We assembled a panel of experts from Business Transformation Partner, HSO, to discuss the state of finance functions in Financial Services and the role of AI and Autonomous Finance in this traditionally conservative industry.
On current challenges facing the Financial Services industry
Tom Berger: I think if we were to start a conversation about challenges in the financial services industry today, it starts with the overarching theme of doing more with less.
Jim Bretschneider: Absolutely. And underneath that, I think there are a couple of things going on. There is a good amount of consolidation – a lot of M&A activity. Often the business case is really around economies of scale and as a result, the cost of doing business needs to go down through automation. Finance is constantly working to attract the best talent. To address all these things, you need to have the best systems in place and reduce the amount of manual work as much as possible. Those jump out as some of the main challenges.
Also, if you are talking specifically about Insurance, I think there is a push towards more integrated systems. For instance, if you are in a Brokerage you need to have full visibility from your operational system – your AMS, your CRM – into the finance systems to be able to deal with producer compensation and other things in one big workload.
On regional differences across Financial Services
Tobias Menzel: From a European, especially continental European perspective, we are having a tough decade. Bear in mind we had this really low, zero interest policy set by the Central Banks and traditionally the cost to income ratios of European Financial Institutions have been traditionally bad. So, if part of your income equation went down, the focus over the past decade was really managing your cost base. So even though there were market demands for more digitalization on one side you could not simply react to that demand.
Instead, you basically were trying to manage your cost base and changing regulations – work that did not help a bit to transform that digital experience for customers, for finance teams, for anyone. From my perspective, we are facing a huge traffic jam around digitalization and automation and – speaking for the continental European market – we really have to catch up with Anglo-Saxon Banks and Insurers.
Tom Berger: We’ve absolutely seen slower growth in Financial Services in Europe whereas US Financial Services firms, in particular capital markets in the US, have seen tremendous growth and as a result, have displayed some irrational behaviors. Interestingly, that hasn’t been the case in Insurance necessarily.
On the Autonomous Finance journey
Tom Berger: The phrase Autonomous Finance is an interesting one because I think that the Finance suite, especially in conservative industries like Banking or Insurance, is probably the most conservative part of the of the business so I have not really seen anyone move toward an autonomous state. I think that people are interested in and learning about it. But no one wants to be the first one to do it. I don’t know about what you both think.
Jim Bretschneider: Yes, I agree with you – there is still a ways to go when it comes to automating finance. A lot of organizations are still dependent on disconnected, personal Excel sheets. To have a reasonable chance of using AI and other tools to automate your processes you’ve got to have a shared system, a corporate or organization-wide system, and not be dependent on the hundreds of Excel sheets that people are using to do their jobs.
You’ve got to get your processes in order first and then you can start moving from manual processes into automated processes then from automated process into AI-supported or AI-managed processes. Even the best organizations are still somewhere on that journey. I don’t think anybody is getting to the point where all this is happening from end-to end, but you’re starting to see pockets.
I would say one example is AP invoice processing, where organizations are starting to process these using AI. As the comfort level gets high enough, teams are starting to simply review the work that AI produces.
Tobias Menzel: From a European perspective, we haven’t seen that many Financial Services institutions who have already moved towards an Autonomous Finance function. From my personal perspective, this is because they simply haven’t recovered from the finance transformation programs that have taken place over the past 12-15 years where you had a lot of systems and underlying processes standardized and transferred into low-cost shared service centers. Here the ultimate goal was to free up valuable time for onshore finance resources to truly act as a business partner and strategic leader for the organization. But very often, unfortunately, the corresponding governance procedures were not in place, so you had this tremendous amount of investment, but the original business case failed to materialize.
Now, I think a key challenge, as I think about Autonomous Finance, is how to explain the difference between traditional finance transformation and an Autonomous Finance journey. What’s the gap and what is the promise of Autonomous Finance? How do we help companies pursue Autonomous Finance in a way that avoids the huge investments undertaken in the past that failed to fully deliver? I think this is really key.
On the relative infancy of AI
Tom Berger: It feels like longer, but we’ve only really been talking about AI for a maximum of twelve months! We have to remember we’re talking about – not a sales function, not a marketing function – we’re talking about applying AI to the financial core of the business. Giving a machine autonomous authority over the finance function is a far leap
from where finance functions are currently at. Yes, finance teams are eager to understand how they can do it, but they need the assurance that if they do it in a highly regulated environment, it won’t end their business.
We’ve all seen cautionary tales of trades happening or decisions made without human oversight. At the end of the day Finance teams want to make sure that this is done in a comprehensive and secure way and with partners that they can truly trust with their business. It’s a lot to put on the table.
Jim Bretschneider: I think it’s going to be a journey. The systems that companies are using 3 to 5 years from now are going to look very different than the systems that we’re seeing today.
