Enjoying the podcast? Don’t miss out on future episodes! Please hit that subscribe button on Apple, Spotify, or your favorite podcast platform to stay updated with our latest content. Thank you for your support!

In this episode, Carol Hamilton, Chief Product and Strategy Officer at Provenir, discusses how AI-powered decisioning platforms are transforming risk management across financial services. She explores the critical balance financial institutions must strike between managing evolving risk threats and maintaining seamless customer experiences, a challenge that has become increasingly complex in today’s uncertain macroeconomic environment. She explains how Provenir’s platform helps organizations make intelligent decisions across the entire customer lifecycle, from onboarding and fraud prevention to collections, by orchestrating real-time data and AI to provide contextual insights that enable both risk mitigation and opportunity optimization.
The conversation delves into key findings from Provenir’s 2025 Global Risk Decisioning Survey, revealing that over half of respondents struggle with data integration while 60% find it difficult to deploy and maintain risk models. Hamilton emphasizes how generative AI is being leveraged not just as a trend but to drive tangible outcomes, speeding up decision-making processes, enhancing model explainability, and analyzing unstructured data. Looking ahead, she describes 2025 as “the year of intelligent decisioning,” where organizations can move beyond traditional rules-based systems to achieve the perfect contextual understanding needed for hyper-personalized customer interactions that unlock value while effectively managing risk.
In this podcast you will learn:
- What Provenir does exactly.
- The types of risk decisions they help their customers make.
- The biggest pain points in risk management for financial institutions today.
- The different geographies where Provenir operates.
- Why they have focused on enterprise businesses.
- How they approach the tension between preventing fraud and seamless customer experiences.
- How their clients are managing risks in today’s uncertain environment.
- The different types of simulations they can run inside their platform.
- How they are using generative AI in their decisioning engine.
- How their approach differs from others in the market.
- The purpose of their 2025 Global Risk Decisioning Survey.
- Some the of the core findings from the survey.
- How they help their clients deploy their risk decisioning models.
- How they are working with lenders with credit risk and detecting fraud.
- What they mean by calling 2025 the “year of intelligent decisioning.”
- How they approach product development given the different needs of their customers.
- What is next for Provenir.
Read a transcription of our conversation below.
FINTECH ONE-ON-ONE PODCAST NO. 539 – CAROL HAMILTON
Carol Hamilton: The biggest challenge we most commonly hear about is the balancing of that managing risk and the view of risk and how those threats are changing whilst maintaining and not maintaining a great experience and not sacrificing on that customer experience. So, balancing the two, that’s the major pain point. Another major pain point is really the growing urgency we see to modernize legacy systems and be able to adapt to fast-changing risk environments.
Peter Renton: This is the Fintech One-on-One Podcast, the show for fintech enthusiasts looking to better understand the leaders shaping fintech and banking today. My name is Peter Renton, and since 2013, I’ve been conducting in-depth interviews with fintech founders and banking executives. On the show today, I am delighted to welcome Carol Hamilton. She is the Chief Product and Strategy Officer at Provenir as we discuss AI-powered risk decisions. We cover how their decisioning platform works, the challenges faced by financial institutions today, how the company is leveraging AI and generative AI to enhance the decision-making process, findings from their recent Global Risk Decisioning Survey, and much more. Now let’s get on with the show.
Welcome to the podcast, Carol.
CH: Hi, Peter.
PR: Good to see you again. And let’s kick it off by giving listeners some background about yourself. Hit on some of the high points of your career to date.
CH: Sure, thank you. It’s great to be here. I am Carol Hamilton. I’m currently at Provenir as the Chief Product and Strategy Officer. I’ve spent about 20 years working across the world to help market-leading organizations in mostly financial services, and some other sectors, unlock value from data and analytics. So, I’m a mathematician by training, but have been in and around the banking world, consulting and predominantly software industries leading teams to engage with customers and deliver great solutions. So, highlights have been working with very, very big organizations. I was at SAS Global Lead and Analytics through to much smaller and different types of organizations. So different types of people in the past.
