Jon: back to Transform. joining us for part two of a three part conversation ⁓ a recent trip that Tom and I took to ⁓ to talk about data, information and technology ⁓ 27 of ⁓ current clients ⁓ all asset ⁓ well. ⁓ In first part, talked about some of the common challenges, what we saw, what's common
Tash Wilson: You've been listening to Transform by GHD. Big ideas, real impact. Welcome to Transform, tomorrow's thinking today. Real conversations on the innovation shaping businesses and our society.
Jon: Australia and UK, what's common between clients in Australia, time we're going to a little more the data versus applications platforms piece. Data goes into our applications, some of which in infrastructure are quite old. ⁓ because they're safety critical and because they're part of the safety case just because ⁓ put them in a long time ago and that was really painful and we don't want to do it again. Data applications. Because those applications are difficult and expensive to change, lot of the time we also then put a platform on top of the applications to aggregate and integrate that data and turn it into insights, or at least that's ⁓ the We can't get rid of the legacy systems, but by putting the platform on top, hopefully we can make things better. But sometimes that just adds complication. what did we see in terms of where were at in that process and what ⁓ some of common themes there?
Tom: Yeah, yeah, yeah, yeah. Well, it kind of feels like almost quite an exciting time in the industry in general, that the kind of solutions and it's just like more software. And I think people are coming to the realization of that. So it's going to less, but better. And then also a move away. So like make better use of ⁓ the systems software that you have, ⁓ and potentially
Jon: Indeed.
Tom: your kind key transformation isn't the new software, which can cost ⁓ a lot of money, actually modernizing for where that software is. So kind of like less customization, ⁓ because you procure that in the first place, and in some cases that's kind of 10 years ago, 12 years ago, whenever, ⁓ it highly customized to do the things that you needed to do to operate safely your infrastructure. ⁓ And over that time, all of these systems now have built in that capability as part of their standard offering. if you can have less customization, then that process of updating the system over time ⁓ less and less painful. So you're there with the latest version of it ⁓ you have minimal customization, which think is really exciting if organizations can. ⁓ get to that stage, we've definitely seen in the UK organisations putting off that question about updating system and effectively they do it until it's not supported anymore and then ⁓ there's big flurry because there's a big risk associated with it.
Jon: Yeah, the changes that AI is going to bring in terms of the costs of new and particular the functionality that can come out of the box in the platform piece and the aggregation piece is ⁓ going to be real game changer. It does, however, referencing previous conversation, bring the focus back to... what are you putting into that box? like AI can and kind of the new technologies and kind of as system development and application development costs kind of dramatically reduce and accelerate in speed, the focus comes back to we will ⁓ be able to ourselves of that burden at point over the next kind of five years. ⁓
Tom: Yeah, yeah.
Jon: three years, one year, maybe next week, who knows it's AI, right? But the focus then comes back to are we putting into that whole chain? What is the data that we are feeding that? What does it represent in reality? And is it good quality? ⁓ that's not a piece that AI is going to be able to sell for us because that's the bit where ⁓ actually create a digital representation of our real world. And ⁓ the bit where the data goes in. I think that's ⁓ really ⁓
Tom: Yeah, yeah, maybe next ⁓ Yeah. Yeah. Yeah.
Jon: area of discussion. think some clients very advanced platforms and were ready for that ⁓ had some foundations that would make them resilient to that. I think there were some clients that had very advanced platforms but hadn't solved the underlying, are we putting good stuff in, ⁓ piece, which was a challenge.
Tom: Yeah, yeah, yeah, yeah, which is exactly what we see here in the UK. So yeah, it's probably just the shape of where the industry is.
Jon: Absolutely. Yeah. Yeah, And, know, the ⁓ space, we want to clients new stuff, tools. Azure, AWS, Google, Snowflake, pick your name, right? we can buy new fancy stuff to go in the platform. It's harder to sort out the data ⁓ and go through that process of really in detail what does it mean and what is it telling us?
