dick and Mike Callender. You know you're still there. 580 00:30:40,280 --> 00:30:42,000 Speaker 1: Is that The mindset really is that if we want 581 00:30:42,040 --> 00:30:45,400 Speaker 1: to be competitive, more competitive than the incumbents, we have 582 00:30:45,520 --> 00:30:48,200 Speaker 1: to be bespoke. We can't just buy something off the 583 00:30:48,240 --> 00:30:50,800 Speaker 1: shelf and have a great experience. We've got to build ourselves. 584 00:30:51,800 --> 00:30:54,320 Speaker 3: Yeah. I thinks, yeah, there is a mindset to do 585 00:30:54,360 --> 00:30:57,440 Speaker 3: things differently. You know. One of our values is say 586 00:30:57,480 --> 00:30:59,640 Speaker 3: no to the status quo. So we like to innovate, 587 00:30:59,680 --> 00:31:04,320 Speaker 3: do things differently Bespoke Yeah, I think I still think 588 00:31:04,320 --> 00:31:06,680 Speaker 3: of it like a product. So it's actually what we're 589 00:31:06,680 --> 00:31:09,360 Speaker 3: building is a product in its own right. So it's 590 00:31:09,360 --> 00:31:13,240 Speaker 3: not a kind of customized you know bespoke aspect. It's 591 00:31:13,240 --> 00:31:17,320 Speaker 3: actually something that's strategically important for us. And we don't 592 00:31:17,840 --> 00:31:21,440 Speaker 3: build everything. We use a lot of open source, We 593 00:31:21,520 --> 00:31:24,720 Speaker 3: do use a lot of other software to complement an ecosystem, 594 00:31:24,760 --> 00:31:27,120 Speaker 3: but the core of it we do actually build and own, 595 00:31:27,960 --> 00:31:31,280 Speaker 3: which I do think is unique. To give you an 596 00:31:31,320 --> 00:31:34,440 Speaker 3: example of you know, we don't necessarily you know, build 597 00:31:34,520 --> 00:31:37,840 Speaker 3: versus by aspect. Right, Building's not always great because it 598 00:31:37,880 --> 00:31:39,080 Speaker 3: can take a lot of time. You don't want to 599 00:31:39,080 --> 00:31:43,040 Speaker 3: reinvent the wheel. So when we were looking at we 600 00:31:43,120 --> 00:31:45,680 Speaker 3: built this actually a few years ago to do network 601 00:31:45,680 --> 00:31:48,640 Speaker 3: as a service. So when a customer, particularly business customers, 602 00:31:48,640 --> 00:31:51,000 Speaker 3: we have a portal called Flex, they can actually log 603 00:31:51,040 --> 00:31:55,479 Speaker 3: in get networks on demand, so you can provision a 604 00:31:55,520 --> 00:31:58,480 Speaker 3: network service from a data center to the cloud as 605 00:31:58,520 --> 00:32:02,600 Speaker 3: your AWS in real time. Typically the approach would be 606 00:32:02,680 --> 00:32:05,160 Speaker 3: to go to a vendor for your network orchestration layer. 607 00:32:05,600 --> 00:32:07,960 Speaker 3: We actually looked across what other big tech companies were 608 00:32:07,960 --> 00:32:10,160 Speaker 3: doing and we saw what Uber we're doing in terms 609 00:32:10,160 --> 00:32:14,040 Speaker 3: of their workflow orchestration to book a taxi we go, 610 00:32:14,160 --> 00:32:16,760 Speaker 3: why don't we look at using that technology for the 611 00:32:16,800 --> 00:32:20,080 Speaker 3: network and we did and that was actually a huge 612 00:32:20,080 --> 00:32:22,640 Speaker 3: game changer for us. So I'm just looking wider than 613 00:32:22,720 --> 00:32:25,640 Speaker 3: kind of a traditional talco. And then we've used that 614 00:32:25,720 --> 00:32:31,320 Speaker 3: same workflow engine that we Uber open sourced it to 615 00:32:31,440 --> 00:32:34,480 Speaker 3: do mobile provisioning as well. So since the merger, we've 616 00:32:34,480 --> 00:32:37,640 