Nov. 4, 2024

Hazem Abolrous: The three pillars for AI mobilisation

Hazem Abolrous: The three pillars for AI mobilisation
Hazem Abolrous: The three pillars for AI mobilisation
AI Pathfinder for Private Equity Podcast
Hazem Abolrous: The three pillars for AI mobilisation

Summary In this episode of the AI Pathfinder for Private Equity podcast, host Steve Budd discusses the critical role of AI in private equity with Hazem Aboris, Managing Partner at RingStone. They explore the importance of AI mobilisation, the evolution of AI technology, and the necessary steps for successful implementation. The conversation highlights the significance of data quality, the need for a top-down approach, and the challenges organisations face in adopting AI. They also discuss the...

Summary
In this episode of the AI Pathfinder for Private Equity podcast, host Steve Budd discusses the critical role of AI in private equity with Hazem Aboris, Managing Partner at RingStone. They explore the importance of AI mobilisation, the evolution of AI technology, and the necessary steps for successful implementation. The conversation highlights the significance of data quality, the need for a top-down approach, and the challenges organisations face in adopting AI. They also discuss the metrics for measuring its impact. The episode concludes with insights on the future of AI in organizations and the importance of a continuous learning mindset.

Takeaways

  • AI is crucial for business transformation in private equity
  • A significant percentage of AI projects fail due to lack of readiness
  • Data quality and strategy are foundational for AI success
  • AI mobilisation requires a top-down approach from the C-suite
  • An experimental mindset is necessary to explore AI possibilities
  • Metrics like R&D spend and revenue per employee will change with AI
  • Common challenges include messy data and nascent use cases
  • Organisations must adapt their ways of working to leverage AI effectively

AI Pathfinder Private Equity AI Strategy Meet-ups:

https://www.aipathfinder.co/private-equity-ai-events

Have a question or would like to suggest improvements, please contact steve@aipathfinder.co

SPEAKER_00

Hello and welcome to the AI Pathfinder for Private Equity podcast. I'm on a mission to lead private equity from AI bewilderment to AI empowerment. And I'm doing that by building a dedicated network that will piece by piece create a playbook for successful AI adoption across private equity and their portfolio. In building this network, I'm meeting and having conversations with some very interesting people. They may work in or with PE or provide a solution to their portfolio. What unites them is their curiosity of this emerging technology. And I'm really pleased to bring some of these conversations to life through this podcast. Today we're going to talk about the building blocks for successful AI mobilization, the theme for my next private equity AI strategy breakfast briefing in London on the 3rd of December. Now a recent report by Rand found that 84% of business leaders believe AI will have a significant impact on their business, but the majority aren't ready, leading to as many as 80% of AI projects failing, twice the failure rate of other IT projects. So how might private equity reverse this trend? Well, we might get close to answering uh that question today, and I'm very excited to have joining me today is Hazem Aboros, managing partner at Ringstone, or incidentally visionary partners for my PAI Breakfast Briefing Series. Welcome, Hazem. Thanks so much for joining me. Thanks, Steve. Thanks for the invitation. Good to be here. Well, look, first things first, I know I'm in the presence of a lot of relevant experience here. Can you briefly tell me a bit about your background and what Ringstone does, please?

SPEAKER_01

Absolutely, yeah. So, you know, I I come from a um I come from a software background mostly, spent about 30 years uh in industry. Uh a good portion of my career has been at Microsoft, uh working across enterprise groups. Um, but really uh most of my focus in the latter years have been on AI experimentations and data uh through incubating new products and team transformation, uh, which eventually led me to the MA work that I was doing at Microsoft and eventually partnering with private equity um to help with uh technical due diligence, uh cell side preparation, and also company transformations. Um and that that's kind of what we do at Ringstone. Uh for the past um you know 10 years I've been doing this. Uh we're a group of CTOs and CIOs with kind of what we like to uh call real-world experience. Um, and we help in technical due diligence um and uh company transformations.

SPEAKER_00

So thanks for having me. Yeah, okay, that's that's really good context. Thank you for that. Now, AI is not new, however, the landscape has changed quite a bit over the last couple of years. I think it's almost two years to the day that that ChatGPT went public. What's changed in the way that you support private equity with this new uh development?