Tom: Hmm. Yeah, And then really, if you've got a programme, you're kind of making that better and better over time. So it's not kind of a big bang cost you're planning for it.
Jon: ⁓ think ⁓ ability to take things a piece at a time and understand how you solve an issue, have stay relatively well solved and be able to step forward from that is something that we've ⁓ spent lot of time on with our clients here in the UK over the last kind two or three years. ⁓ And was interesting to see there were clients that were at that same stage of We can do a use case, but somehow it just dies and we go back to square one. How do we solve an issue and then another issue and then another issue and be able to get that kind of really solid incremental progress? think that comes back ⁓ the basics of what data are putting in? What does it really mean? What is it really telling us? ⁓
Tom: Yeah, yeah Yeah.
Jon: Are we processing and developing that data ⁓ through a domain cognizant lens, i.e. having people that really understand what it means look at it and process it out?
Tom: Yeah, yeah, which could be transformative for the industry over the next couple of years, that you're effectively that, you know, you're no longer having quite clunky systems that you're not wanting to touch too much. And then the data sits behind that, whereas actually the data is elevated. And then the clients are owners and stewards of the data of which the systems facilitate. And then the power is in the the client's hands.
Jon: Yeah, yeah, absolutely. I mean, ⁓ as say, ⁓ AI, right? So make predictions that are peril, ⁓ but I think that it's clear that what data you really have? does it really mean? What are you really trying to do with it? And which specific business problems is it required? to solve that's where the conversation is going to move to the technology, the applications, the individual pieces of software the platforms that they run through that stuff is going to, ⁓ you know, the there is just phenomenal ⁓ enormous. So ⁓ how are we planning for? Yeah, a fresh new world where there isn't, ⁓ yeah, we've got data quality and that's an issue, but we also need the right applications and they need to be set up in the right way. And then we need to do loads of work in the platform piece as well. And actually we've got sort of a whole string of problems to, we just have data. We just have data problems. We just need to put in really good stuff. And what does that look like in terms of business capability, in terms of technology capability?
Tom: you Yeah, yeah, it was really good being in the pub last night with lots of the data team. of them was saying even March compared to December, the ability to use AI to facilitate the cogeneration that they're doing is just transformative. And that's just in the space of three or four months. so that's really good because then we can support clients to ⁓ kind shape things the way they need,
Jon: Yeah, I think of just thinking back to the trip at the end of November, start of December, we clearly didn't talk about AI enough. ⁓ We did about AI, but we didn't talk about it enough. ⁓ that ⁓ and again, you know, as we talked about in the first part, pace of progress ⁓ that that presents a huge, challenge to the infrastructure sector ⁓ to a critical
Tom: No, Yeah. Yeah.
Jon: engineering sector anyway, because it just simply isn't safe, frankly, for organisations to move that fast. It's just like, that's just never going to be okay. So, know, there is probably a new discipline of thinking of like, what does change management look like in this space? Like, how do you resiliently try and
Tom: Yeah. Yeah.
Jon: a lasso around a technology piece that's moving way faster than your organization ever will be able to. I don't know what the solutions are in that space. I'm not sure that's a mature offering. I think it's a ⁓ fascinating area to think about. ⁓ ⁓ going to need to be some experimentation to see ⁓ works and what doesn't.
Tom: Hmm. This sounds like we need another trip, ⁓
Jon: I think we should go back. I think we should go to Adelaide. The people from Adelaide were a bit upset we didn't go there. I think we should go back.
Tom: we can add Adelaide onto the list and talk to clients about AI.