Speaker 3: actually been using the same workflow engine to do mobile 617 00:32:38,080 --> 00:32:42,760 Speaker 3: workflow orchestration. So that's one example where you're leveraging technology 618 00:32:42,800 --> 00:32:46,520 Speaker 3: not necessarily building all of the building blocks, which is 619 00:32:46,520 --> 00:32:49,440 Speaker 3: pretty exciting. The team are excited about it, and they're 620 00:32:49,480 --> 00:32:52,000 Speaker 3: all just across the road actually from where we're recording 621 00:32:52,040 --> 00:32:54,920 Speaker 3: this podcast. Most of them they're all in New Zealand 622 00:32:55,280 --> 00:32:58,720 Speaker 3: software developers and they're really bought into the vision of 623 00:32:58,800 --> 00:33:00,840 Speaker 3: two degrees. So I think that's a really important part 624 00:33:00,960 --> 00:33:03,320 Speaker 3: is they're there to fight for fear. They know how 625 00:33:03,360 --> 00:33:05,920 Speaker 3: important it is for two degrees to exist in this market, 626 00:33:06,240 --> 00:33:09,080 Speaker 3: and they come to work every day to actually provide 627 00:33:09,160 --> 00:33:12,800 Speaker 3: value to our customers, and that quite often can be 628 00:33:14,280 --> 00:33:16,840 Speaker 3: you know, as I've been a previously a software developer, 629 00:33:16,880 --> 00:33:19,600 Speaker 3: a network engineer, you can be stuck in the on 630 00:33:19,640 --> 00:33:22,600 Speaker 3: your keyboard, not really understanding why you're doing it. So 631 00:33:22,720 --> 00:33:25,040 Speaker 3: a big part of what we're doing is the cultural aspect. 632 00:33:25,040 --> 00:33:28,840 Speaker 3: It's actually eighty percent cultural. When you kind of break 633 00:33:28,880 --> 00:33:29,440 Speaker 3: it down. 634 00:33:29,600 --> 00:33:34,959 Speaker 2: When you're doing such a big transformation integration project and 635 00:33:35,000 --> 00:33:39,240 Speaker 2: you really hyper focused on trying to go from one 636 00:33:39,280 --> 00:33:43,160 Speaker 2: thing to another, how do you balance just focusing the 637 00:33:43,240 --> 00:33:49,600 Speaker 2: resource on kind of the transformation aspect versus bringing in 638 00:33:49,640 --> 00:33:55,200 Speaker 2: new features upgrading, adding on to create new experiences. How 639 00:33:55,200 --> 00:33:59,600 Speaker 2: do you decide where to go big and new and 640 00:33:59,640 --> 00:34:01,640 Speaker 2: where to just stick to your bread and butter. 641 00:34:02,040 --> 00:34:04,280 Speaker 3: It's a great question, and it's a yeah. We do 642 00:34:04,360 --> 00:34:09,840 Speaker 3: have this discussion a lot. Yeah, So we laser focused 643 00:34:09,840 --> 00:34:13,080 Speaker 3: on some key deliverables and that integration and getting onto 644 00:34:13,080 --> 00:34:15,319 Speaker 3: a single stack is such a game changer for us. 645 00:34:15,360 --> 00:34:17,879 Speaker 3: So we're trying to be really really focused, but we 646 00:34:18,040 --> 00:34:22,440 Speaker 3: see opportunity, we may take it. I guess one example 647 00:34:22,480 --> 00:34:25,279 Speaker 3: we've had of that when we were building some of 648 00:34:25,320 --> 00:34:28,920 Speaker 3: the new mobile capability in TAHI, we saw an opportunity 649 00:34:28,920 --> 00:34:31,680 Speaker 3: to launch a new product out as kind of a 650 00:34:31,719 --> 00:34:36,920 Speaker 3: relatively simple product, a traveler some package. So when overseas 651 00:34:36,920 --> 00:34:38,759 Speaker 3: people come in and they want to roam around New 652 00:34:38,840 --> 00:34:41,160 Speaker 3: Zealand and have the great