Ep. 241: Dan DeGolier - Adapting to AI in Accounting

November 06, 2023 | 20 Minutes

Welcome to the Count Me In podcast with your host Adam Larson and special guest Dan DeGolier! In this episode, Adam and Dan, founder and CEO of Ascent CFO Solutions, dive into the fascinating world of AI and its application in the finance and accounting sectors. Discover how AI is enhancing efficiency and reducing errors, while also exploring the potential challenges and ethical considerations it presents. Join us as we explore the evolving landscape of AI in fractional leadership. Tune in now for an engaging discussion you won't want to miss!

Full Episode Transcript:
Adam:            Welcome back for another exciting episode of Count Me In. I'm your host, Adam Larson, and today we have a special guest joining us, Dan DeGolier. The founder and CEO of Ascent CFO Solutions. We start off by exploring current use cases of AI in the industry. Such as coding transactions and streamlining forecasting processes. 
But as Dan points out, we're only scratching the surface of what AI can do. The potential for growth and efficiency is immense. But it's important to proceed with caution and be aware of the biases and ethical considerations that come along with it. 
Throughout this episode we highlight the evolving role of finance and accounting professionals, in the age of AI, and how they can adapt to leverage its benefits. From bookkeepers, to CFOs, to fractional CFOs, AI has the power to enhance efficiency and transform the way we approach financial management. So grab your headphones, and join us as we uncover the exciting world of AI in accounting. Let's dive in.
< Music >
Well, Dan, we're so excited to have you on the podcast today, as we're going to talk about AI and fractional leadership. And just to get started, as we think about AI, how is it currently being applied to finance and accounting sectors? Obviously, it does things like enhance efficiency and reduce errors, but how is it being applied in those areas?
Dan:                Yes, thanks for having me on, Adam. It's a pleasure to meet you, pleasure to be here. I think we're just getting started, for one thing. AI, even though it's been around for a while, ChatGPT, GPT 4, and all those things, are relatively new to the mainstream. And, so, a lot of this stuff we're just starting to figure out right now. 
Definitely, in the accounting side, we're starting to see some use cases for coding transactions and things like that. I think there are a lot of opportunities in our world, in the finance realm. When it comes to forecasting, to be able to streamline multiple scenarios and make iterations to financial models and forecasts. 
I think that's an area that we're starting to see develop. And, then, things like pricing strategy and looking at different ways to price and run different scenarios around that. Using large language models, and data, and being able to bring in data and run multiple scenarios and see what things look like there. I think those are all some areas that we're starting to see. 
But, honestly, because it's so early, what is really going to be the biggest use cases, two years from now, is probably something we haven't thought of. Or somebody's thought of but hasn't really been implemented, yet.
Adam:            Yes, that's a great point, that we're so early in the generative AI phase that some organizations are adapting quickly, other ones aren't. And software companies are trying to integrate it into there but it's still in the early phases. So our traditional role-
Dan:                And it's still prone to errors as well.
Adam:            Exactly.
Dan:                Yes, we've all read the articles about the lawyer who tried to use it for briefs and got in huge trouble, and the hallucinations are still rampant. So I think proceed with caution, but recognize that it has enormous potential and don't be left behind. 
I was going to say, I've heard that it's been compared to if you look at Web 1.0, the emergence of the Internet, and commercial use, that this could be a 10x-type of opportunity. From a growth potential, from an efficiency potential, et cetera, it's just fascinating to me, just how massive this could be, and how life-changing this is.
Adam:            Well, and also the bias that's implicit in there, in the AI. Because there are so many biases among how people think, wording, that's out there in the Internet and how it's learning. There's going to be that bias that you have to get over as well. Because it's going to be embedded in there because of how it is societally.
Dan:                Correct, yes, I agree with that. I think one other ethical consideration that needs to be taken into account, when you're implementing AI, is things around copyright infringement, and intellectual property, and protection there. I think the chatbots aren't necessarily aware of what's IP and protected and what's not. 
