Generative AI: Fundamentals for Finance – Transcript

Jim O’Connor:                  

If you’re an FP&A – what we call financial planning and analysis – think about how Gen AI could turn you into a finance partner and advisor, right? We’ve got clients already that are streamlining their reporting and creating digital assistants to help the business and finance to make better decisions.

Announcer:                       

Welcome to The Hackett Group’s “Business Excelleration Podcast.” Week after week, you’ll hear from top experts on how to achieve Digital World Class^® performance.

Shawn Fitzgerald:           

Hello and welcome to The Hackett Group’s “Business Excelleration Podcast” session today. I’m your host, Shawn Fitzgerald, research director responsible for the finance enterprise performance management, account-to-report and customer-to-cash research programs here at The Hackett Group. And today we’ll be talking about generative AI in finance, specifically foundational fundamentals for finance. And this is a Gen AI podcast that’s part of a Finance Executive Series that we’re doing as part of The Hackett Group’s “Business Excelleration Podcast” series, which we’re titling “From Foundational Fundamentals to Creating Sustainable Competitive Advantage By Leveraging Gen AI Technologies Across the Finance Function.” And I’m joined today by Jim O’Connor. Jim is a principal in our Global Finance and GBS Advisory practices here at The Hackett Group. And Jim, do you want to introduce yourself and then we’ll get started?

Jim O’Connor:                  

Thanks, Shawn. I lead our overall Global Finance Advisory practice here at The Hackett Group, and certainly glad to be here. This is an exciting topic – one that has been now for a couple years and really hit the market this year – so I’m ready to roll.

Shawn Fitzgerald:           

Great. Looking forward to getting in today’s discussion. So for our audience members, please know that this series is specifically targeted for CFOs and financial leaders who have authority, influence, and budget across the executive leadership suite and within the finance area for educating, investing, and deploying Gen AI technologies both today and over time. And what’s really important is that Gen AI is a strategic imperative for both finance and the enterprise leadership at large. At The Hackett Group, we feel that a failure to educate, invest in, and deploy these Gen AI technologies is to literally risk extinction as a business – in the near term and future. So with that, let’s start off today’s foundational fundamental session. And so Jim, Gen AI is being used exhaustively in the popular and business media today. And while we have all heard the term, it’s important that we start our discussion with a detailed and contextual explanation of what we mean when we say Gen AI here at The Hackett Group. So Jim, can you provide this to our listeners as a starting point for today’s discussion?

Jim O’Connor:                  

Thanks, Shawn. I sure can. And as you mentioned, it is a hot topic, and at Hackett we’ve put a lot of energy into this – into researching this, learning about this. We’ve done surveys. We’ve done research. We’ve talked to vendors. We’ve talked to over 300 organizations. And, as you say, you just really can’t go through a day without hearing about it. I personally love to kind of start my day with the Squawk Box and see what’s going on in the market – and every day it’s a topic.

So I always say, the beginning of wisdom starts with a definition of terms. So let’s just talk a little bit about what is Gen AI, right? To simplify it, it’s really as simple as it’s the ability to generate new content. And many of us have used it already. What is that content – text, images, videos? There’s a lot of technical terms, large language models, neural networks, RAG, prompt engineering. But what should matter to you as a finance professional is it’s generative, and that’s different than other AI and other technologies. It’s probabilistic and generative versus programmatic. It’s going to make inference versus algorithmic. And you can start to think about how that’s going to open up all kinds of improvement for the finance organization.

Shawn Fitzgerald:           

So when we talk about Gen AI – because it is both a hot topic and it’s got broad implications – everybody’s talking about, well, how do we reimagine work and finance and other parts of the organization. We’ve got to understand the data requirements, the technology, the IT ecosystem. How do we think about our HR staffing and talent models with the introduction and evolution and adoption of Gen AI? Lots of different things across lots of different areas that finance leaders need to be thinking about – both by themselves and for their own organizations – but partnering with others both in IT and HR and other parts of the organization. So it’s just such a kind of big, hairy, audacious topic to start getting traction around. So what should people be thinking, again, this is for a finance audience, but what should people in finance be thinking about and engaging immediately as part of the commencement of their Gen AI journey, if you will?

Jim O’Connor:                  

I think first – from a finance point of view – to your point, it impacts HR. It impacts sales. It impacts product development. And finance isn’t isolated. And certainly a chief financial officer – or CFO – or many of the leaders within finance, they want to be thinking enterprise. So to your point, I think it’s OK to think about how is this going to impact the whole company? And what we’re seeing, particularly with our Hackett lens of G&A, is high-impact areas out of the gate are sales, are call centers, are product development, are IT, not surprisingly, coding and that type of thing. One client – we have a financial services company – they amped up their call center using it, and they’ve lowered costs by 25% and increased customer satisfaction by 10%.