experience, they could easily just 653 00:34:41,200 --> 00:34:44,719 Speaker 3: get on boarded and have a traveler prepay like a 654 00:34:44,719 --> 00:34:47,960 Speaker 3: thirty day unlimited use, and we built that on the 655 00:34:48,000 --> 00:34:51,600 Speaker 3: new system. In some respects, it was to both test 656 00:34:51,680 --> 00:34:54,120 Speaker 3: the system, but also there was a gap in the 657 00:34:54,160 --> 00:34:56,799 Speaker 3: market we saw and we can actually provide a really 658 00:34:56,840 --> 00:35:01,200 Speaker 3: great digital experience you can sign up with eSIM. So 659 00:35:01,239 --> 00:35:03,080 Speaker 3: there was an example of where it was a win win. 660 00:35:03,280 --> 00:35:07,640 Speaker 3: You could actually test your integration thesis while also building 661 00:35:07,719 --> 00:35:11,200 Speaker 3: something something unique in the market. But we don't do 662 00:35:11,280 --> 00:35:13,680 Speaker 3: that too often because we're trying to stay trying to 663 00:35:13,680 --> 00:35:17,160 Speaker 3: stay focused on what we're trying to achieve. But yeah, 664 00:35:17,200 --> 00:35:19,520 Speaker 3: it's a daily discussion. But I think that's the important 665 00:35:19,520 --> 00:35:22,080 Speaker 3: part is you get the right people together to have 666 00:35:22,160 --> 00:35:27,719 Speaker 3: that intentional discussion around these decisions. I think that's the 667 00:35:27,760 --> 00:35:30,520 Speaker 3: most important part. So where we may deviate, it's for 668 00:35:30,560 --> 00:35:33,480 Speaker 3: a good reason, we're following some sort of value, or 669 00:35:34,080 --> 00:35:37,040 Speaker 3: it's giving us ability to test and learn for where 670 00:35:37,080 --> 00:35:41,120 Speaker 3: our future is. But yeah, there's many many aspects of that. 671 00:35:41,160 --> 00:35:43,520 Speaker 3: You can't really break it down to a recipe or playbook. 672 00:35:43,719 --> 00:35:46,520 Speaker 3: It could just be an opportunity that is in front 673 00:35:46,560 --> 00:35:49,440 Speaker 3: of you. And I think, yes, again, it's really important 674 00:35:49,480 --> 00:35:52,439 Speaker 3: you have a plan, but you have to be able 675 00:35:52,520 --> 00:35:56,160 Speaker 3: to adapt. And quite often over the last two years, 676 00:35:56,200 --> 00:35:59,600 Speaker 3: it's been the actual constraints and the impediments or the 677 00:35:59,680 --> 00:36:03,960 Speaker 3: oppertunities that actually have created our journey for us. So 678 00:36:04,040 --> 00:36:06,400 Speaker 3: not being rigid on a plan, which I think is 679 00:36:06,440 --> 00:36:09,680 Speaker 3: also learning that I've had in my history, and I 680 00:36:09,719 --> 00:36:13,520 Speaker 3: think many other CEOs and many organizations where you do 681 00:36:13,600 --> 00:36:15,959 Speaker 3: follow a plan rigidly, that's where things can go wrong. 682 00:36:16,080 --> 00:36:18,440 Speaker 3: You have to definitely adapt. You know, AI wasn't really 683 00:36:18,480 --> 00:36:21,239 Speaker 3: a thing when we started off on the merger either, 684 00:36:22,719 --> 00:36:25,719 Speaker 3: so that you know that all you know became blew 685 00:36:25,800 --> 00:36:28,920 Speaker 3: up in October November twenty twenty two. Actually, can we 686 00:36:29,000 --> 00:36:33,080 Speaker 3: leverage this technology to accelerate our integration strategy And we've 687 00:36:33,120 --> 00:36:36,319 Speaker 3: been looking at aspects in that area too. One of 688 00:36:36,360 --> 00:36:38,640 Speaker 3: those has really been around software development, which I think 689 00:36:38,760 --> 