And, so, it's important that we take into that, that there's a human overseeing that, and making sure that there's nothing being taken out of context or being utilized improperly. And along the same lines, research is another area. Tax research and other types of accounting research is a place where there is a lot of use cases for AI. 
But, again, this is where you need to be very careful around trusting that research and validating that it is accurate. So we don't end up in a situation, where something that's not valid is being utilized.
Adam:            It's going to be very difficult to understand what has been verified and what hasn't, and as you're doing research and as you're looking at things online. I imagine new tools are going to have to be developed to verify, "Yes, this is valid." Or "No, it's not." And how do you trust those as you go forward?
Dan:                Yes, that's really important, and there are going to be mistakes made. As we start to adopt this, we're going to see mistakes being made. And, as humans, we need to learn from our mistakes and learn from others' mistakes, that's how we evolve.
Adam:            Mh-hmm. Do you think that the traditional roles in finance and accounting are going to change because of these? I mean, obviously, they are. But how can we adapt as we go forward?
Dan:                Yes, I think, first thing I would suggest is pay attention to what's going on, see what's evolving, see where things are taking it. I think it's going to definitely change the accounting side, the day-to-day transactional stuff. There's a YouTuber out there, Hector Garcia, who has done some demos of how you can plug in a ChatGPT tool into QuickBooks Online, and how that can help ease the coding of transactions and things like that. So it's definitely going to change that bookkeeper and junior accountant role significantly, I think it'll change all aspects.
The CFO's desk, it's going to still require somebody with experience, and knowledge, and understanding, to validate what's coming out of it. Just like in any other industry, there's a lot of need to confirm, and double-check, and be heavily involved at that strategic level. But I think it'll make us more efficient.
Adam:            Yes, I definitely agree with that. And as you're talking about things like analysis and looking at it from that higher level. I mean, obviously, the AI has a better computing power, but we still need that human element. And how does that traditional human analysis going to affect, as we look at the output from the AI?
Dan:                Yes, that it's still going to be critical. Machines are going to do a lot of the analysis and it'll find pattern. It's better at pattern recognition than us, especially. with large data sets. But when it comes back to that human element of truly understanding, and the uniqueness of certain things, it's going to require a human element. 
In preparation for this call, I was thinking a lot about fraud detection, and you got large data sets out there. I think, again, back to pattern recognition, AI can be really good at identifying things that stand out and look unusual. I mean, if you think about, maybe, purchase orders or sales orders that look unusual. 
Maybe have overrides from managers and they can look for patterns there, where particular users, within an accounting system or ERP system, might see that something that a particular manager might tend to override things more often. Or looking at addresses, and zip codes, and understanding if there might be some inappropriate payments made that match up to addresses, vendors matchup to employee addresses or things like that. So that could be bogus, that could be fraudulent. 
I think those things are going to be a huge area for auditors, both, internal and external auditors starting to use those data sets. Where that AI tool can go in and start digging around and finding some unusual patterns. 
Adam:            Yes, and thinking about implementing AI, within your organization, if you're really considering this, you've done all the research. What are some challenges or ethical considerations that should be addressed, when implementing it?
Dan:                The first thing that comes to mind is security. Right now, I've been reading some things that we're trying to be able to bring it inside your intranet, bring in those tools inside your internet. But you don't want to have breaches of data, things that go out, where the chatbot is getting a hold of your corporate data and then utilizing that in the greater universe. And, so, that's going to be really critical, is that we solve for security concerns where things stay within the four walls very clearly. That's the first thing that comes to mind. 
And I think the other one is touched on earlier, which is just trusting it too much and seeing that something that comes out of it is just trustworthy, as opposed to really validating it. Whether that's research around case law, when it comes to tax law, or whether it has to do with... Just what comes out of a financial model, and what's practical from a pipeline perspective and things like that, when you're forecasting your financials.