SPEAKER_01

Yeah, no, I mean you're right, Steve, is in the sense that it's not new, you know, it's been a journey for AI and has been it has been long coming, but absolutely what what what transpired in the past uh couple of years uh took us all by surprise with Gen AI and uh the capabilities that we all have at our fingertips um as people and uh companies. Um so I would say in the past data was important, AI was sometimes important where it provided roadmap um advantages to certain companies. Uh but today I have to admit that both AI and the associated data for a company are uh really part of almost every other project. And to really bring that uh into light, I I would like to distinguish between uh two different types of companies, some that may be AI companies that provide platforms and and and solutions and models, et cetera. Uh while the others uh are, you know, frankly, everybody else uh as AI-enabled companies. So in DDs, you know, in tech diligences, we typically we typically take an approach of tying the technology and looking at it through a business lens. Uh for example, tying things to uh the impact on revenue, uh you know, how how much that could impact positive growth uh or negative uh impact on EBITA uh and uh efficiencies, and also ultimately on a company value, exit multiples and this sort of thing. Uh but I would say AI in particular these days has uh a very interesting uh impact on companies' efficiency and productivity. So there is a very direct link between that and Ebitamartians, and that's really the AI-enabled companies that I'm talking about.

SPEAKER_00

And and the other companies, how is there growing demand or appetite for them to move into that AI company space?

SPEAKER_01

The reality is is uh you know the change the change is big. Um and it's one of those changes that really impact the C-suite um and and and the company strategy. Um so for the AI companies for that category one, we see a lot of young companies and a lot of startups. The ecosystem is so busy at the moment, right? I can't even count the numbers of smaller companies that are seeking seed money and this sort of thing. But but I I really want to focus on the AI-enabled ones because this is rather the opportunity um or the threat for some companies. So it it does, as I mentioned, impact um the strategy of the company, the C-suite, and ultimately those who will participate and figure out how to actually be tech uh uh AI enabled are gonna be uh the winners in this game, uh as opposed to everyone else. And there are some, I mean, there's a playbook for this. It's not, you know, it's not it's not the Wild West. There is definitely a good playbook for it. Uh there are already companies uh that we see on a day-to-day basis that have made some fairly good progress and leveraged um some some of the efficiencies that AI could could uh could provide them.

SPEAKER_00

Yes, uh definitely there's more or more use cases uh coming out. And I think that I suppose what we um should all be doing is making preparations rather than just dreams of of where we might uh be able to end up. So if we were to to come to the subject of AI mobilization and and the the building blocks that I mentioned as the theme for the next breakfast briefing, what does mobilization mean to you? And is that we should be looking back at the um the digital transformations that we've run before? Is there something uh a new approach that we should be taking here?

SPEAKER_01

Yes, that's a great question. And it is a little bit different, you know, than I think of the digital um, you know, enablement or even things like agile transformations. I mean, these definitely had an impact, but the AI one is a bit different because it really it really starts from the top down. I don't want to be little digital transformations because you know they have been uh really important for companies in in in the recent decade. Uh but with AI, it really does require uh uh a lot of top-down shift in thinking at the C-suite level and also some actionable steps that they need to take uh into their strategy and their approach. And I I would say there are really three components to it. So component one is data, which oftentimes we take for granted, the data story, you know, from the past. Uh a lot of it I would say you see a lot of lip service. Yeah, we're data driven, this, this, that. But really, when it comes down to it, it's much larger than that. It's it's really about quantity, quality of data, it's about continuous cleansing, it's about mix and matching product data and operational data and customer data. And and really something magical happens once you begin to put that together in a machine-readable way, that that um some insights begin to surface. And then in addition to that, I would say the the second big pillar here is really around starting the conversations. And that's where I would say a lot of companies may be lagging. Um, and this is the conversations around use cases. And whilst there are a bunch of use cases or things that AI and Gen AI are good at today that are pretty clear to us, uh, the reality is some of the use cases are going to be very specific to a particular vertical, to a particular line of business or or a company for that matter. So the most important thing is from the executive level down is that they start the use case conversation as soon as possible. And it's not an event, it's rather a journey, a continuous conversation. But then the third pillar, which really the one that takes it all home, is adopting an experimental mindset, what we call concept testing. And and concept tests have been around for a long time, and and you know, just to touch on that a bit, it's really a way to spend the least amount of time and money and figure out whether a particular initiative or or an experiment or a feature or what have you can really um is worthwhile investing in further. And in the age of AI and Gen AI, um, I would say that having that mindset and really mastering the experimentation is what's going to make a difference for a lot of companies. So just to kind of zoom out and and and summarize this, it's about having the right data in place and and then having a continuous discussion on identifying use cases, and then thirdly, having an experimental mindset to go test these use cases to understand what's possible. You would think you know, AI is is is not as deterministic as software is, and therefore the experimentation is key because you really get to understand what are the possibilities. So these are the three pillars for mobilization in my head.