00:36:43,080 Speaker 3: is a massive game changer for anyone using AI to 690 00:36:43,160 --> 00:36:46,880 Speaker 3: actually improve productivity and software development, which we did invest 691 00:36:46,920 --> 00:36:49,960 Speaker 3: a lot of time in. That's you know one example, 692 00:36:50,000 --> 00:36:51,680 Speaker 3: things change and you've got to adapt to it. 693 00:36:52,000 --> 00:36:56,040 Speaker 1: Yeah, that's a massive one. Andy Jesse, the CEO of Amazon, 694 00:36:56,120 --> 00:36:59,239 Speaker 1: said a couple of weeks ago that using their in 695 00:36:59,360 --> 00:37:03,480 Speaker 1: house sort of copilot for software development, I think it's 696 00:37:03,480 --> 00:37:05,680 Speaker 1: called Q, they were able to save four and a 697 00:37:05,760 --> 00:37:11,120 Speaker 1: half thousand developer years just on upgrading Java apps. At 698 00:37:11,800 --> 00:37:14,319 Speaker 1: Amazon they have tens of thousands of Java apps, so 699 00:37:14,400 --> 00:37:16,759 Speaker 1: this is they said, it's going to save about a 700 00:37:16,840 --> 00:37:19,319 Speaker 1: quarter of a billion dollars a year, not just in 701 00:37:19,360 --> 00:37:23,000 Speaker 1: getting rid of developers, all the infrastructure the apps run 702 00:37:23,040 --> 00:37:26,640 Speaker 1: more efficiently on their big server farms and ad as well, 703 00:37:26,680 --> 00:37:30,040 Speaker 1: so the energy use is lower. So are you seeing 704 00:37:30,080 --> 00:37:32,000 Speaker 1: that you say you're a sort of a software firm 705 00:37:32,000 --> 00:37:34,840 Speaker 1: in many respects, are you seeing big efficiencies and software development? 706 00:37:34,920 --> 00:37:38,799 Speaker 3: Yet we are seeing some maybe not claim so many 707 00:37:38,840 --> 00:37:44,680 Speaker 3: hours in productivity. Yeah, I think early on, yeah, we 708 00:37:44,719 --> 00:37:47,720 Speaker 3: saw around twenty percent as we measured kind of by hours, 709 00:37:48,600 --> 00:37:51,600 Speaker 3: and it's probably got to about ten percent now in 710 00:37:51,680 --> 00:37:56,719 Speaker 3: terms of productivity. But it's hard to measure software development productivity. 711 00:37:56,760 --> 00:37:59,080 Speaker 3: It's not an easy thing to do, and there are 712 00:37:59,080 --> 00:38:01,960 Speaker 3: other constraints. So this is something I think about a lot. 713 00:38:02,000 --> 00:38:07,000 Speaker 3: I've been on the journey for too long. And yeah, 714 00:38:07,040 --> 00:38:08,839 Speaker 3: you've got to think about how you can figure your 715 00:38:08,840 --> 00:38:11,719 Speaker 3: teams and how you can figure your overarching system. An 716 00:38:11,800 --> 00:38:16,560 Speaker 3: organization is actually a complex adaptive system and it can 717 00:38:16,600 --> 00:38:18,640 Speaker 3: be immune and you can poke it and it can 718 00:38:18,680 --> 00:38:21,440 Speaker 3: do things. But you've got to look at the end 719 00:38:21,560 --> 00:38:25,399 Speaker 3: end value. So software development can improve by maybe even 720 00:38:25,400 --> 00:38:28,200 Speaker 3: fifty percent, But if you've got constraints to go to market, 721 00:38:28,320 --> 00:38:30,719 Speaker 3: or you've got constraints at the front end and your 722 00:38:30,719 --> 00:38:34,480 Speaker 3: business casing, your finance modeling, how you actually do business 723 00:38:34,480 --> 00:38:36,800 Speaker 3: cases you won't get the true end to end value. 