Adam:            Yes, so as we look to the future, when it comes to AI. What are some of the breakthroughs that you think will happen within the finance and accounting industry, as we look to the future with AI?
Dan:                Automation, in general, and that can take multiple forms. We touched on the accounting coding of transactions and things like that, I think that's a big part of it. There can be a lot more automation around all of the accounting cycles. Whether it be payroll, invoicing, accounts payable, there can be a tremendous amount of automation on that side. 
Variance analysis when it comes to your soft close of the books, your initial review of a month-end close. I think there can definitely be an analysis and digging in a transaction, and looking for those variances to prior periods variances, to budget variances, to forecast, and pulling those out. So I think there can be some automation around that. And, then, again, on the financial modeling piece, the forecast piece, there will be automation there as well.
Adam:            So one area of expertise that you have a lot of expertise in, is the fractional leadership, the fractional executive, and especially the fractional CFO. And as we're talking about AI and the changing of how that CFO looks. How do you see the ability to have this AI as a fractional CFO? How does that really enhance your ability to help the organizations, that you're within that fractional capacity?
Dan:                Yes, well at our firm, we're technology first, and we've always been focused on automation where we can. So I think for us, it's going to be those same types of approaches. Where we find ways to be more efficient, to be more cost-effective, to really implement these tools. Identify the best use cases for these tools, kind of trust but verify. Make sure that you still got that adult supervision, with that AI tool. But really leaning into it and making it a tool that speeds up data for the C-suite. 
The faster you can close your books, the faster you can update your model, the faster you can make adjustments. When you see something change with your pipeline, I think, more agile executive team can act.
Adam:            So when you're coming in as a fractional executive, a lot of times the best place for that model is an organization in transition. And, so, that's what I've been reading when I've had other conversations. It seems like it's organizations that are in transition, and when you're in that transition, it seems like you would be looking at all your systems. But how do you come in and say, "Hey, I want to have this technology first and utilize these tools." But they have never used those before. How do you bridge that gap?
Dan:                Yes, it's an incremental process. I mean, when we look at working with a company, they are often going through a transition. Maybe, they're looking to raise an additional round of capital. They've recently raised another round of capital. They've got a new board reporting requirements. They need better discipline when it comes to forecasting their cash flow. 
So if they're a little behind the 8-ball, when it comes to technology, it's going to be incremental steps. You first have to get a really solid ERP, or accounting system in place that is trustworthy and fully GAAP. Whether they're audited or not, you want them to be fully on accrual GAAP basis. 
Once you have that, then, you start to put in place those data visualization tools. That's something we've been leaning into really heavily the last year or two, is creating really robust dashboards and data visualization, that not only show your historical financials, but your forecast, and your HR, and your payroll, and your sales pipeline. And, so, those technologies first need to have really reliable actuals, before you can lean heavily into some of the other newer technologies, and more robust technologies.
Adam:            That makes me think of how important it is to have good data. Because you don't want to have garbage in, then, it'll just be garbage out. So you have to really make sure your data is in a good spot.
Dan:                You don't even want to start to forecast or implement those better tools until your historicals are accurate, for sure. And it's not just plain GAAP financials, it's also what your KPIs look like. What are the real drivers of your business? And that's one of the things we look at when we come into a new client, is really take the time to look at the true drivers of the business. 
They may not be obvious at first, every company is a little bit different. What's driving their growth, and their revenue, and their cash flow. So we really lean into that. And, so, we'll often start with what we call an assessment phase. We'll spend 20 to 40 hours just really digging in deep, to understand every component of the business.
Adam:            Do you think that all businesses would benefit from some a fractional executive coming in and relooking at things? A lot of times people bring in consultants to do that. But it's just like they look at everything, give you a PowerPoint, and head out the door. But that fractional seems to be like that person who partners with you for a period of time.
