Ep 52: Foundations Before Acceleration - a Planning Aces Episode

Speaker A: Hello and welcome to Planning Aces, where finance leaders share how modern FP&A is becoming a true performance driver. In this episode, we feature insights and commentary from three accomplished CFOs who are rethinking how planning fuels growth, sharpens capital allocation and strengthens decision velocity across uh industries. Our Planning Aces discuss how forecasting, discipline, operational visibility and business partnership are reshaping the finance agenda. The message is consistent. Planning must be continuous, dynamic and embedded in strategy. I'm, um, Jack Sweeney, your co host, Joe, joined as always by author and finance strategist Glenn Hopper, our resident thought leader. Together we highlight the takeaways from this episode. Glenn joins us after this message from our sponsor. Ready to plan confidently, close faster and report more accurately. Here's what sets Planful apart. Uh, imagine purpose built applications for, for every department from FP and A to accounting, marketing to hr. Ah, all with built in financial intelligence. It's easy to implement Planful in just weeks and with minimal IT involvement. So you can rapidly and seamlessly engage everyone across the business on your key financial progress. Best of all, Planful grows with you. Its unparalleled scalability ensures that no matter how fast your business expands, Planful can keep up. See why over 1500 customers worldwide choose Planful as their flexible, user friendly end to end financial performance management platform. Go to planful.com CFO Thought Leader to see how you can reach peak financial performance with Planful. I'm, um, joined as always by our resident thought leader, finance author, former CFO and operators operator, Glenn Hopper.

Speaker B: Glenn, how are you and where are you?

Speaker A: I'm doing great by the way.

Speaker B: Uh, so the good news for me the last couple of weeks. Well, that's not true. I was going to say I haven't traveled much. I did take two trips to Arizona for different clients, but if you've got to go somewhere in January, Arizona is the time to go. But I have been home the past couple of weeks, so uh, I'm excited to be, uh, back to Planning Aces with you.

Speaker A: So I mentioned before we got on that I thought we had three strong cfo, uh, uh, Planning Aces this week. And you quickly agreed. But tell us something, Glenn. What, what jumped out immediately, uh, to you as you heard the conversations beforehand.

Speaker B: Uh, first off, these three guys probably don't need to read my upcoming book because it sounds like they're, they're running the playbook by it. Seriously though, it's um, what hit me was how deliberate all three of these CFOs are. They're not rushing to Deploy. They're not chasing after some, you know, AI magic wand fairy dust that you're going to sprinkle on your processes. But what I love about the clips you pulled is they're each doing something that I think the best finance leaders do instinctively and they, they touch on each of them. So Kevin, building a centralized governance function. And I, I've got chapters in the book dedicated to that and that is exactly the right approach. Um, Bruce taken a year to redesign business processes before even picking a platform. And I just, I felt like preach, brother, preach. And, and um, Razak warning people against bolting on seven point solutions and calling that a strategy. And you do see that a lot of times where if you don't have a centralized approach to how you're going to roll out AI, it ends up just being chaos. And so with these three, I think we saw three different companies, three different stages, but that same operating discipline. And that's really, that's the FPA mindset through and through. So yeah, they were, I was loving every, every bit of what they said.

Speaker A: Yeah, the interviews are getting interesting because the CFOs, uh, clearly finally they've given it a lot of turn back the clock 12 months and it was so fluid and they weren't really coming forward with any position. These guys all have sort of a position in my way of thinking anyway. And it was less about tools, which is uh, also interesting. But uh, let's tee up our first, uh, planning ace, who will be Kevin Rubin. Um, and he's CFO of zscaler. Kevin has led finance through multiple transformation chapters and here he frames AI not as a breakthrough moment, but as a budgeting and prioritization discipline. Again, this is Kevin Rubin.

Speaker C: I completely agree with um, the notion that AI is an experimentation phase. Right. You're going to start with a use case, you're going to test that use case. Um, you're going to build conviction around a use case and then you're going to deploy. Um, AI is still early and new. There's not a lot of mature technologies that you can just uh, drop in and say that I'm leveraging a Gentec technology in this particular workflow. So there is an iterative process that exists and in order to do that you've got to find room, uh, in your budget to be able to support that. So we spent a lot of time uh, going through our current fiscal, uh, budgeting process to be able to prioritize investments in AI. Uh, I think one of the decisions we made that works well is that there is a central focus for how we think about, uh, AI and how we think about, uh, deploying it through the organization. So, um, that's intended to prevent a lot of, uh, technology sprawl and duplicate technologies providing similar outcomes. And we really do stress test if somebody presents a new AI oriented technology. Do we already have something in house that we have tested, certified and deployed that can satisfy a particular use case foreign.