724 00:38:36,920 --> 00:38:39,560 Speaker 3: So we do think a lot about that idea to 725 00:38:39,680 --> 00:38:43,439 Speaker 3: value and measure that flow. That's a really key part 726 00:38:43,440 --> 00:38:45,960 Speaker 3: of what we look at. So while you can optimize 727 00:38:46,000 --> 00:38:48,239 Speaker 3: and that wedge in the middle for software development, you've 728 00:38:48,239 --> 00:38:51,080 Speaker 3: got to optimize the full end to end, which I 729 00:38:51,080 --> 00:38:54,960 Speaker 3: actually think is the whole thing around generative AI. There 730 00:38:55,040 --> 00:38:59,040 Speaker 3: is a huge software is completely revolutionized. I believe from 731 00:38:59,040 --> 00:39:02,720 Speaker 3: what happened kind of in the nineteen seventies, software actually 732 00:39:02,719 --> 00:39:05,720 Speaker 3: hasn't changed too much and in many respects you're still 733 00:39:05,800 --> 00:39:08,840 Speaker 3: kind of was very traditional, but with the advent of 734 00:39:08,920 --> 00:39:11,880 Speaker 3: large language models, it's turning more into a natural language. 735 00:39:12,200 --> 00:39:15,359 Speaker 3: We have co pilot assistance next to our software developers. 736 00:39:15,480 --> 00:39:18,480 Speaker 3: I really think the multi modal aspect is going to 737 00:39:18,480 --> 00:39:21,239 Speaker 3: be huge. So you can actually have a voice assistant 738 00:39:21,360 --> 00:39:24,839 Speaker 3: while you're coding. That's going to be significant. So yeah, 739 00:39:24,880 --> 00:39:27,319 Speaker 3: it's going to be a huge game changer, and there's 740 00:39:27,360 --> 00:39:30,200 Speaker 3: a whole lot of I did not necessarily threaten my team, 741 00:39:30,239 --> 00:39:33,719 Speaker 3: but I provocated that, you know, software development profession will 742 00:39:33,760 --> 00:39:36,319 Speaker 3: be dead in five years, you know, and think about 743 00:39:36,320 --> 00:39:39,239 Speaker 3: the evolution of it. I don't think that's actually a case, 744 00:39:39,280 --> 00:39:42,280 Speaker 3: but it will definitely adapt and change. So the teams 745 00:39:42,320 --> 00:39:44,759 Speaker 3: are you know, they're just getting ready, getting the skills 746 00:39:45,120 --> 00:39:48,880 Speaker 3: to adapt, but it is the next step. Can you 747 00:39:48,920 --> 00:39:51,120 Speaker 3: have these we call them value streams. Can you have 748 00:39:51,200 --> 00:39:55,200 Speaker 3: value stream agents that can connect the requirements to the 749 00:39:55,239 --> 00:39:56,920 Speaker 3: teams that are doing the work. Can you then get 750 00:39:56,960 --> 00:39:59,279 Speaker 3: strategy agents over the top. So if you can get 751 00:39:59,680 --> 00:40:02,759 Speaker 3: a X improvement and software development, can you get the 752 00:40:02,800 --> 00:40:05,440 Speaker 3: team working better at the value stream? That's another ten X. 753 00:40:05,480 --> 00:40:09,000 Speaker 3: Can you get the strategy connected to execution better. That's 754 00:40:09,040 --> 00:40:12,560 Speaker 3: you know, could be a thousand X improvement, which I 755 00:40:12,600 --> 00:40:14,960 Speaker 3: think would be massive. And then you've got the other 756 00:40:15,040 --> 00:40:17,880 Speaker 3: case where AI is now I play with it all 757 00:40:17,880 --> 00:40:21,800 Speaker 3: the time. You can build apps and software so quickly 758 00:40:21,800 --> 00:40:25,319 Speaker 3: as an individual contributor, So you'll see the advent of these, 759 00:40:25,360 --> 00:40:28,880 Speaker 3: you know, one person companies, which I think it's going 760 00:40:28,920 --> 00:40:31,279 Speaker 3: to be great for startups in the future as well. Yeah, 761 00:40:32,960 --> 00:40:37,120 Speaker 3: but our team's assisting them day to