"I can't explain to my board why the AI made that decision." – CFO, $180M company

Most AI tools can't. Maximor agents can. Their agents automate journal entries, accruals, reconciliations, and reporting. And when they’re uncertain, they escalate with context before anything posts. 

No black box. Every entry is traceable.

Trusted by CFOs at Sierra, Gusto, Perplexity, Eightfold, and more.

See what automation looks like on your data for free 

Picture the scene... A handsome young Secret CFO sat in the hallway outside the BU CFO's office of a global mega-corp.

I was three layers below… merely a young FP&A analyst. I'd only met Matthew - the BU CFO - a few times. He was a strange, introverted dude. Despite only being in his mid-thirties, he was already every bit the stereotypical accountant. Like he'd heard about eye contact... but had always landed on "why risk it."

I had no idea why I was there, but I'd clearly been chosen for something.

He told me the Group CFO had just signed a major CapEx to implement a new group-wide FP&A system. And that I would be the project lead for his business unit. It was the first time I'd come across anything beyond the day job… it felt exciting.

The group was actually pretty good at FP&A. But exceptional people, rather than great tooling, made it outstanding for its time. This was twenty years ago. Everything ran through Excel, or Access databases (RIP) for the moments when we needed 65,537 or more rows. With a lot of reliance on fragile VBA code.

It was a function that understood the business well and delivered what was needed, on time and in the right place. But through brute force rather than finesse.

Matthew told me I'd keep my day job; I'd been automating a lot and had created some space. He told me to expect to spend 30% of my time on the project. Red flag number one. Not that I knew that at the time. This was a $30B division… it was insane to expect to happen without dedicated resource.

The system we were to implement was an extension module of our ERP. The selection rationale wasn't entirely clear, but it had the same name as our ERP in the title, so that was enough... Right? Right?! Red flag number two.

I'd be working with Katie, an up-and-coming Finance Director, seconded onto the central project team. I was the implant in my BU.

We went through months of process design, chart of accounts mapping, and requirements building. That this work was happening after the CapEx had already been signed was, of course, red flag number three. I was on a roll.

The pace started to slow after the first six months. Lots of meetings. Nothing seemed to happen. I was too far down the food chain to understand what was blocking it, and had every reason to assume that smart, important people were doing smart, important things somewhere above me.

Then, about a year in, I read an internal announcement. Katie had been promoted to VP in a completely different part of the business. I emailed to congratulate her and asked what this meant for the project.

I'm still waiting for a reply… twenty years later.

A few weeks after that, the SVP running the whole program got moved into an international CFO role.

The already-fading momentum became a total vacuum. Nothing was happening, and I didn't even know who to ask. Matthew had moved on. The people I worked for in my day job didn't know and didn't much care. They just wanted my time back.

I was bewildered. I did the napkin math on what had been spent, based just on what I'd seen, and it ran into the millions. This had been badged as the most important initiative for the finance team twelve months earlier. Now it had evaporated. When I mentioned it, eyes shuffled awkwardly, the way they do when someone brings up a disgraced family member at the Christmas table.

I kept waiting for an announcement. Some kind of official closure. A post-mortem. Anything.

Again… twenty years on, I'm still waiting.

This wasn't just my first exposure to a failed systems project. It was my first exposure to egregious corporate waste. I was certain we were the only business on the planet where this had happened.

How naive I was.

Over the years that followed, I saw similar themes play out on different systems projects. While none were quite as harrowing as that first experience, I came to understand how rare truly successful, efficient tech rollouts in finance are. Meanwhile, the tech debt just kept compounding, with Excel doing the heroic heavy lifting to keep everything glued together.

I knew I wanted it to be different when I was in a chair to call the shots. But I also learned how hard that was.

And the most important lesson of all: while it wasn't impossible to win, it was harder than it should have been… Because the whole game was rigged.

Welcome to a new series of The Secret CFO's Playbook. This month: Tackling CFO Tech Debt.

If you come from the tech industry, the term will be second nature. Technical debt is fancy words for where your technology is lagging behind where it should be. In the tech sector, that's mission-critical, because tech is the product. Crap tech = crap product = crap business.

But operational tech debt doesn't get the same attention. And I'd argue no function has accumulated more of it than the Office of the CFO.

Put simply, CFO tech debt is the gap between the systems your finance function runs on and the systems it should be running on. It accumulates quietly, compounds constantly, and gets more expensive to fix the longer you leave it.

CFO tech debt is difficult to solve because, while the cost to service the debt is huge, so is the cost of addressing it. It's a little like being an overleveraged operating company. You can't afford to pay down the debt, but you also can't afford the interest.

Where it differs is that there is no hard maturity on tech debt like there is on financial debt. Nothing that ever forces the cycle to an end. No restructuring agreement, no creditor haircut. It just kind of… stumbles on.

Well, in this series, we are going to help you find that tech debt in your finance function and equip you to de-lever it.

The true cost of your tech debt isn't easy to see. But the symptoms are.

The biggest symptom sits on your shared drive as an .xlsx file. Excel is the pain relief for tech debt. It is not the cure.