Speaker A: So AI discipline beats AI speed. In the words of, uh, Kevin, I believe Kevin Rubin of Zscaler, he talks about AI as an experimentation phase, testing, building conviction, then deploying. Glenn, what does that signal about how FP&A should engage AI though right now?

Speaker B: Yeah. So I want to go a step further on the experimentation phase because I did appreciate that. And I think one, one way that we could look at investing in AI is not putting all of our eggs in one basket. Think of AI Projects as a, uh, portfolio. So you have 25 investments, 25 stocks, whatever in your portfolio. And you know, you're trying to hedge and balance out the portfolio so, so that you, you can have wins. And I think that just as, as he was talking, I thought about that portfolio approach. Um, but it's really, he's talking about treating AI investment the same way a good CFO treats any capital allocation decision. So as, as you were discussing with fpa, it's the same thing. You, you test, you validate, and then you scale so that you're not, you know, trying to eat the elephant all in one bite. And I think he's doing something specific at Zscaler that more companies need to copy. Um, I loved how they created a centralized AI function that acts as a filter. So somebody pitches a new tool internally. The team stress tests whether Zscaler already has something deployed. Because if you have multiple solutions going at one project, it just creates chaos and confusion. So validate it and then say, okay, we need this or we don't. And then, you know, that covers the use case and that's really how you prevent the tool sprawl problem. Um, and you know, listening to the rest of the podcast, keep in mind, Kevin led Alteryx through its ipo, scaled their ARR to a billion dollars or whatever. And he's been, so he's been in these hyper growth environments where ungoverned tech spending can spiral really fast. So when he says the landscape's immature and there's no plug and play sol, that's uh, that's not, uh, him just guessing, that's caution. Uh, because he's, he's been down that road before. And that's really the, I guess you'd say the pattern recognition from a CFO who's seen what happens if you skip that governance step. So great, great insights from him there.

Speaker A: I love that. Avoiding the uh, the tool sprawl, uh, which many organizations are just catching on that this could be a real problem. So our next planning ace. All right, let me tee him up. Bruce Schumann, CFO of Universal Technical Institute. Bruce is candid about being early in the AI journey and very intentional about why, uh, Bruce came out of Intel. Great episode where he looked back in time at his intel years. Uh, but it's interesting. Each finance, uh, leader clearly their past experience shapes how they think a little bit about uh, the AI world as well. So.

Speaker D: The, the goal for this year, Jack, and I'll be transparent with you, maybe other companies are much further down the path than we are, but we have two divisions and we're on two different ERP systems. So it's really, we really want to get to a place where we have kind of an AI driven, driver based forecasting model for the company. Uh, we're not quite there yet. Because of those two ERP systems, we had to make some pretty critical foundational data investments first. Having a really strong data foundation, good data governance, having a data lake, for lack of a better way of saying it, of three to five years of really solid metrics across every aspect of the business. We had to get that done first before we just throw an AI model at it because it's kind of garbage in, garbage out. So that was the investment we made last year and into this year. And we hope one of our goals for 26 is to have a true AI driven, uh, forecasting model to kind of help FP&A. The same way that accounting has been helped in some of the reconciliation work that AI has helped us with the erp.

Speaker A: The two ERP systems that you mentioned, uh, we would imagine, okay, obviously you want to get it down to one ERP system in the future, but at the same time, uh, was AI part of the consideration? As you go forward, you're looking at both these. Which ones are really adding value with AI? What's a practical approach with AI?

Speaker B: Absolutely, yeah.

Speaker D: In fact, we've chosen one. I'll kind of, I'll maybe not say the name because I'm trying to do an advertisement for an erp, but the platform we chose, um, how, how integrated AI was it in the platform and opportunities that's going to unlock for us over the next five, 10 years. That was a front and center consideration for us. We have a, a wonderful CIO who just joined us about, about the same time I, I joined and she and I are very closely partnered on this and that's a forefront consideration for us and how we chose our erp.