PR: Okay, so then let’s talk about Provenir. What do you guys do exactly? How do you describe the company, and what’s your role?
CH: Sure. Provenir transforms how organizations manage their customers with AI. So what’s behind this is that the idea that organizations around the world, they’re having a very continuous set of interactions with their customers, and they need to very dynamically balance those interactions with decisions around where’s the risk and how to manage that with actually, there could be some opportunity here to unlock for their business or the customer. So we have a decisioning platform that helps organizations optimize those customer interactions. And it’s driven by sort of an AI sophistication and the need to be very customer-centric and very personalized with those decisions.
PR: Okay, so then what type of decisions are we talking about? Every business interacts with their customers in one form or another. So, what specifically are you talking about?
CH: Sure, really across the customer lifecycle, there are pockets of sort of problem areas that we work to tackle our use cases. So everything from that initial onboarding of a customer trying to maximize the acceptance rates at that point whilst keeping friction sort of low and manageable, that’s a very strong area where the need for speed and accuracy is super important. And then beyond onboarding throughout customers’ sort of life cycles with organizations, it might be that there’s a very low risk that is monitored, and therefore, there’s the opportunity for upselling and engaging with them with other products, but also there could be risk and an evolving view of risk for that customer. So it’s about trying to mitigate and make great decisions to try and minimize the impact of that risk on the business and sort of introduce various processes to intervene then with the customer. And then let’s say the risk has increased quite far, and the customer is then in collections; there are decisions to be made about how to optimize that collection activity through tailoring and personalizing channels of communication strategies to go and collect that money. So, use cases across the customer life cycle. That’s really with any organization who is offering a financial service. So yes, you think of banks and other obvious financial services providers, but this is also what we do for telco retailers and other parts of the world as well.
PR: Okay. So then let’s talk about financial institutions cause that’s the focus of this podcast, and where most of the listeners reside. What are some of the biggest pain points when you’re talking with financial institutions regarding risk management today?
CH: Sure. So the biggest challenge we most commonly hear about is the balancing of that managing risk and the view of risk and how those threats are changing whilst maintaining a great experience and not sacrificing on that customer experience. So, balancing the two, that’s the major pain point. Another major pain point is really the growing urgency we see to modernize legacy systems and be able to adapt to fast-changing risk environments.
PR: Right. So your software, I presume, obviously can work with legacy systems as well as the latest software, because I know that you have traditional banks, and fintechs as customers of your organization. So, how are you kind of interfacing with the existing systems that are out there?
CH: Sure, so it’s API-driven technology. To be honest, whether or not it’s a sort of a cutting-edge fintech or a more traditional type setup in a large bank, it’s really the same system that’s interacting in the same way. And that’s because there’s an emphasis on that flexible sort of API microservices-driven approach that has scalability and agility at its core. So, how we connect and engage with different-sized organizations is the same. It’s about using modern connections, modern connectors to be able to digest and serve back that information really, really quickly.
PR: Right, gotcha, gotcha. Okay. What geographies are you operating in? I mean, are you a UK company? Cause I think you’re, you’re based in the UK, right? Are you across the world? Where are you operating?
CH: Yes, I am in the UK, but we are very much a global business. We have tier-one and market-leading organizations that we’re serving across the world. We operate really through four main geographies. We have a North American business, a Latin American business, a European, Middle East, and Africa business, and then APAC. And I would say that whilst historically, 20 years ago, in our sort of early essence, we grew out of North America, we have a footprint of people and customers across the world. We serve over 140 customers around the world, and that’s in 40 different countries. So, very much a global footprint and helping market leaders and stronger innovative players across the suite.
PR: So then are you mainly serving enterprise level customers or do you ever work with startups?