Jon: conclusion, that acceleration, will the geographies closer together in terms of the that they're facing because it ⁓ is just going to make a lot of existing problems disappear. you can deploy it correctly, legacy systems, the specifics of how do we take data and process it out? How do we make data, how do we make our insights accessible? All ⁓ of mechanic stuff of going from raw data to the insight is going to, if you get it right, massively accelerate. that whole space to a degree becomes automated. The question is, ⁓ what does data really mean? ⁓ What instruction you giving to the ⁓ AI and agents and the automation that takes it from raw data to ⁓ insight? How you assuring that that's correct? So I think challenge with the challenges will become ⁓ more common, because will really be a question of how do we get
Tom: Yeah.
Jon: our specific automation to reflect our specific circumstances and deal with our specific data and tell us what we specifically need to know.
Tom: exciting times, transformational times.
Jon: Yeah, it is. an incredibly bright future here, which is ⁓ we don't have to ⁓ terrible problems we need to pay like tens ⁓ and sometimes hundreds of millions of dollars or even pounds to basically refresh a and have the same functionality we had ⁓ like ⁓ a ago much fuller wallet.
Tom: Yeah.
Jon: we can focus all of that effort and all of that money on how do we improve our data quality? How do we make our data the best possible representation of our physical reality? And how do we give the right insights to the right people to manage and operate the infrastructure effectively? there's a huge upside here, right? Like there will be huge disruption and it's gonna be chaotic in places, there's a real utopia that's... that feels like it's just becoming like, can sense ⁓ coming down the track. And I think that should very exciting for everyone and particular, the taxpayers. ⁓ you know, gonna be able to make some real gains here.
Tom: Yeah. Yeah. yeah for not massive massive amounts of money. There is one part of me that's slightly dying inside because I did spend the five years traveling every day from far east London to far west London basically doing system customization which maybe wasn't needed but it's the pace of change isn't it?
Jon: Yeah, people will look back in history and there's pre-AI and there's post-AI and there's the technology revolution pre-AI and a lot will look like going round in circles perhaps and then there's post-AI. I'm sure we will find a new way to go round in circles in some places. But certainly some of the challenges that we have today are going to
Tom: Yeah.
Jon: become trivial and that's a very exciting proposition.
Tom: Yeah, that is. What are we going to talk about next?
Jon: In part three, we're going to talk about what happens next ⁓ are next steps and what we're to do next.
Tash Wilson: You've been listening to Transform by GHD. Big ideas, real impact. Welcome to Transform, tomorrow's thinking today. Real conversations on the innovation shaping businesses and our society.
Jon: Australia and UK, what's common between clients in Australia, time we're going to a little more the data versus applications platforms piece. Data goes into our applications, some of which in infrastructure are quite old. ⁓ because they're safety critical and because they're part of the safety case just because ⁓ put them in a long time ago and that was really painful and we don't want to do it again. Data applications. Because those applications are difficult and expensive to change, lot of the time we also then put a platform on top of the applications to aggregate and integrate that data and turn it into insights, or at least that's ⁓ the We can't get rid of the legacy systems, but by putting the platform on top, hopefully we can make things better. But sometimes that just adds complication. what did we see in terms of where were at in that process and what ⁓ some of common themes there?
Tom: Yeah, yeah, yeah, yeah. Well, it kind of feels like almost quite an exciting time in the industry in general, that the kind of solutions and it's just like more software. And I think people are coming to the realization of that. So it's going to less, but better. And then also a move away. So like make better use of ⁓ the systems software that you have, ⁓ and potentially
Jon: Indeed.
Tom: your kind key transformation isn't the new software, which can cost ⁓ a lot of money, actually modernizing for where that software is. So kind of like less customization, ⁓ because you procure that in the first place, and in some cases that's kind of 10 years ago, 12 years ago, whenever, ⁓ it highly customized to do the things that you needed to do to operate safely your infrastructure. ⁓ And over that time, all of these systems now have built in that capability as part of their standard offering. if you can have less customization, then that process of updating the system over time ⁓ less and less painful. So you're there with the latest version of it ⁓ you have minimal customization, which think is really exciting if organizations can. ⁓ get to that stage, we've definitely seen in the UK organisations putting off that question about updating system and effectively they do it until it's not supported anymore and then ⁓ there's big flurry because there's a big risk associated with it.