SPEAKER_00

Yeah, that last um point around um experimentation, it's curiosity as well, isn't it? With such an emerging technology, um, we don't know quite what falls into um what's possible or not possible. And so that curiosity and that continuous um what you call concept testing is is really important to know what path to move uh towards. Um I suppose that brings into focus as well is what should be bought, um, what should be um uh built as well. So where does that come into the conversation? Is it that concept phase or is that something that you know depends on other factors?

SPEAKER_01

That's uh that's also a very interesting one, and one that we often hear. I mean, the reality is the the buy versus build uh you know cannot be overemphasized here. It's it's it's really it's really an important balance that companies strike. So even even in software, that's really important because it it can determine your your speed to market, your quality or focus, for example. I mean, you don't want to be building stuff that uh is not your core business because that could be a distraction and others may be able to do it better. And with AI, I would say that's still very much true, even more so, because not every company is gonna be going and building models and housing AI experts and specialists and what have you. So it really I like to think about it that most companies um should focus on and on on their own businesses uh and and business models, uh, but pay attention to the data and house only the minimal um resources needed internally to again address those three uh pillars that I mentioned around data and use cases and experimentation, you know, be it a product manager, be it a data analyst or a data scientist, um, and and then finally you know a small team. And you know, just to be clear, when I talk about concept tests, I mean there is minimal firepower here. This is not big teams or anything. But but to your point, um I I should probably just underscore that because it's a big change for executives and C-suite, um all of a sudden really a lot of them to have that mind shift, they really gonna need a trusted um resour internal resource to to work with them, whether it's a dedicated resource or a shared resource. But really they you know there's an element of education and brainstorming. Um it's not something that every CEO, for example, is expected to be deeply familiar with. So the existence of these other resources is important. Um and related to that, just to uh answer your question very directly, the the right move is to understand what's an ecosystem, and those resources can help with that, um, and then ultimately decide well, what is it that we're building and what is it that we're not. So what good looks like is that companies are actually building the glue. They understand their their data, their ecosystem, their business, but they're building the glue to allow them to leverage AI to its fullest capabilities, you know, so it's minimal stuff internally.

SPEAKER_00

When we're building up to potentially the ambitions of being AI enabled and that we can see uh a portco can see an opportunity to to leverage AI with their proprietary data that they've they've um uncovered. AI is not cheap. So how how well aligned is private equity and the portfolio at the moment in terms of knowing how how much this might cost ultimately in the end? So we've moved away from those those tests and now we're into the production and and maybe that's build, um, maybe maybe it's buy. But there's going to have to be at some point a realization that this is going to be quite a bit of additional investment that's going to have to be um added here. How ready um RP for that?

SPEAKER_01

I could tell you what we see. Um I could tell you that most private equity are making investments in one form or another. Um, some of them are helping uh their pork's get the conversation going or facilitating the conversation. Uh are a little bit more sophisticated where they're actually housing shared resources, shared experts that they can lend to companies to partner with executives and and other uh members on the teams to to really uh make some progress on some of these experimentations or or uh or goals that they may have outlined. Um and and and and some other private equity, um, not too many, but from what we see, some are actually building uh shared uh components that could be leveraged. Um but there is one thing most of them are doing, which uh as you can guess, you know, the ecosystem again it goes back to the buy versus build. You could see the the ecosystem of line of business tools, for example, um is quickly developing and everybody is adding uh AI enablement in a lot of these tools. So, you know, be it Salesforce or Copilot or you know, all the vendors are doing that and slowly that's evolving, but surely it's headed in that direction. And the reality is a lot of the companies are gonna be leveraging these tools anyways, because if you're using Salesforce, well, you you you now have additional capabilities. But if your data is also well enabled, you can even take more advantage of those uh of the AI capabilities and in the line of business tools. So, really, again, it's it's about four different um things that private equities are doing today to to kind of help their porkos uh leverage the power of AI.