CH: We could work with startups, they often don’t have the scale breadth of the enterprise that really most beneficial, so benefit sorry, from the system. So I would say predominantly our audience is the enterprise organizations. Why? Because they are the ones who feel the pain of trying to operate at scale across different product lines, different geographies, and different customer segments at speed, and be agile to adapt in fast-changing risk environments as well as meeting customer expectations, which are forever on the up. So the enterprise business does suit us well. Also, because we have a single platform that serves all the use cases across that customer lifecycle, it means that within an enterprise, we can satisfy requirements for credit risk and fraud and collections and the customer personalization engagement areas as well. So it very much suits a bigger business. It doesn’t mean that we sort of the tech wouldn’t work for a smaller scale, but our business model is to work with enterprise tier ones.
PR: Right. Gotcha. Okay. So one of the subjects I think it’s really critical today, more than ever before, it seems, is fraud prevention. And you have this natural tension, it seems, when you’re talking fraud prevention between, you know, actually preventing all fraud versus providing an experience that is seamless for your customers. How are you guys approaching that tension?
CH: It is a real tension. In fact, we did have a fraud roundtable yesterday with some customers and industry leaders, and they still talk about the tension that exists between different departments. Let me explain it this way. So imagine there’s a front door to the organization, and one camp is trying to keep it shut to keep all the fraud out, and the other camp, driving new business, is trying to swing it open and let everyone through. So it’s about how closely do you sort of close that door and sort of where do you keep it as the environment changes. And, in fact, yesterday, another example of that was the tension being described between sort of payments departments and the need for speed, as well as, again, serving customers in a way that protects them from that fraud, which, if instant, might mean instant losses. So it’s a real balance to find where that’s right. So, how we help is in a few ways. We’re firstly unifying and bringing together in real time the orchestration of lots of data that’s relevant, as well as AI. Why does that help? Because together, they give very strong contextual decisions, and that leads to a good understanding of who’s at the front door. So you can swing it open and let them fly in quickly, or you can close it quite quickly as well if you need to. But for fraud, it’s not so black and white. Sometimes, the door needs to hover open because there are gray areas. So, it could be that a journey needs to be disrupted a little bit by introducing an extra step or two rather than stopping it in its entirety. So, trying to navigate that tension is about bringing the right tech approach to bring clarity and context to the decision being made so that it is really serving the best possible outcome.
PR: Right, right. Okay. Okay. So this year, 2025, has seen, let’s just say, lots of uncertainty in the macroeconomic environment. We’ve certainly seen, you know, wild swings in public markets for both the stock market and the bond market, and people are worrying about recessions and all that sort of thing. How does this impact your products, and how are your clients managing risk through this uncertain environment?
CH: Sure, well, it’s definitely a struggle because whether or not we’re talking about credit risk or fraud risk, the landscape is changing, the behaviors are changing, and also there’s consumer nervousness at the same time. So, there’s lots of different factors to manage. And I think the way that we’re successfully working through this with customers is helping to ensure what we can offer them and how they can operate is a very flexible way. Because if you have something that’s rigid, either a process or a tech platform, that does not allow for the adaptability that’s required for today. So by focusing on flexibility in our platform as an example, that’s allowing our customers, and we’re helping them see the benefit of quickly adjusting models or integrating new data sources or treating a certain customer segment in a different way that month, just to try and give that sort of that control back to them as much as possible. Other things that we’re doing is encouraging them to help really give them power to simulate outcomes because by simulating and understanding, you know, if we change this sort of approach, that might increase the risk in a certain way and change our customer experience and they can almost try and evaluate to the best of their ability before they then action things. So, I think that simulation is important. But overall, it’s just being agile. I guess I see agility as that key to staying ahead of volatility that’s out there in the market.
PR: So, does your platform allow for those different types of simulation natively inside your platform?
CH: Yes, absolutely. In fact, that’s one of the things that we’ve been investing a lot in at present, and we have a forthcoming roadmap that will deliver more on that topic. But it is about giving, I guess it’s because there’s the juxtaposition of managing risk and opportunity, sort of optimization. And so by bringing those together through simulations, we help people understand the impact on both. Because if you’re looking in a siloed way, just at risk and how that might change, you sort of miss how that affects the opportunity you have to manage your customers in a better way and deliver more value to them and your business. So by creating more sort of interactive pieces in our platform for our customers on simulation, we’re able to help them understand and try out the art of the possible, which is proving very, very important.