Jon: Yeah, the changes that AI is going to bring in terms of the costs of new and particular the functionality that can come out of the box in the platform piece and the aggregation piece is ⁓ going to be real game changer. It does, however, referencing previous conversation, bring the focus back to... what are you putting into that box? like AI can and kind of the new technologies and kind of as system development and application development costs kind of dramatically reduce and accelerate in speed, the focus comes back to we will ⁓ be able to ourselves of that burden at point over the next kind of five years. ⁓
Tom: Yeah, yeah.
Jon: three years, one year, maybe next week, who knows it's AI, right? But the focus then comes back to are we putting into that whole chain? What is the data that we are feeding that? What does it represent in reality? And is it good quality? ⁓ that's not a piece that AI is going to be able to sell for us because that's the bit where ⁓ actually create a digital representation of our real world. And ⁓ the bit where the data goes in. I think that's ⁓ really ⁓
Tom: Yeah, yeah, maybe next ⁓ Yeah. Yeah. Yeah.
Jon: area of discussion. think some clients very advanced platforms and were ready for that ⁓ had some foundations that would make them resilient to that. I think there were some clients that had very advanced platforms but hadn't solved the underlying, are we putting good stuff in, ⁓ piece, which was a challenge.
Tom: Yeah, yeah, yeah, yeah, which is exactly what we see here in the UK. So yeah, it's probably just the shape of where the industry is.
Jon: Absolutely. Yeah. Yeah, And, know, the ⁓ space, we want to clients new stuff, tools. Azure, AWS, Google, Snowflake, pick your name, right? we can buy new fancy stuff to go in the platform. It's harder to sort out the data ⁓ and go through that process of really in detail what does it mean and what is it telling us?
Tom: Hmm. Yeah, And then really, if you've got a programme, you're kind of making that better and better over time. So it's not kind of a big bang cost you're planning for it.
Jon: ⁓ think ⁓ ability to take things a piece at a time and understand how you solve an issue, have stay relatively well solved and be able to step forward from that is something that we've ⁓ spent lot of time on with our clients here in the UK over the last kind two or three years. ⁓ And was interesting to see there were clients that were at that same stage of We can do a use case, but somehow it just dies and we go back to square one. How do we solve an issue and then another issue and then another issue and be able to get that kind of really solid incremental progress? think that comes back ⁓ the basics of what data are putting in? What does it really mean? What is it really telling us? ⁓
Tom: Yeah, yeah Yeah.
Jon: Are we processing and developing that data ⁓ through a domain cognizant lens, i.e. having people that really understand what it means look at it and process it out?
Tom: Yeah, yeah, which could be transformative for the industry over the next couple of years, that you're effectively that, you know, you're no longer having quite clunky systems that you're not wanting to touch too much. And then the data sits behind that, whereas actually the data is elevated. And then the clients are owners and stewards of the data of which the systems facilitate. And then the power is in the the client's hands.
Jon: Yeah, yeah, absolutely. I mean, ⁓ as say, ⁓ AI, right? So make predictions that are peril, ⁓ but I think that it's clear that what data you really have? does it really mean? What are you really trying to do with it? And which specific business problems is it required? to solve that's where the conversation is going to move to the technology, the applications, the individual pieces of software the platforms that they run through that stuff is going to, ⁓ you know, the there is just phenomenal ⁓ enormous. So ⁓ how are we planning for? Yeah, a fresh new world where there isn't, ⁓ yeah, we've got data quality and that's an issue, but we also need the right applications and they need to be set up in the right way. And then we need to do loads of work in the platform piece as well. And actually we've got sort of a whole string of problems to, we just have data. We just have data problems. We just need to put in really good stuff. And what does that look like in terms of business capability, in terms of technology capability?
