GL coding in the Chart of Accounts(COA)

Find out interesting insights with Jon Naseath, CEO/Founder Cantu Capital Inc

Moderated by Sherry, Financial Technology Consultant at Hyperbots

Don’t want to watch a video? Read the interview transcript below.

Sherry: Hello, and welcome to all our viewers on CFO  Insight. I am Sherry, a financial technology consultant at Hyperbots, and today we are speaking with Jon Naseath, who is an accomplished executive with expertise in AI, machine learning, and computer vision driving impactful technology solutions in education, healthcare, and business. Thank you so much for joining us today, John. Today we’ll be discussing the importance of GL coding in the chart of accounts, common mistakes companies make, and how technology can help maintain coding integrity. So let’s dive right in. Can you briefly explain what GL coding is, and why it is so important in the chart of accounts?

Jon Naseath: So just briefly, what GL coding is, and I’ll tell you an example of the importance. GL throughout the entire business. People are doing their work. They’re spending money, earning revenue, buying things, or investing in product development. Whatever you’re doing throughout the entire business. That translates into equity. You’re making money for investors or making customers happy. Whatever is your role, there’s a representation of that in the GL. We have to make sure that what you’re doing, what you’re spending, and what revenue you’re making is accurately represented in what’s going to turn out later to be the P&L or your balance sheet. You know, what did we invest in on the balance sheet? And what did we spend, and how much revenue did we make on the P&L? If we don’t map things correctly to the GL, then the rolled-up reporting numbers will be wrong. There are all these very formal GAP accounting rules over how things exactly have to be mapped. The story example I wanted to raise real quick was just a few hours ago. I got an email from a very good friend of mine. I used to work at a large organization, and he’s a leader in that organization. He spends his whole day talking to big companies around the world about how to reduce the cost of IT and cloud spending, and how to help them with strategy. He pinged me and asked me, “Where can I find the right definition of revenue?” I chuckled when I saw that because, I mean, there are people’s whole careers spent helping companies figure out how to define revenue for their business. He asked, “What’s the definition of net revenue?” I spent a year working with a company where there were actually seven definitions of revenue just within that company. They called it gross revenue, sales revenue, RevOps, net revenue, and all these different things. But the reality is, if you ask an accounting person, there’s only one definition of revenue, and there are actual policies around how that’s defined. The reason I bring it up is, that if you’ve mapped all these different things that you do in your business to the GL, and then you map that GL to management reporting, different management reporting people will want to see how what they’re doing impacts revenue in this example. They can make up their definitions, but they should define that there’s one thing that is accounting GAP, you know, audited revenue. All the other stuff is management reporting. If you’ve mapped stuff in wrong, then the core number is wrong. If you map stuff wrong, then the other pieces are wrong. Just bluntly, if you do it intentionally and you’re trying to hide things, people go to jail. So that’s why it matters. But also so investors can understand what they’re doing, and management can have clear views of how they’re managing the business. Long answer to a short question, but it was a pretty loaded question.

Sherry: And in your opinion, what are some common coding schemes that companies should follow when setting up the chart of accounts? Can you provide examples from different industries?

Jon Naseath: Most simply, think about it as your P&L and your balance sheet. As you walk down your P&L, you know, revenue, cost of goods, operating expenses, whatever you have in your P&L. As you walk down your balance sheet, assets, liabilities, etc. All it is is a number that represents walking down your P&L and balance sheet.  The root of your question, though, is within different industries, there’s become quasi-standards around how they’re doing their business. Every company within that industry is also going to try to find its competitive advantage, so they’ll do something unique to them. When they do roll up their P&L and balance sheet, I remember I was in a job where my job was to take that from the accounting team, roll it up into these investor analyst reports, and then we’d hit submit to go live at quarter-end. There was an army of like 60 investor analysts whose whole job seemed to be to find any errors we had in any of those numbers. What they’re doing is mapping that financial statement to other ones in the same industry and seeing if we’re different, wanting to compare them. So to some degree, you want to be similar to your peers in the industry, but for others, you want to innovate. I was at Equinix, which is an innovator in space. We impacted how the industry looks at metrics like FFO and other REIT-related revenues. We made sure we were compliant with revenue. There were lots of things besides just data center buildings that we had as revenue, and we had to account for them correctly.

Sherry: What are some common mistakes you’ve seen companies make with their GL coding structures? Can you provide examples from various industries?

Jon Naseath: One mistake is in the way companies add dimensions to the GL, like region, department, or sales channel. You might see the initial chart of the account code, and then different dimensions get added as “dash something else.” This lets you slice and dice to understand costs by department, region, or product type. But bloating the GL by adding too much detail, like putting all your product SKUs into the chart of accounts, can make it unmanageable. Over time, old codes and new codes can create complications. Another issue is the lack of standardization. It’s important to align with industry standards so reporting can be compared. Insufficient detail is another problem, where management wants specific insights to reduce costs or invest more, but all they have is a generic code for product costs. You need to break things up by what should be capitalized and what should be expensed. Finally, there’s inadequate training and documentation. If people aren’t trained well, they can tag transactions incorrectly, which impacts the rolled-up reporting. That’s why visibility and proper training are key.

Sherry: How do these mistakes impact a company’s financial management and reporting?