PR: Right, for sure. So, you know, it’s on your homepage, your website, you talk about an AI decisioning engine that you guys have. I’m curious about generative AI. You’ll see AI has been around for a long time, but I’m curious about, you know, generative AI has really come to the fore in the last two and a half years. How are you using generative AI in your decisioning engine?
CH: Sure. So we get asked this quite a lot. I mean, for years, we’ve been using AI through machine learning models, but GenAI, you’re right, is a different beast, very different, and cannot be ignored. I mean, as individuals and consumers, we’re in and out of GenAI all the time. Why? Because it’s firstly increasing the speed at which we can do things and just bringing intelligence and insights that we might not have known. So that’s really how we’re using it in the product. So we’re trying to leverage GenAI in a few ways. Whilst not jumping on a trend in a bandwagon, it’s always about tying it to the outcomes we’re driving anyway. So, trying to drive the accuracy of the outcome. So, as an example, we might be trying to, through the dashboards that are explaining how the models are working, we’ll use GenAI to instantly create summaries so people can digest and understand that information much quicker, so that helps the user understand something they were going to look at anyway, but with greater speed and sort of context. And we’re also using it to sort of speed up just common tasks, you know, in a similar sort of format, just helping the user who is still going to be there interact with the system in a better way. And then the final way we’re looking at using GenAI is to really explore what we can do with unstructured data and how we analyze that and again, bring that to the decisioning moment to improve the outcome.
PR: Interesting. So then, you obviously operate in a market that has competitors. I’m curious about what your approach is that distinguishes it from others in this sort of decisioning engine marketplace.
CH: Sure. The three things really that stand us apart. The first is that it’s not necessarily all about risk, which, you know, it was for so many for a long time. It’s about really maximizing that duality of understanding the risk, but also, in the moment, trying to work out the opportunity. You know, customers, through their engagement with organizations, will change. Things aren’t so black and white as we’d like them to be. So actually, it’s about understanding that continuous engagement, which I think through our AI investments, we’ve been able to really hone in on and deliver that holistic approach. The second is the demonstrable value we’ve been able to deliver across a breadth of sectors and sizes of organizations, because, at the end of the day, whether or not you’re a bank or a payments company or telco, you need to be able to interact with your customers where something is about to happen, either an action is about to happen, or some sort of product or service is about to release. Everyone’s trying to just make the right decision in the right moment about those interactions. And so, the fact that we can do that in a flexible way across multiple sectors, but bring the SME and the depth of expertise in there is, I think, very, very powerful. And then the third is that I think we’ve been able to actually craft our own unique position in the market. You know, we’re often asked, who are our competitors and you could look at credit risk, decisioning competitors, and we sometimes overlap in that space. We’re often asked, are we sort of AI platform and model builders? And because we have that sort of capability, you know, we often get compared to those, get compared to fraud tools, and orchestration platforms. I think that whilst that almost doesn’t make us overlap very distinctly over one area, I think that’s our strength because we bring a very unique view to interacting with customers for the organizations we work with that is all driven on improving those interactions from a risk management and also unlocking value perspective, which helps them grow their business and thinking holistically in that way, positioning ourselves, and carving that out and driving it forward with AI is, I think, our third uniqueness.
PR: Okay, so I want to switch gears a little bit and talk about the 2025 Global Risk Decisioning Survey that was recently released by you guys. Maybe talk about what was behind this report, why you decided to produce it, and what your methodology was.
CH: Sure. We surveyed, I believe, just over 200 decision makers globally to capture challenges, priorities, innovations. We like to do these because they provide quite actionable insights and benchmarking just to help us understand trends across the landscape. We’re building technology. We want to make sure it’s relevant and is going to our values. So it’s something we try and do every year just to do a bit of a temperature gauge.
PR: And were there any surprising learnings? Tell us some of the core findings that you wanted to highlight here.