day to write 762 00:40:37,120 --> 00:40:42,799 Speaker 3: better software, also for testing software, doing our reviews of 763 00:40:42,880 --> 00:40:46,360 Speaker 3: software as a complementary like a copilot. And that's something 764 00:40:46,400 --> 00:40:49,000 Speaker 3: we've actually talked about a lot, is around the co 765 00:40:49,239 --> 00:40:53,680 Speaker 3: pilot nature of AI. Actually being an assistant next to you, saying, 766 00:40:53,719 --> 00:40:56,120 Speaker 3: with our care team actually having a co pilot next 767 00:40:56,160 --> 00:40:58,800 Speaker 3: to our agents that are answering the phone to support 768 00:40:58,800 --> 00:41:00,879 Speaker 3: our customers, they've actually got a co pilot that can 769 00:41:00,880 --> 00:41:03,399 Speaker 3: coach them and work with them. That's our next step 770 00:41:03,400 --> 00:41:07,319 Speaker 3: in our journey, particularly in the care environment. So that's 771 00:41:07,360 --> 00:41:09,520 Speaker 3: I think that's the approach I think we'll move forward on. 772 00:41:09,840 --> 00:41:13,799 Speaker 2: So it would be in quotes, listening to the conversation 773 00:41:14,320 --> 00:41:17,040 Speaker 2: as you're having it and say, oh, I noticed that 774 00:41:17,080 --> 00:41:19,799 Speaker 2: you've talked about this. You know, don't forget to tell 775 00:41:19,840 --> 00:41:22,640 Speaker 2: the customer that this is an option that kind of thing. 776 00:41:23,080 --> 00:41:24,279 Speaker 3: Yes, definitely, Yeah, that's it. 777 00:41:24,440 --> 00:41:24,759 Speaker 1: Yeah, ye. 778 00:41:25,719 --> 00:41:30,000 Speaker 3: And we've already got a AI auto summary working with 779 00:41:30,080 --> 00:41:33,960 Speaker 3: our care team, so is actually it's got the real 780 00:41:33,960 --> 00:41:40,000 Speaker 3: time transcription happening. It actually summarizes the entire call for 781 00:41:40,120 --> 00:41:42,560 Speaker 3: the agent. It used to be the human and the 782 00:41:42,600 --> 00:41:44,399 Speaker 3: agent used to type it all out, so they get 783 00:41:44,400 --> 00:41:46,160 Speaker 3: a little bit distracted while they're trying to talk to 784 00:41:46,200 --> 00:41:49,440 Speaker 3: a customer while they're writing notes. So now our humans 785 00:41:50,000 --> 00:41:52,760 Speaker 3: can actually have a real conversation, and then the AI 786 00:41:52,880 --> 00:41:56,600 Speaker 3: is actually writing up the notes in real time, which 787 00:41:56,640 --> 00:41:59,720 Speaker 3: gives us really good data. It actually captures sentiment and intent. 788 00:42:00,040 --> 00:42:02,120 Speaker 3: You can then put that into our feedback loop and 789 00:42:02,120 --> 00:42:05,120 Speaker 3: actually improve the situation. We haven't quite got to the 790 00:42:05,120 --> 00:42:08,040 Speaker 3: point where we actually have a really true co pilot, 791 00:42:08,080 --> 00:42:09,960 Speaker 3: but that's our next step is to have a co 792 00:42:10,080 --> 00:42:14,239 Speaker 3: pilot alongside our human agent to give that sort of 793 00:42:14,280 --> 00:42:17,800 Speaker 3: coaching aspect and go okay, you should maybe think about 794 00:42:17,840 --> 00:42:22,080 Speaker 3: talking about this. It actually a near term for us, 795 00:42:22,080 --> 00:42:24,920 Speaker 3: it's not far away, which is pretty exciting. 