Jon Naseath: From a financial perspective, it creates a lot of unnecessary work. Ideally, you could just roll things up and have it reconciled, and everything makes sense. But often, you do plan vs. actuals or month-over-month, and something’s off. You might have a gut feeling that a number isn’t right, and sure enough, you unpack it, find a miscode, and need to reclassify. The impact includes inaccurate or inefficient financial statements, upset executives, and hours of rework. It can lead to compliance risks, resource wastage, and worse, damage a company’s credibility. If a CFO has too many reporting errors, they could lose their job.

Sherry: How do you think technology can help maintain GL coding integrity and reduce these mistakes?

Jon Naseath: Technology is excellent at reconciling across different dimensions and sources. It can tie everything together, but it’s not perfect—AI can sometimes hallucinate. There are automation tools and AI that help, but there needs to be a balance between understanding numbers and producing accurate outputs. For example, I was talking to the controller of a large global organization. They have a complex ERP system and are transitioning to a new version. They’ve gone through the business planning, and they know what they want that future state GL and reporting metrics to look like. And they finish that. But it’s gonna be another at least a year and a half, maybe 2 or 3 years, until they get this full ERP fully implemented, all trained, and are in the new future state system. In the meantime, they’ve got a year and a half and 2 years or more, because things always go wrong in those projects. There’s always some reporting that’s missing, even if they say they go live. It’s never right at first. However, even with what I just said, it’s never right at first. I believe there’s an opportunity here where if you define what you want your future state reporting to look like, and you have that data coming in. And you’ve created this mapping thing that you’re gonna give off to some developers and they’re gonna rebuild the system based on that new mapping tables and then you have to extract the data from the old table, load it into the new table once it’s developed and then wait. Maybe the reports work, but they don’t. And it’s huge UAT testing.  Anyone who’s been through that knows it’s painful. I think AI can do a lot of that stuff. I think you can take from your legacy system, your current system. You can say, here’s what I want my master tables to look like. Here’s what I want my reporting outputs to look like and it can help produce those. Now, again, it’s going to hallucinate. You have to code it the right way to make sure that it gives you the right outputs. But a lot of the pain, which is real pain, or a lot of the late nights because there’s an error, and there’s rework. You have to go through a lot of the cost of hiring an army of people to fix an error that was in there historically and rebuilding reporting.  A lot of that, I believe, will be able to be fixed by AI. And it’s no longer for me just a belief. I know it’s real, because I’m seeing it happening in different companies I’m talking to or working with and it’s fun. And frankly, I’ll just give you guys a shout-out. You’re on your track with the products you guys develop. I’m seeing good things. It’s exciting to see what you’re building, and where this will lead to.

Sherry: Thank you so much, Jon, and from your experience in the finance industry, can you provide an example of how automation might improve GL coding practices in a specific industry?

Jon Naseath: Yeah. The manufacturing industry is complex a bit, because it’s not all just kind of in the cloud, SaaS, and I’ll say relatively easy. So there are so many different stages of raw materials, work in progress, finished goods, making sure you’re coding things all to the right place and then, making sure it’s current. I think that speed aspect is really important, because then, if you don’t get it right in time, you’re making accruals, and you have to fix them later. So I think that can reduce a lot of human error and complexities when things go away. And I think automation will fix a lot of that stuff.

Sherry: And what best practices would you recommend for companies, or for our viewers looking to implement or improve their GL coding system?

Jon Naseath: Sure. Don’t get overloaded is the way I like to describe it in your GL. Keep it relatively simple. Think about what is the core dimension of your GL, and then what the sub-dimensions, and sub-ledgers that tie into that. Make sure people have the training they need so that they’re not screwing your stuff up. I remember a friend of mine was in accounts payable, and he had a paper on his screen. He would just keep, and all he was doing was coding things to those key numbers, and that master table. When I come into a new company, I’ll, based on that, go around and ask the people in these types of roles: “Show me your kind of cheat sheet. What’s that master thing for mapping that you rely on?” And they all have them. They all pull out, “Well, this is what I look at, this is what I rely on.”I like to take those cheat sheets, standardize them into policy, and make them real. Use technology wherever possible. I do think AI is great, but I think that there’s lots and lots, you know, throughout my whole career there’s been automation of things. So there are lots of things that are proven as technology automation, use them. We don’t need to reinvent AI just to do something that’s already fully automated. And then use AI for stuff that couldn’t have been automated previously. That is now enabled. And the combination of those two things is where you get really powerful results and then just basic, I’ll call them controls. Reconciliations, making sure that things are coded correctly, doing budgeting, and making sure you’re doing those plans versus actuals. The best call out here is to partner with FP&A, finance, and accounting. Accounting wants to get their numbers right. They’re very proud of that. But then it’s FP&A that is creating the management reporting a lot of times and a lot of the forecasting. So if those numbers are off, work together to make sure they’re right before you go spread it all over the business and tell them that they’re idiots because they screwed up some number when it was an accounting-finance disconnect. That never happens, but just hypothetically.

Sherry: Thank you so much, Jon, for these valuable insights on GL coding practices, and how technology can play a crucial role in maintaining financial integrity.

Jon Naseath: My pleasure, always fun.