SPEAKER_00

Yeah, I think um I I've been um privy to you know some conversations that uh are describing quite a uh you know a steep learning curve, essentially, for uh private equity to to better understand what it takes and and often it can be the you know the port code leading the uh the discussion and hence why you know the the need for um the capabilities of of Ringstone. You mentioned at the top around um you know a key metric um for us to measure with AI is is ebit DA ultimately. But what else should we be measuring here to understand the impact?

SPEAKER_01

Well, there's there's a number of number of metrics that are common. I mean, Ebitah, the impact of AI enablement and how it affects EBITDA, but if you parse that, there are a lot of different things under it. So for example, um companies could focus on RD spend. Um one recent study uh from a partner of Ring Stone that I was exposed to, uh that it's uh effectively uh a development house. Um they help with outsourcing and and really constructing real teams. You know, I I like to call that their model partner shoring because it's really setting up teams for companies that work on roadmap as opposed to the classical stuff. But the study that I saw was quite interesting, it's very recent, actually. They they have conducted some experiments internally to measure the development speed with and without AI enablement. And the data that I saw was just stunning. Uh for example, um in the latest round of of experiment, the data showed 65% improvement in speed. So that's kind of that's a that's kind of a major one because when you really think about it, there is a quality piece attached to it and then there is a cost piece. So if you have your developers and or engineers in general able to work 65% faster, whether do it you whether doing unit tests, for example, or writing actual code for features, that's a massive improvement. I mean, even if you capture 30% of that or 40%, that's still massive. So then the RD spend is going to shift for a lot of the companies. And and if you really consider it for a bit, what's gonna happen is some of the benchmarks that are out there are no longer gonna apply. For example, those who really take advantage of that, they're gonna be much further ahead, delivering quicker uh products to market, delivering better quality at a cheaper cost. Uh so you know, from a competitive stance, this is massive. Um so that's I would say that's that's the biggest one. I did speak of just R D right here, but the reality is that it impacts teams broadly. Um so for example, you could I could see scenarios where companies may reduce um certain staff, but but it's not it's not something that I'm promoting in any format. But I'm just calling it out. And it's it's absolutely the wrong thing to do to jump to that first. I would rather companies um begin to educate and enable their staff and train them on how to how to leverage uh AI in their day-to-day so they can be more efficient. And and that may have very positive implications. So I would say overall R D, um overall um revenue by employee, um, we could see that metric changing as well. These are two that come to mind.

SPEAKER_00

Yes, I think um well, just taking your point there about um giving employees access and and I think that goes back to almost giving that the the opportunity to to experiment and see where there might be opportunities to to improve their productivity but also in improve their their output. In terms of I mentioned right in my introduction about failure and and that um you know a lot of the time um as a as a product person and and having seen a lot of and tried a lot of things out with AI, you know, that that there's a culture of it's okay to fail, but only to a only to a point, you know, you've got to learn from that failure, and when you're starting to pump resources in, um, you know, that the ideal is is that you're um reducing your failure rate. What are the common challenges or mistakes that that companies face when trying to mobilize that that these projects just fall flat before they've almost got started?