It’s a matter of getting started on the journey. In the beginning it’s not about handing over every task to AI. It’s having tools that can support you – that notice when something looks a little different or doesn’t fit a pattern. That dig into the details or the underlying documentation to help you understand an allocation or the makeup of a complex journal entry. That builds a series of accounting rules for your review. I think we’ll see a lot of these copilot scenarios as opposed to just having AI to take over.
Tom Berger: 100% agree. I think another area where AI can help is providing rapid access to data. One of the interesting things I’m hearing my customers ask for right away is the ability to ‘chat’ with their data by using natural language processing to surface data points or to provide a Controller quick access to a particular report, for example. Of course, teams still need to make sure that they are accessing the data in a secure way and that they’re not giving permissions to access certain data to the wrong people. I think everyone just wants to make sure that they’re doing it in the right way. At the end of the day, it’s a journey and it needs to be taken in a careful and systematic way.
On AI regulations – or lack thereof
Tobias Menzel: What about the US regulations? What I’m seeing is the European Union commissioners are still discussing AI – is it moral is it anti-moral, what are the laws that need to govern use? Typical continental European behavior. I would expect that it’s much more ‘liberty and freedom’ from a US and Canadian perspective?
Tom Berger: Honestly the regulators haven’t caught up. The US President just created an office for this, basically a CIO for AI. I think that the regulators are way behind, and it could prove to be very dangerous. For now, from a regulatory perspective, existing regulations hold true, but I don’t think the existing regulations that I’ve read have adequately addressed the AI issue.
I think companies are self-regulating right now. Companies are more concerned about data leakage. They’re more concerned about mistakes. They’re concerned about hallucinations.
I’ll give you an example. I just got off the phone with a very conservative Bank, and they are saying ‘We need to do this right. We need to be very, very careful. We’re not jumping into this, and we are not going to risk our business based on this shiny object.’ At the same time, there’s this acknowledgement that when the floodgates open – and they will open – it will be a race because people will see it as a multiplier and a competitive differentiator. It will be another way to extract a few basis points from the bottom line.
Toby, you said it in the very beginning. The margins are so thin in Financial Services, and everyone is struggling to extract more value. Driving costs out in any way they can is key to the business. You’ve made your workers work harder. You’re extracting as much productivity as you can from them. What else can you possibly do to extract additional productivity? That is the central story, whether you’re above the line or below the line in these businesses.
Jim Bretschneider: Great points, Tom.
On guiding clients into the future of finance
Jim Bretschneider: That reminds me of an Insurance Broker we’re currently working with who was looking to modernize their finance applications. To start they were thinking about this in a relatively traditional way – putting everything they could think of onto their ERP solution because that was how it was always done.
Now we’re working with the Finance team to help them design a solution approach that matches their goals – in this case there are a few key elements. The first is building a finance architecture that supports an effective M&A strategy and allows them to access data from their acquired entities quickly an in a structured way.
The second is attracting and retaining producers, the independent Insurance Agents that sell their products. To do this they need to build an automated data flow to support Insurance Broker transactions to ensure they pay out the correct amount, as quickly as possible.
With these goals clearly established, we’re in a better place to discuss that actually loading up your ERP is not going to help you meet either of those goals. However, with an accounting hub supporting an ERP solution with a thin General Ledger, it’s easier to ingest data from acquired entities and centralize and configure those data flows in a more efficient way – while still using your traditional reporting tools.
Now you can really start to automate a lot of these processes that typically required data to travel from system to system via hard-coded integrations. Each element of the architecture is performing the job it was meant to do with seamless integrations establishing that consistent linkage.
On the value of partnerships
Tobias Menzel: I think a big difference between Microsoft and other major ERP players is the fact that you can still think big, but you can start small. This really is a completely different approach to this idea of having to be full stack from one vendor to ensure everything works.
Tom Berger: That’s a great point, Toby. I think about the composable ERP message that Microsoft VP, Georg Glantschnig has promoted from a finance and operations perspective. It’s completely in line with the best-in-class technology ecosystem that we are delivering through our partnership with both Microsoft and Aptitude.
Tobias Menzel: Exactly. And this underpins the whole premise of Autonomous Finance or any change initiative from my perspective, because you can have this big vision but implement these progressive solutions that drive change without requiring huge, upfront investments prior to seeing any value.
Our partnership with Microsoft and Aptitude offers the chance for Finance functions to create a new finance architecture that combines accounting hub and subledger technology, an accounting rules engine and then the Modern ERP/thin General Ledger offering from Microsoft – all tightly integrated to make reconciliation and drill back so much easier. And now, as of just last month, you have this announcement from Microsoft around Copilot for Finance. It’s extremely exciting to think about how this is going to help finance conduct their daily business. It’s really game changing.
Previously published in CFO Futures: An Autonomous Finance Magazine (Spring 2024)