796 00:42:25,480 --> 00:42:26,200 Speaker 1: It's pretty cool. 797 00:42:26,280 --> 00:42:27,080 Speaker 3: Yeah. 798 00:42:27,120 --> 00:42:31,120 Speaker 1: So just finally, Stephen, what's your advice to executives in 799 00:42:31,160 --> 00:42:33,560 Speaker 1: New Zealand companies that may have just acquired or been 800 00:42:33,600 --> 00:42:36,239 Speaker 1: a process of merging with another company that might be 801 00:42:36,280 --> 00:42:40,040 Speaker 1: a retailer or a construction company. You've talked about the 802 00:42:40,239 --> 00:42:44,160 Speaker 1: sort of the gradual approach versus big bang. It's nice 803 00:42:44,200 --> 00:42:47,239 Speaker 1: to be on one system from an efficiency point of view, 804 00:42:47,280 --> 00:42:50,439 Speaker 1: so there will be this temptation to get everything onto 805 00:42:50,440 --> 00:42:53,319 Speaker 1: one platform. We've seen where that has gone wrong. Some 806 00:42:53,360 --> 00:42:55,600 Speaker 1: projects have gone off the rails have been paused because 807 00:42:55,600 --> 00:42:58,080 Speaker 1: that's too complex. We've seen Jason Parris, you know, your 808 00:42:58,080 --> 00:43:01,200 Speaker 1: competitor from One Ends, had been very upfront and honest 809 00:43:01,239 --> 00:43:04,319 Speaker 1: about the pain that that organization went through with all 810 00:43:04,360 --> 00:43:07,920 Speaker 1: these old systems hanging around. Even if you can bridge 811 00:43:08,160 --> 00:43:11,840 Speaker 1: through APIs and that you can get data shared across 812 00:43:11,880 --> 00:43:15,920 Speaker 1: those systems, how do you approach what sort of thinking 813 00:43:15,960 --> 00:43:18,480 Speaker 1: do you go through? Everything? Is every situation as different 814 00:43:18,520 --> 00:43:20,000 Speaker 1: as you sort of said, But what are some of 815 00:43:20,040 --> 00:43:22,480 Speaker 1: the fundamental questions you need to ask yourself as a 816 00:43:22,520 --> 00:43:26,080 Speaker 1: leadership team when you're in this sort of merged position. 817 00:43:26,160 --> 00:43:30,560 Speaker 3: Now, yeah, yeah, you've got to have belief in a 818 00:43:30,600 --> 00:43:35,239 Speaker 3: commitment to do it. So it's not easy. So integration 819 00:43:35,640 --> 00:43:38,680 Speaker 3: in particular and legacy has to be a priority. It 820 00:43:38,760 --> 00:43:41,640 Speaker 3: has to be the forefront of your strategy. So one 821 00:43:41,640 --> 00:43:45,800 Speaker 3: of our strategies is that one company strategy. So integration 822 00:43:45,960 --> 00:43:49,640 Speaker 3: is everything that's fundamental. I think sometimes leaders can get 823 00:43:49,640 --> 00:43:53,120 Speaker 3: a little bit maybe just focus on the next thing 824 00:43:53,400 --> 00:43:56,799 Speaker 3: they think it's done, But there's a long term commitment 825 00:43:56,920 --> 00:43:59,320 Speaker 3: and you have to reinforce it, and I think leaders 826 00:43:59,320 --> 00:44:02,120 Speaker 3: go first in that respect. Build the belief, build the 827 00:44:02,200 --> 00:44:07,200 Speaker 3: vision of the future. Don't inflict change on your own 828 00:44:07,239 --> 00:44:10,040 Speaker 3: teams because part of it is also you're not just 829 00:44:10,160 --> 00:44:13,480 Speaker 3: changing technology, you're changing how people work. There's so many 830 00:44:13,560 --> 00:44:18,040 Speaker 3: different aspects to you making change happen. It's like it 831 00:44:18,120 --> 00:44:21,560 Speaker 3: is eighty percent of human endeavor, So you really need 832 00:44:21,640 --> 00:44:24,960 Speaker 3: that cultural aspect. I think that's the critical part is 833 00:44:25,080 --> 00:44:28,080 Speaker 3: having a team that are willing to have each other's 834 00:44:28,120 --> 00:44:30,920 Speaker 3: back and actually, you know, it's okay to make mistakes. 