CH: Sure. I think it was a mix of same old problems that continue to be difficult. You know, the orchestration of data, having it not be siloed, doing things quick enough, deploying models, and getting the analytics, you know, at its very best. Those are all issues and pain points that have been around for quite a long time. But we have also seen sort of newer things come in, you know, the mention of AI, the discussion around trying to also maximize customer value whilst trying to focus on risk. So they really are enforcing, you know, I guess our path forward and we know we’re on the right track with that. I mean, I’ll give you one example. There was, I think it was just over half of our respondents are struggling with integrating data sources. And so, you know, when we think of data orchestration, you know, this has been a challenge for a long time, but we then just continue to build out what we have, which is a marketplace to allow instant access to a global set of data providers to sort of help facilitate that. I guess, yeah, some interesting outcomes for sure for debate, but lots of validation to make sure that people continue to innovate and tackle these problems in a certain way.
PR: Right. So, one other thing that struck out to me was 60 % of the respondents finding it difficult to deploy and maintain risk decisioning models. Maybe you can talk about how you’re addressing that. You know, obviously, because there’s two different things there, right? There’s putting it in in the first place, and then there’s making sure it stays relevant and useful. How are you helping your customers do that?
CH: Each customer is quite different, but most of them like to build their own models and then perhaps compare them to models being built by the system and challenge each other and choose a better way forward. So the way that we’re helping them deploy their models, which most of them have, is, we have a model ecosystem. It’s, again, true to our nature of being very simple, low code. It’s in a very attractive UI, easy to use, that allows them to deploy those models through drag and drop into decisioning processes very easily. And that has been a game changer. Rather than building sort of manual connections, to be able to do it via a UI in that ecosystem has been very important for us to help customers do that. But you’re right; that’s one part of it because they might have built that model over six months, and now it’s the moment to deploy it. But then actually what we’ve had to do over the last couple of years to prevent them waiting another six months to get to a second build is to be monitoring those real time, and sort of allow for the retraining in a much more easy fashion. And also, by bringing up champion challenger models as well, which could be deployed instantly as replacements where needed. So, yes, deploying is one thing, but screens and visualizations to bring that explainability and monitoring to allow for quicker retraining have been a mega focus point for us.
PR: Right, right. Okay. I want to talk about lenders real quick because it’s one of the other things that I saw in the survey was, you know, managing credit risk and detecting fraud remain top challenges for lenders. How is Provenir working specifically with lenders today to help them here?
CH: The way that we work for those two problems is about really focusing on the outcomes and the business goals to drive a great tech solution. We’re very aware of how to solve for credit risk management, be it onboarding or in a sort of ongoing monitoring way, because we understand the goals, we understand their KPIs, we understand it’s about maximizing those acceptance rates without faltering on that smooth customer journey. So by understanding those goals, it helps us to focus on the right things in the tech. So it’s about doing things quickly, accurately, in a very sophisticated way, in a cost-effective way. You can’t go and orchestrate every bit of possible data for everybody. So, I think we are able to help organizations through the absolute intense focus on those outcomes for them. And for fraud, it’s sort of similar but different. So, for fraud, the KPIs are a little bit different. Obviously, it’s about reducing losses again, same as credit, but not only those reducing fraud losses, it’s about sort of improving the false positive rates for fraud. It’s about identifying the most sort of high-value frauds. It’s about staying ahead of changing MOs, which is very difficult because if you imagine how much companies invest in tech, think what fraudsters’ budgets might be. So we’re always competing against that. But I think the best way to stay ahead with fraud is not only bringing AI in, which is an obvious point here, to bring that sophistication, but the data is even more valuable for fraud to get right than credit. Because if you want to make a great decision about fraud, the context of the moment and the customer is super important. And how can you achieve that great contextual understanding? It’s by understanding as much as you can about that customer. It’s around their identity, their potential fraud information from third parties, something about their email address, their device, maybe their location, previous behavior. You need a lot of that to make a decision. And, the AI helps you call the right data that’s going to be most effective for that type of customer, type of moment, and type of decision. But I think this is why we’re able to help customers because it’s a focus on those outcomes and then really designing a solution to just deliver on those.