So that doesn’t mean it’s going to fit every time. One of my clients, Michael, his point is, “Hey, I’ve had a lot of this technology – broader AI – around for a while. So right now I’m getting a 60% hit rate in my call centers, but if I could get it to 80, I’m interested.” So first of all, think about the implications, and at Hackett we are expecting, everybody’s given their prediction, but we’re expecting about a 40% overall impact to productivity and related costs and FTE savings for G&A. For finance, we actually think it’s going to be a little bigger – around 42% savings.

Now, again, that doesn’t necessarily mean that’s bottom line because jobs are going to change and how we approach is going to change. So if you’re in finance, we’re going to have a later session where we talk about just implications and very detailed use cases. But you could come back to my definition and you could think, “Gosh, well, where would I apply this?” But I would start with your strategy and think about what do you need and think about Gen AI? In and of its name, it’s a little cheeky, but I talk to my clients about for finance that should be about generating actionable insights.

So something finance is always looking to do is to enable the business through improved insights and improved reporting. And that is exactly the type of thing this technology can do. One metric we’ve always loved in benchmarking is percent time analyzing versus aggregating, and that’s going to change things here. Another kind of broad thing to think about – if you’re an FP&A, what we call financial planning and analysis – think about how Gen AI could turn you into a finance partner and advisor. We’ve got clients already that are streamlining their reporting and creating digital assistants to help the business and finance to make better decisions. So there’s a huge opportunity, and to your point, it’s large and we can talk a little more about how you start there, but I think first you got to kind of think top down what are you trying to accomplish? And then secondly, you can think about bottom up. By process, where are those implications and what can I do?

Shawn Fitzgerald:           

Those are all really great points, Jim. And again, to put some context and further into this for our listeners today, our own research shows that Digital World Class^® organizations are already operating at a 50% lower cost of finance relative to revenue than peers and with half the FTEs of their peers. And just like the world changed dramatically – and I’ll date myself when we went from accounting ledger paper, which I started my career as an intern, I will say, as a qualifier – and now we’re in the world of spreadsheets and ERP systems, and I think Gen AI is going to do the same thing analogous to that going from accounting paper to the spreadsheet ERP world, not only for peer companies, but even for those Digital World Class^® that are already operating at half the cost with half the staffing. Because again, the Gen AI capability when fully deployed in finance and other areas of the business is going to be breakthrough and step change because I know these Digital World Class^® companies have been chasing incremental improvement year in and year out as they’re challenged by leadership to reduce budgets and produce more with less.

But Gen AI is really going to be a step-change force multiplier, even for those highly competitive, very cost-effective in an impactful Digital World Class^® finance organization today. And that said, Jim, I think we talk about these major transformational initiatives – whether it was digital transformation over the last decade or now in the era of Gen AI. And I often talk about to get successful outcomes from these big technology and business transformational initiatives, I always talk about what I call long games and team sports – five- to 10-year journeys – having the right set of strategic partners that both help you figure out the vision, but also implement and execute those capabilities, leveraging

these technologies, specifically Gen AI. So in the context of Gen AI success being that team sport and long game, how should folks be thinking about where and how they partner for education and expertise? Because again, Gen AI is such a new technology category. How do people get smart about this stuff both by themselves and working with others, and where they do that?

Jim O’Connor:                  

Yeah, I love that colloquialism that you just shared, Shawn. That is true. I was actually, maybe another analogy I was talking with someone about the other day actually, I always say I learn from clients sometimes as much as they learn from me, and they used a phone analogy. So you used the ledger. But when I first got my iPhone – just to date myself a little bit – I used it to call people and later I learned there’s all kinds of other uses. I can use it for my home finance. I can track my kids and know where they’re at. I can monitor my sports teams, etc. And there’s all these other great uses. So to use that analogy, I think first people just got to start using it. And so most of our clients, they have a Microsoft Copilot or some equivalent of that where they’re starting to use it.

And so I think when you think about education, obviously you’ve got to learn yourself, you’ve got to work with your company, and we help people, I would say, three things – educate, ideate and execute. And when you think about education, I’ll highlight a couple points, and let me know if I’m going too far, Shawn. But first is there’s got to be three ways that this happens in terms of benefit. There’s going to be incremental, there’s going to be transformational and there’s going to be breakthrough. So some improvements may not be that exciting. Clients are automating some variance analysis. But you got to start to get used to it like I mentioned. And then there’s going to be transformational where you can totally change the game a bit and kind of do that step change. And then breakthrough is doing things in ways that we wouldn’t even imagine. And to do that, you’ve got to become familiar with what Gen AI is, which is why we’re going to do this series, and we publish research, and the other partners you should tap into as well – your vendors, others that you leverage.