Tom: you Yeah, yeah, it was really good being in the pub last night with lots of the data team. of them was saying even March compared to December, the ability to use AI to facilitate the cogeneration that they're doing is just transformative. And that's just in the space of three or four months. so that's really good because then we can support clients to ⁓ kind shape things the way they need,
Jon: Yeah, I think of just thinking back to the trip at the end of November, start of December, we clearly didn't talk about AI enough. ⁓ We did about AI, but we didn't talk about it enough. ⁓ that ⁓ and again, you know, as we talked about in the first part, pace of progress ⁓ that that presents a huge, challenge to the infrastructure sector ⁓ to a critical
Tom: No, Yeah. Yeah.
Jon: engineering sector anyway, because it just simply isn't safe, frankly, for organisations to move that fast. It's just like, that's just never going to be okay. So, know, there is probably a new discipline of thinking of like, what does change management look like in this space? Like, how do you resiliently try and
Tom: Yeah. Yeah.
Jon: a lasso around a technology piece that's moving way faster than your organization ever will be able to. I don't know what the solutions are in that space. I'm not sure that's a mature offering. I think it's a ⁓ fascinating area to think about. ⁓ ⁓ going to need to be some experimentation to see ⁓ works and what doesn't.
Tom: Hmm. This sounds like we need another trip, ⁓
Jon: I think we should go back. I think we should go to Adelaide. The people from Adelaide were a bit upset we didn't go there. I think we should go back.
Tom: we can add Adelaide onto the list and talk to clients about AI.
Jon: conclusion, that acceleration, will the geographies closer together in terms of the that they're facing because it ⁓ is just going to make a lot of existing problems disappear. you can deploy it correctly, legacy systems, the specifics of how do we take data and process it out? How do we make data, how do we make our insights accessible? All ⁓ of mechanic stuff of going from raw data to the insight is going to, if you get it right, massively accelerate. that whole space to a degree becomes automated. The question is, ⁓ what does data really mean? ⁓ What instruction you giving to the ⁓ AI and agents and the automation that takes it from raw data to ⁓ insight? How you assuring that that's correct? So I think challenge with the challenges will become ⁓ more common, because will really be a question of how do we get
Tom: Yeah.
Jon: our specific automation to reflect our specific circumstances and deal with our specific data and tell us what we specifically need to know.
Tom: exciting times, transformational times.
Jon: Yeah, it is. an incredibly bright future here, which is ⁓ we don't have to ⁓ terrible problems we need to pay like tens ⁓ and sometimes hundreds of millions of dollars or even pounds to basically refresh a and have the same functionality we had ⁓ like ⁓ a ago much fuller wallet.
Tom: Yeah.
Jon: we can focus all of that effort and all of that money on how do we improve our data quality? How do we make our data the best possible representation of our physical reality? And how do we give the right insights to the right people to manage and operate the infrastructure effectively? there's a huge upside here, right? Like there will be huge disruption and it's gonna be chaotic in places, there's a real utopia that's... that feels like it's just becoming like, can sense ⁓ coming down the track. And I think that should very exciting for everyone and particular, the taxpayers. ⁓ you know, gonna be able to make some real gains here.
Tom: Yeah. Yeah. yeah for not massive massive amounts of money. There is one part of me that's slightly dying inside because I did spend the five years traveling every day from far east London to far west London basically doing system customization which maybe wasn't needed but it's the pace of change isn't it?
Jon: Yeah, people will look back in history and there's pre-AI and there's post-AI and there's the technology revolution pre-AI and a lot will look like going round in circles perhaps and then there's post-AI. I'm sure we will find a new way to go round in circles in some places. But certainly some of the challenges that we have today are going to
Tom: Yeah.
Jon: become trivial and that's a very exciting proposition.
Tom: Yeah, that is. What are we going to talk about next?
Jon: In part three, we're going to talk about what happens next ⁓ are next steps and what we're to do next.