Speaker A: Could you share, share with us a little bit about, um, you know, was finance and that CIO and yourself naturally. But then you had maybe a member or two from your finance team also tasked with trying to evaluate and understand better. Here's why the system might be of value to us more. Um, yeah. Are these FP and A professionals? Who are these? Or they just sort of AI champions on your team? Who.

Speaker D: It was a mix of FP and A and accounting professionals kind of working with our data analytics team and our cio. It's kind of a partnership of those four, uh, folks working together. Honestly Jack, it took us about a year. We were very thoughtful about this. We really started with the business process redesign first. What did we want for the next five, ten years? Let's just not go choose the, the shiniest system or the best sales pitch we got and okay, we'll just go with it. We really spent about a year going through the evaluation process on the business process redesign first. That was really important for us and I think that yielded a great outcome. We're really excited. 26 is going to be a year of kind of consolidating up to one year erp, uh, system for the company.

Speaker A: I guess we're not surprised by this, but at the same time he makes a really strong case for pausing AI and ambitions until data governance and ERP foundations are in place. Glenn, the right path, given, just given, you know, hey, time is of the essence here as, uh, AI, you know, disrupts the world.

Speaker B: Yeah. So we're all under pressure. We're all having this existential crisis around AI and what it means for our departments, our companies, our jobs and all that. But what he called being behind is really, I mean maybe, maybe they are behind on some people who are really out there on the leading edge. But long term, I think that this is the right approach. Uh, they're running two divisions on two separate ERP systems right now. Before he can build AI driven driver based forecasting model and all the ambitions they have, they have to clean the data across every part of the business. So they took a beat in 2025, invested in data governance, stood up their data lake, accumulated what he said, three to five years of reliable metrics. And it is framing, I think was Blunt, incorrect. An AI model that's trained on bad data is just giving you garbage faster. So you have to have that data foundation layer. And, um, you know, we talked about before, that's a concept I cover extensively in my books. And the AI will amplify whatever foundation it sits on. If the data is clean, processes are sound, then great AI accelerates good decision making. But if the foundation is shaky, then suddenly you're just automating errors at scale.

Speaker A: It's funny, sometimes I'm phishing, and that's sort of what I was doing there, where I ask him, you know, tell us who's coming together here. You mentioned your CIO, but were there FP&A people sitting in? And again, I want him to challenge and say, no, not really. Uh, uh, just to get a clear idea how organizations are making this rather collaborative adoption and, uh, uh, thinking around AI happen. I mean, in your experience, uh, who's at the table?

Speaker B: Yeah. So, Jack, I gotta tell you, you are timely with your selection of questions and thoughts here, because this is crazy. Just this morning, I do a lot of work with a large, uh, EPM provider. Um, and again, not trying to do commercials for any particular software, I was talking to them just this morning about how important the CFO CIO relationship is. And what I love is the way Bruce approached this, where he and the CIO joined. I guess he said they joined around the same time at uti, and then they partnered on the ERP selection process. And I've got to tell you, as a cfo, I was at a company once and I wasn't there for a lot longer beyond this. We were doing an ERP implementation. And the CIO said, I'm not participating in this because this is a finance platform, not a business platform. And I knew at that moment that we were doomed. And so. And right now, I see it every day, the CFO and the CIO have to be together because, um, as a consultant, when I go out and talk to companies, I'm usually, because I speak the language, I'm coming in through the office of the cfo, and we'll go all the way down the line with the cfo. If they're not coordinating with the CIO as it goes, then the CIO is going to come in at the end and say, whoa, whoa, whoa, wait a minute. You can't do this because xyz. So I loved hearing about that relationship, and I love. This is another big one, that they didn't start by evaluating vendors. They spent a full year designing those Business processes first asking what UTI would need for the next five to 10 years. And that's like whether you're selecting an AI tool or an ERP, whatever you're selecting, the wrong approach is just to, to pick the platform first and then say here we're going to slap this brand new shiny ERP on top of it. It's going to fix all our problems. Well you haven't even identified your problems. So um, yeah so once that work was done then you look at designing the platforms and AI integration capability and he said was a primary selection criterion. So the sequence really is processes, understand those processes, then figure out your platform to do those processes and then you can put the AI on top. So that is all uh, the perfect right framing and thinking and you can see how these uh, how these folks got to the CFO role because uh, that's the perfect approach to it.