SPEAKER_01

Yeah, that that's that's a good one. You mentioned the experimentation mindset and kind of what I touched on at the very beginning, which is really it's a mind shift, especially for you know executives. So acquiring that mind shift is really important, and and the three initial pillars around clean data and and and identifying use cases. You know, so these are the core things, but it goes deeper than that. Um we actually outline, interestingly enough, the the top five things that that companies challenges that companies typically face when they're implementing AI. So one of them is the the top one actually is messy data or data with major gaps, incomplete data and this sort of thing. Because of the nature of AI being probabilistic, not having the right data, it may not be able to give you what you need exactly. So therefore, again, the the the cleanliness of data and the the quant the quantity as well, that's a big deal. Um the second one is is the is towards that mind shift. You know, we're all used to software, and you know, you you tell it to A and it does A and a few bugs and until you get it right, very deterministic in nature. But with AI there is a how-to deal with AI mindset you know, to to understand uh probabilities and and how the outputs are, and therefore the experimentations are key. Um the third one is around nascent use cases. Um you know it's greenfield tech in general, you know, fewer best practices exist and people are learning as they go. So having having deeper conversations and allocating the time on the agenda is important, ensuring that you know, component of that, for example, is ensuring that those conversations are actually part of the company's strategy and the roadmap, frankly, uh, you know, to identify that white space, that the the use cases that are very specific to us, and being able to measure the value around those. The fourth one, it goes back to this concept test. Uh you know, AI development is iterative, and the timelines are definitely less certain. So it takes time to find patterns and identify and and tune different models, be you know, be i w whether these models are are you know belong to the company, which is the unlikely scenario, or you know, third-party models, but but ultimately there is a fine-tuning exercise that must happen. So there's a lot of iterations. And often we see companies kind of given up quickly, or they just run one test and then you know they give up or quickly determine that it doesn't have any merit. So there's a there's really a lot to it. And then the the final one is around that transformation mindset. You know, for every company today, I would say it is gonna take time to develop new skills in the organization, both at the C-suite level but also at the team level, to figure out what resources they actually need to house inside or uh what to leverage from outside. There is a balance there. And I think because it's because it's fairly new to most companies, that will take time to figure out. Um and then the only other thing related to that is it really does mean that companies have to slightly modify their ways of working. Um I mean it's already happening, but but it will. If you think back to agile transformations, although it it's not the best analogy, but it's the closest one I can think of. That did modify the way the executive suite worked a little bit, it modified how teams worked and collaborated together. AI is gonna be much more powerful than that. And a typical agile transformation to be successful requires rigor, um, continuous effort, uh, fine-tuning. Uh same with AI. It's gonna happen where it's gonna take a good amount of time for an organization to mind shift and be able to work with AI and Gen AI. So in summary, there are really the the five the five aspects that could uh lead to uh AI initiatives failing is not having the right data or or or clean data. It's not understanding the the difference between deterministic software development and probabilistic AI and expecting the wrong out you know, expecting different outcomes. Uh having nascent use cases or use cases that are not well developed, not well thought out, or you know, uh not having any at all. The uh understanding that development itself is different because it's very iterative and it's very agile uh as an approach with continuous fine-tuning. And then finally, it's around the ways of working in an organization and how that ultimately requires an organization to shift their DNA a little bit.

SPEAKER_00

Uh I was talking to a CEO yesterday, he's so enthusiastic about this technology, and it's just giving him his renewed and renewed uh approach to his business. And I think that if we all have even a small amount of that to think that how our work could change, but for the better, and how much more we could perhaps enjoy the process. We all know that that learning motivates us, and there's a real opportunity to do that. And I just thought that conversation yesterday really put it into picture. If you uh if you lean into this conversation and you can see how you could break some of the uh maybe some of the habits that you've had for years and years, and that there's a different answer uh to the question, I think it's quite exciting. Um look, final question, uh Hasa, is um when your things are changing so much, even in the last 24 hours, um, you know, uh ChatGPT have come up with a a a new product. So you know it's it's hard to stay on top of how things are moving. How do you do that then in in your job?

SPEAKER_01

Well, I I think you just nailed it with the example you gave of the CEO. I think it's a continuous learning mindset. Both at the individual level, at the organization level, and uh ask for help when it's needed. So I I completely concur with what you said. In fact, uh it's exciting to hear that because that that is the reality is that everybody has a responsibility to, or or put it this way, everybody has a short-term opportunity to grow and learn about this new quickly developing ecosystem, but also they have a long-term responsibility to make it work for their organization. Uh so it that continuous learning mindset and accepting that yes, that we have some things to learn. It's gonna be iterative, we're gonna, you know, we're we're gonna experiment and until we understand what works for us and what ROI um you know we can get from it. So it, you know, in a nutshell, it is that continuous learning.

SPEAKER_00

And that's what this podcast is about as well, is to help um people understand um and navigate um through this this really interesting um path that we're on. Thank you so much for your time as of today. I really enjoyed uh the conversation. Thank you very much.

SPEAKER_01

Thank you very much, Steve. Uh really appreciate the time and uh intriguing conversation and uh looking forward to um you know seeing the podcast. Thank you very much.

SPEAKER_00

I've enjoyed it. It's been um a pleasure to have you on. And people can connect with you on LinkedIn or check out ringstone at ringstone tech.com. But for now that's it. Thank you so much for listening and see you soon.