PR: Right. Okay. So, you’ve described 2025 as the year of intelligent decisioning. I’d like to know what you actually mean by that and maybe what specific metrics or outcomes that, you know, financial institutions listening to this podcast, what should they be targeting?
CH: The year of intelligent decisioning, yes, you’re right. I think that for so long, we have had systems that were rules-based, that were perhaps doing things over time, maybe in batch. And, I think, you know, lots of evolution has happened over the last few years to really bring us to a place now where it is possible to achieve that perfect context to a decision, that perfect understanding about how a customer is operating and the risk they pose. So, when sort of focusing on trying to make the decision more intelligent, the gains to look for and measure will be speed to that decision, and then focusing on losses, reduction in losses from a fraud and credit perspective, as well as on the modeling side, time saved in those model deployment and update cycles as well.
PR: So, I want to ask about product development. Provenir has a range of different customers. You talked about telcos, fintechs, banks, different customers with different needs. So, what is your approach to product development in the context of the different needs of your customers?
CH: I guess there’s two parts to that that I’ll sort of offer up. The first one is that when we are innovating and evolving our platform, it is with flexibility at its core. And a real emphasis on scalability and agility, which then truly does allow us to serve organizations who want to deploy simple decisioning processes through to those who want to enact much more sophisticated sort of multi-model ecosystems and processes. So that focus on flexibility, scalability, and agility is very important. And I think the second part is for our product development, we have really grown our global centers of excellence and, in particular, over the last year or two, really invested in product and design functions, which really allow us to have much more of a customer-centric approach to development of new features so that we can see the synergies across those markets. We can work with customers more closely. We know the market moves very quickly. We have to focus on giving value to customers with new capabilities in months, not years.
PR: Okay, so final question, we’re recording this just almost the middle of the year here in 2025. So, as you look out over the next 12 months and 18 months into 2026, what’s coming down the pipe at Provenir?
CH: Well, what’s coming down the pipe is very influenced by sort of an event I went to in December, which is still stuck in my head. And it’s largely driven around AI. I think we can’t escape the potential positive impact of AI, as well as the sort of unknowns that are still being discussed and explored. And I remember very clearly hearing and witnessing that organizations have two types of fears around AI. One is that they feared not doing it and being left out. And the second is they fear doing it wrong. So when we look at our product roadmap, and what’s coming down the line, we’re very aware that that is the contextual thinking of our customers as they want to embark on projects with us and use AI-driven technology. So we have to, in the product, be developing things that help give them the comfort of embarking on an AI journey, if that’s a newer or sort of just a part of their evolution, but also being able to deliver early results. So I think that’s why we’re focused on; I mean, I’ve mentioned a few things; let me bring them back for the roadmap. So, more investment in the simulation, which is helping people sort of better plan and explore and predict and then enact very quickly. So that’s one. Strengthening the data orchestration is another. I think for organizations, as we’ve seen in the survey, it continues to be a mega challenge, as it has always been. But what’s the next gen of data orchestration? How can AI actually make that a smarter, more intelligent process to go and get the right data in the right moment for the sort of product and the customer in the moment being sort of decisioned on? And then the third is around bringing that sort of contextual understanding through better use of data. Not data being called in in that moment, but over time, how can organizations better leverage everything they know about their customers to, again, be ready to serve them in a very hyper-personalized way, which can unlock that value that they seek in a way that they’ve not been able to achieve before.
PR: Okay, well, we’ll have to leave it there, Carol. Really appreciates your insights today. Great to chat with you and thanks so much for coming on the show.
CH: Thank you, Peter. Thanks.
PR: See ya. We live in a world today where important business decisions for financial institutions are becoming ever more complex. This is mainly because of the massive amounts of data that are available. Everyone needs to make sense of all this data, which makes a decisioning platform like Provenir ever more valuable. With a focus on flexibility, scalability and leveraging generative AI, the company is well-positioned to help navigate the complexities of today’s financial landscape.
Anyway, that’s it for today’s show. If you enjoy these episodes, please go ahead and subscribe, tell a friend or leave a review. And thank you so much for listening.