835 00:44:30,960 --> 00:44:34,719 Speaker 3: Actually having a psychologically safe environment where it's okay sometimes 836 00:44:35,000 --> 00:44:38,920 Speaker 3: maybe not do the right thing, and amplify those signals quickly. 837 00:44:39,840 --> 00:44:41,840 Speaker 3: I think that's a key part of it. But you 838 00:44:41,880 --> 00:44:43,920 Speaker 3: have to move fast. You've got there's a pace to it, 839 00:44:44,960 --> 00:44:48,160 Speaker 3: and that can be you know, people are stronger on 840 00:44:48,239 --> 00:44:51,479 Speaker 3: loss of version right, two times more impactful losing something 841 00:44:51,520 --> 00:44:54,040 Speaker 3: than gaining something, So you do have to build that belief. 842 00:44:54,280 --> 00:44:56,760 Speaker 3: You have to create a really safe environment and empower 843 00:44:56,840 --> 00:44:58,840 Speaker 3: people to go on that journey. So I think you 844 00:44:59,000 --> 00:45:03,799 Speaker 3: focus on the outcomes and why incremental is critical. It's 845 00:45:03,840 --> 00:45:07,160 Speaker 3: really easy to follow a plan and might be a 846 00:45:07,160 --> 00:45:11,320 Speaker 3: big modernization plan, but it may feel painful at the beginning, 847 00:45:11,360 --> 00:45:15,359 Speaker 3: actually questioning things, really do we really need to do this? 848 00:45:15,640 --> 00:45:18,800 Speaker 3: You know, is that requirement really necessary. It's quite easy 849 00:45:18,840 --> 00:45:20,920 Speaker 3: for people to add in all sorts of things as 850 00:45:20,920 --> 00:45:22,440 Speaker 3: you go on the journey, so you need to be 851 00:45:22,520 --> 00:45:26,000 Speaker 3: deleting things removing things. The best way to achieve it 852 00:45:26,000 --> 00:45:29,279 Speaker 3: actually not to build anything. So you have to be 853 00:45:29,440 --> 00:45:33,600 Speaker 3: really really focused on it and really ask those tough questions, 854 00:45:33,800 --> 00:45:38,400 Speaker 3: and yeah, focus on the value, focus on the customer. 855 00:45:38,480 --> 00:45:41,000 Speaker 3: That's ultimately why we're doing what we're doing. Yeah, why 856 00:45:41,080 --> 00:45:44,080 Speaker 3: we're doing what we're doing is get onto a single stack, 857 00:45:44,120 --> 00:45:47,640 Speaker 3: be the customer experience for the future for our customers. 858 00:45:47,640 --> 00:45:51,200 Speaker 3: So yeah, easy as that, so simple. 859 00:45:51,320 --> 00:45:55,799 Speaker 2: Right, Well, that's it for this week's episode. Thank you 860 00:45:55,800 --> 00:45:58,640 Speaker 2: to Stephen Kajerv for joining us and to two Degrees 861 00:45:58,680 --> 00:46:01,120 Speaker 2: for its ongoings. Abort the Business of Tech. 862 00:46:01,280 --> 00:46:04,480 Speaker 1: Subscribe to get all of the episodes of The Business 863 00:46:04,480 --> 00:46:08,200 Speaker 1: Off Tech in your favorite podcast app or from iHeartRadio, 864 00:46:08,440 --> 00:46:11,120 Speaker 1: where you can stream every episode. Show notes for the 865 00:46:11,120 --> 00:46:13,360 Speaker 1: Business of Tech are in the Tech section on the 866 00:46:13,360 --> 00:46:14,440 Speaker 1: Business Desk website. 867 00:46:14,480 --> 00:46:17,200 Speaker 2: Get in touch with your feedback, ideas, topics, and guest 868 00:46:17,280 --> 00:46:21,000 Speaker 2: suggestions email me on banat Businessdesk dot co, dot z 869 00:46:21,239 --> 00:46:24,319 Speaker 2: or find both of us on LinkedIn and sometimes on. 870 00:46:24,520 --> 00:46:26,960 Speaker 1: X and another dose of The Business Off Tech coming 871 00:46:26,960 --> 00:46:28,280 Speaker 1: your way next Thursday. 872 00:46:28,360 --> 00:46:28,920 Speaker 3: Catch you then,