Speaker A: Yeah. These uh, established ERP players, it's interesting to see what they think should be the AI, uh enhancement to their platform. Whether they succeed in identifying the right, the right AI enhancement. Spent all this time developing something that discover you know, finance departments aren't cool on this. It's not something they, they really felt like they need. It's not a must have. Uh, our third planning ace is Rizak Jalo, CFO of floqast. Rizak uh, brings a platform first lens shaped by experience and he does have a really interesting background at Adobe when they had, when they first developed this subscription model. Apple and Looker are on his resume. Great interview with Rizzak a little while back. Here's, here's our clinic.

Speaker E: Very different answers for different, different size of companies. Right. So if you are a larger enterprise company or even just a public company, you've got a lot of your processes figured out through that process of ipo and so you're in a good spot to really embrace a lot of automation. It's finding, finding the right platform to build around. I would caution finance leaders from going out and cherry picking seven different small tools that do one little thing in AI automation because you're going to run into a lot of issues there. When they, if you ever want them to talk to each other, they're not going, there's no orchestration layer. They're not going to talk. Well, you're going to have seven different levels of auditability and you're not going to have a platform to people to build around. Every time you have a new problem solved, you're going to be looking at what's a new tool and it's going to get spaghetti AI pretty quickly. Um, and so if that's the case I'd really look for around two or three platforms to build and so something that your team can say, hey everybody, this is a new skill set. You're an accountant. The old new skill set was learning how to use Excel from calculators. The new skill set is learning how to use AI as an accountant and invest. Here we have a platform, we have a tool. You can invest your time in learning how to use it and it's going to pay off both in your day to day work and your long term career. For smaller companies they might not have all those processes. So looking for a solution that can one help figure out your workflows and processes and then uh, easily change from. I've got this checklist of workflows, I can now look at each one, click this button and then start using AI to automate each one of these steps in my close process. I think that's really powerful for teams that a lot of smaller teams aren't going to have like a giant company wide 2 year digital transformation process where you've got these external consultants coming in and it's years of work. They're going to do things incrementally and so having the right basis to do those incremental improvements can be really helpful.

Speaker A: Platforms scale, thinking tools do not. Or so uh, Rizzak explains. Rizzak cautions against spaghetti AI. Kind of like that, uh, too many tools, no orchestration. Glenn, is that something real, something you've gotten the buzz around before? Absolutely.

Speaker B: And I've got to say I love the visual uh, of spaghetti AI. There's um, within the AI community there's a, well not necessarily with an AI community but there's um, a video that was AI generated a few years ago from one of the early video generators of AI, um content and it was Will Smith eating spaghetti and it was just nightmare fuel. It looked terrible. But then later Sora and whatever other video models uh, were able to make a video that looked like a real person eating spaghetti. So when you say spaghetti AI, that's the first thing I think of which is the disaster of the early days. And so I think it's actually a, a perfect visual for this. And I think the problem is, or the risk is that you lose explainability and trust when you just have a million projects going in a million different directions. And he painted a really vivid picture of what that spaghetti AI looks like in practice. Seven different point solutions. Each is handling Some small, narrow function, none of them are talking to each other. And you've got seven different levels of auditability. Um, and every time a problem surfaces, then somebody goes shopping for tool number eight. So it really, um, the planning around AI is like with any other big project, it depends on coherence. And if a board member asks, hey, how did you generate this forecast? And the answer involves, you have to stitch together outputs from half a dozen disconnected tools that nobody's, um, sure which came from what, then nobody's going to trust that number. So the recommendation, and I thought this was great, was to consolidate around two or three core platforms that your teams can actually invest in learning. And then again that, uh, except you had to learn Excel, now you have to learn AI. Uh, I think is great advice and it's coming from a very credible source because at Adobe he was on the finance team during the migration from box software to SaaS. And that had to be one of the most complex platform transitions in software, uh, history. So what did he say? He said like 27 R&D teams had to ship product on the same day under the old model. So he knows what orcus looks like at scale and he knows the cost of the fragmentation. So again, just fantastic insights.

Speaker A: Yeah, he was very plain spoken when he sort of reframed AI as a career skill for accountants. And to be honest, I don't know, I'm a little skeptical there. I mean, you know, the accounting department's not where you go for dynamic presenters or sales quality folks. Not that that's the same as AI. I'm just saying AI is something very different than, um, the accounting discipline. Yeah.