Dan:                Yes, definitely, our model is based on long-term but part-time. So we're, generally, looking at companies in the SMB market. So, generally, we work with companies between 2 million and 100 million in revenue. 
And, so, as that company scales, some companies are too quick to hire a full-time CFO. Where they might really just need a full-time controller, a really solid controller, and an accounting team. And, then, a fractional CFO for a couple of days a week, who's extremely well qualified and very experienced, could be a great fit for them. To bring in that true executive-level oversight, with decades of experience, to help them navigate, again, what those critical KPIs are. Where the holes are and the different strategies that are being considered and things like that. 
So, yes, companies that are worth 75 or 100 million very likely to have a full-time CFO, a very qualified CFO. But companies under 75 million or depending upon the transaction complexity, and transaction volume, companies that are small and medium-sized businesses can really benefit from having a top-notch CFO on their team. But it may not need to be a 40-hour or 60-hour-a-week job. It can maybe be a 20-hour-a-week type of engagement.
Adam:            So you get that full-time experience, that experienced person there. But you may not, necessarily, be able to afford the salary that would require to have that person on full-time.
Dan:                Yes, and there may not be enough, truly strategic, CFO-level work for that person, that you need to have someone on a full-time basis. That's what our whole model is based on. As you grow and evolve, you get the resources you need on a fractional basis. 
Our team is CFOs, and VPs of Finance, and controllers, accounting managers, financial analysts, senior accountants. So we've got a full stack of people with different levels of experience, who can come in and support a company during its growth phases, and you pay for what you need. As opposed to having a real heavy fixed cost on your G&A budget, G&A financials.
Adam:            Yes, that seems like a really good benefit, especially, for the small to medium-sized businesses. Is it beneficial for a startup? If a startup is just getting going; as an entrepreneur, is it good to bring in fractional folks, or do you think full-time would be more beneficial?
Dan:                Yes, fractional makes a lot of sense for an early-stage company. I look at as a step function. You start out maybe you just need a part-time accountant to make sure things are being coded properly. Once you have revenue and you're ready to raise around the capital, then, you probably want a strategic fractional CFO or VP of finance, who can help you with that capital fundraise, help you with a really robust financial model, and understanding what your KPIs and drivers are. 
And, then, over time, you start to fill in some of those roles on a full-time basis, as you get a growth cycle. So it's not uncommon, maybe, you start with a fractional senior accountant and a little bit of oversight, from a fractional controller. And then that evolves into one or two full-time accountants and, then, a fractional CFO, and, then, eventually, you get a full-time controller, and it just builds as you go up the ladder in revenue, and fundraising.
Adam:            So this does not have to do with fractional CFO. But I want to throw this question out there and you feel free to answer it or not. But do you think that the evolution of AI will help bridge the gap between US GAAP and IFRS, to make it a more international accounting standard?
Dan:                I think it definitely has potential to. I think that there's logic in the way that things like RevRec and other things are being handled, between IFRS and US GAAP. So, I think, there's definitely some good potential there.
Adam:            Yes, I don't know. Because I just feel like as we become a more global world and how we would do business and everything. It would make more sense to have a globally recognized accounting standard, so that everybody's doing the same, has the same standards that they live up to. Obviously different countries have different beliefs and stuff like that, but it would make sense for us to think globally.
Dan:                Yes, I like that. I hadn't given that a lot of thought before, but that does make a lot of sense to me.
Adam:            Mh-hmm, well, Dan, I want to thank you so much for coming on the podcast. It's been great talking with you. Thanks so much for sharing your knowledge and expertise with our audience.
Dan:                My pleasure Adam. Really it was fun to meet you and fun to discuss these emerging technologies with you.
< Outro >
Announcer:    This has been Count Me In, IMA's podcast, providing you with the latest perspectives of thought leaders, from the accounting and finance profession. If you like what you heard and you'd like to be counted in for more relevant accounting in finance education, visit IMA's website at www.imainet.org.