But the other thing is how is this going to manifest itself into the market? And it’s really going to come in three ways. One, you’re going to see it come up, and many of you already have embedded in tools like Copilot that I mentioned, like SAP, like Oracle, like other point solutions. You’re also going to see people do a custom enterprise approach. And many of you may be doing that already in finance, but probably for sure in some of those other functions I talked about. And then there’s going to be domains where companies will offer you subscriptions to things like sales tax or tax knowledge. And so how you get started, I think first think about that phone analogy and think about what am I going to do? But then you got to really, when we work with our clients, it’s education first and then it’s ideation, and then you really need to put that plan together.

And hey, look, I know it’s not easy. Shortly before this, I saw a salesforce.com poll where over half of the people found it difficult to get what they want from AI, and over half of them don’t trust the data. So we’re still in a fairly formative

phase, but boy, there’s a lot of progress out there already that is absolutely impressive. And for those that haven’t started, I had many clients at the beginning of the year say, “Ah, I’m going to do a wait and see.” I think that’s changed. At the beginning of the year, we had 20% that said they were thinking about it. Now our latest poll, 25% are actually doing something about it. And anecdotally, my clients that have decided to wait are realizing that cost of inaction is pretty high, and so at minimum, they’re putting a plan together now for ‘25 and beyond.

Shawn Fitzgerald:           

Yeah, that’s really good stuff. I’ll color a little further. We haven’t published it yet, but we just finished this internal control study and one of the key data points was 81% of account-to-report organizations are evaluating Gen AI to supercharge their internal controls capabilities because quite frankly we know there’s a very strong correlation between automating centralized testing of internal controls and substantially reducing your audit fees per billion dollars of revenue if you’re able to get above 20% control automation. So significant real value impact, hard dollar accretive to earnings type savings by leveraging this technology, in this case in internal controls for the account-to-report area.

But going back to that, how do you educate and partner, and not to be a shill, but I think we’ve got this podcast series. This is the first of several sessions Jim and I and others are going to provide to our listening audience. I will say we’ve got an extensive Gen AI in finance report series that we have published. We continue to publish and add different additions to that. I think we’ve got three parts in the works for the balance of this year. And then obviously we’ll continue to write and research on this topic going into 2025 and beyond.

And again, we’ve got this really cool AI XPLR^™ tool at Hackett that helps companies look at areas where they can apply Gen AI technology where they maybe hadn’t thought about it, whether, to your point, Jim, is it incremental, transformational or breakthrough? So with that said, we’re kind of really coming to the end of today’s session, but I think it’s really important that you use these Hackett resources that I described to educate yourself and your leadership team both across finance and the larger enterprise. Start doing your homework by looking at some of that Hackett IP on the subject area. Schedule discussions with your advisor. Get that AI XPLR^™ demo.

And internally to kind of pull your own rope, I think you want to form or think about forming a Gen AI special projects team to lead your understanding of both the technologies and the opportunities as part of that strategy, effort, and construct. And most importantly, stay tuned for part two in our Gen AI finance series because as Jim said, the data is a big challenge. So in session two, it’s really about organizing the organization for Gen AI readiness, which means getting started – means getting your data, your processes, and your people in order for this transformational technology that is here – and companies are adopting and you may be adopting or have already adopted and look to scale in the future. So Jim, I’m going to leave it at that unless there’s anything else you want to say in closing.

Jim O’Connor:                  

I think you closed well. I just encourage everyone I know, it’s going to be a range of expertise on the phone. Stick with us through this whole series. For those of you that are just getting started, educate. For those of you that are starting to get a feel, work internally, like you said. Get a team together, tap into some of the other initiatives are going on and get moving here in finance.

Shawn Fitzgerald:           

Great advice, Jim, and thank you to you and thank you to all our listeners for joining us in today’s session.

Announcer:                       

Thanks for listening. If you liked this episode, please share it. You can find show notes, transcripts and related research at www.thehackettgroup.com/podcast. Subscribe on your favorite listening app so you never miss an episode. We welcome your feedback by rating this or any episode, or send us an email at podcast@thehackettgroup.com. The Hackett Group is a strategic consulting and executive advisory firm. Learn how we can architect your digital transformation journey, including Gen AI, at www.thehackettgroup.com.