Speaker B: And so, you know, I spend a lot of time, uh, doing training for accountants. I work with AICPA and with several other groups and we talk about this a lot. And also the, the nature of, if you just limit it to generative AI, that is the opposite of the nature of an accountant. Generative AI is probabilistic and you're never getting the same answer twice. That doesn't work on your trial balance. So I, I get that. But that said, I think expand AI to mean not just generative AI, but AI can be automation, it can be, um, really think of any knowledge, work, task that we do and that accountants even do. And yes, there may be a generative AI sort of front end to it that you can speak to the data, but you don't want generative AI to be what is actually, uh, you know, making, making your journal entries necessarily, but automating platforms and also Broadly, I do think that it's. Accountants are just like everyone else. We have to understand how the tools work, where they can be applied safely, where they can't sort of know what those guardrails are. Um, and I do for all of us, AI proficiency, it's not going to be a nice to have. It's going to be table stakes for finance, uh, professional and accounting, uh, professionals. And I think that what's. We're also going to see something happening here where, you know, we've for years now in the US at least had a shortage of new accountants coming in. And as older accountants are retiring and now we have this new technology, I think there's going to be a push. Okay, if we can't hire for it, what are we going to build? And the build for accounting is going to be different than build, uh, for other departments, like you said. But what I appreciate, appreciate about his framing is he also tailors the advice by company size too. So he says, you know, if you're a large enterprise that already went through the IPO gauntlet and you have your processes defined and you, you meet all your, uh, requirements there, then really the challenge is platform selection. But if you're a smaller company, then you need a solution that helps you figure out what your workflows are before you can even layer in AI. And you can't even figure out how to layer in AI until you understand what it does. So I understand where he's coming from there. And um, I do think sort of his, you know, checklist approach, where you look at each workflow in your close process and then you start applying that automation step by step, that makes sense. And that's far more realistic for lean teams and the smaller businesses and just kind of a way to frame it and think about how accountants will use and interact with AI.

Speaker A: And just in regards to the planning process, I think across these three conversations, plus even more broadly, uh, we've never heard that AI is going to replace the planning process or the planning process is going to go away and instead we're going to have this new process AI driven. That's not what we're hearing. Right?

Speaker B: I mean, it, there's so many cliches around AI. It's that whole, you know, uh, AI is not going to replace you. Someone who uses AI is going to replace someone who doesn't. And then my framing for AI, at least right now, is I don't think of it as a full bot that's going to replace me. I think of it as like, um, an exoskeleton that you wear that makes you, gives you, you know, more strength and ability to do things. And if you think about what each of these guys brought to the conversation. So Bruce is talking about the data layer. Clean your foundation, consolidate your ERPs, get the three to five years of reliable metrics. And then Kevin's talking about that governance layer. And governance, we could do a, um, whole episode on just the governance part of it, but build that centralized function that filters and validates every tool before it gets deployed. And then, uh, Razak is talking about the platform layer. Consolidate around a few core systems. Build your skills incrementally and avoid that spaghetti AI that he talked about. And you stack those together. Then you have a maturity model that any FPA leader can use. That's what planning is going to be built on. Fix the data, govern the adoption, consolidate the platforms, and that's the setup. So our planning Aces today aren't predicting the future faster, they're designing systems that learn better.

Speaker A: These folks are telling their organizations to resist shortcuts to invest in foundations. Uh, that's such a, uh, finance leadership mindset.

Speaker B: It's asking what should we trust and when. So every One of these CFOs is building trust into their process before they add anything to the model. Kevin's testing and certifying tools. Bruce is accumulating the historical data. Razak is demanding platform level coherence. I mean this is all what we need to move forward. And that's the discipline that is going to separate real AI adoption from expensive experimentation. So there's companies that may seem like they're ahead right now. If they haven't, if they've already done this foundational work, great. But if they haven't, then they're just, they're automating chaos. And good luck with that. And so honestly from these planning aces, it's the kind of thinking that's always separated the best life finance leaders from the rest.

Speaker A: Nice. That's a great, uh, place to end, I think, for this episode. Doug, as always, discipline over hype and foundations before acceleration. Thanks for listening to Planning aces. Where modern FP&A leaders design what's next Thoughtfully.