The direct cost is visible enough. It's the army of people manipulating spreadsheets and manual processes all day. And that cost is real. The Hackett Group (in 2024) found that best-in-class finance functions run at 0.54% of revenue. Poor ones run at 1.76%. On a $500m business, that's a $6m annual gap just in the cost of running finance.

Not nothing, but it's not the point either…

You are never going to move the profit needle for your business with a finance headcount saving effort alone. The real cost is much harder to quantify, and much harder to stomach:

These examples touch every line of the financial statements. They affect the whole organization at a strategic, operational, and cultural level. The true interest cost on your tech debt is impossible to calculate precisely. All I'll say is, for most businesses, compared to this list, the headcount savings is likely a rounding error.

For the last thirty years, tech for the CFO’s office (and the whole organization) has been dominated by three little letters. E-R-P.

The ERP dream that was sold was intoxicating: one system to do everything. One version of the truth for every piece of information in your business. One system to rule them all.

In most businesses, it never quite worked out like that (more on why next week). But it worked just enough to get its claws in.

To be fair, ERP did solve some real problems.

It ran the billing cycles. Paid the suppliers. Measured inventory (if you were lucky). And owned the GL. For some, it went deeper: Material Requirement Planning in the operations.

It just never came anywhere near delivering its promise.

And getting these systems in was hard. A kind of corporate Darwinism meant most of us got most of the way there eventually.

But if you think getting them in was hard, wait until you try to leave…

In 2024, Morningstar analysts covering SAP defined the entire competitive moat of the business as switching costs. Not the product. Not the technology. Not the innovation roadmap. The cost of leaving.

And to be fair, we helped build those walls ourselves. Twenty years of customizations, integrations, and process workarounds will do that.

Think about that for a second. A business worth over a quarter of a trillion dollars, whose entire competitive advantage is how painful it is for customers to walk out the door.

It turns out half-assed system implementations are big business.

It’s like checking into a hotel and discovering the door handles only work from the outside... and your rack rate goes up every time you call reception for help.

And while I'm pointing at ERP vendors, they are just the center of the stack. The FP&A platforms, consolidation tools, and data layers built on top followed exactly the same playbook. They didn't solve the problem. They upsold on top of it.

Then there's the ecosystem that grew around all of that. Implementation partners who get paid by the hour and for the complexity. Support contracts that auto-renew, whether you need them or not (you do). Upgrade cycles that arrive just as you've finished the last one. "Are you in Excel hell?" asks the software salesperson (who used to work for the Company who put you there.)

And the Big 4 or Gartner will always be on hand to tell your board exactly what went wrong at the price of $5,000 per slide.

I'll be honest, if I sound bitter, it's because I am. I've sat in too many meetings with vendors from that era cheerfully announcing a significant price increase on a system I can't leave, for a product I don't love, delivered by an account manager who doesn’t care about my success.

The ERP Industrial Complex has created tremendous shareholder value over the last thirty years. It’s one of the greatest business models ever built. But they don't have customers. They have hostages.

I'm sure this wasn't the original intention. But incentives win. They always do. (More on that in a moment)

Well… it would be nice to pretend. But no.

This has happened on our watch. Most CFOs have either abdicated ownership of their tech stack to IT entirely or IT reports to them, and they've never given it what it needs to win. We've digitized too slowly or focused on the wrong things. Patching over the symptoms by investing in insight layers and dashboards, without addressing the real problem underneath: shitty data.

A minor sniffle in your master data or AP workflow turns into a full-blown raging flu by the time it reaches your analytics layer.

But how to fix it?

Valuable finance resources are tied up for four weeks a month, closing, reporting, and analyzing, leaving zero time for the project work that could actually change that. Sure, you could bring in more headcount. But it's the people doing the work who understand the problem well enough to actually fix it. The ones you need on the project are precisely the ones you can't release.

It's a rat trap.

And many have quietly stopped trying… or haven’t tried hard enough (myself included).

Psychologist Martin Seligman called it learned helplessness. In his famous 1960s experiments, dogs subjected to repeated uncontrollable electric shocks would eventually stop trying to escape, even when the barrier was lowered, and escape became simple. They'd already learned that trying made no difference.

For a lot of finance teams, that's exactly what happened with technology. Not laziness. Not negligence. A perfectly rational response to repeated failure in a game that was stacked against them.

And embracing serious tech debt has historically carried asymmetric career risk. F*ck up an implementation badly enough and you wreck your reputation. If it goes well, you’ll work extremely hard, no-one will ever really thank you, and it’ll likely be your successor who reaps the benefit.

But, finance has fallen behind on technology. A long way behind. I believe we are now the most archaic department in the C-Suite. Guys … we are behind HR…. are you cool with that? I’m not.

It's time for a change.

Legendary business thinker Clayton Christensen described the Innovator's Dilemma like this:

"The very decision-making and resource-allocation processes that are key to the success of established companies are the very processes that reject disruptive technologies."

We can apply this directly to the ERP industrial complex. They innovate just enough to protect today's margin. No more. That keeps the business humming, but over time, a vulnerability builds. As the incumbent loses touch with the problem their customer is actually trying to solve, others focus harder on it. Hungrier. Cheaper. Unencumbered by a business model built on complexity.

That vulnerability can take a long time to show up, even a decade or two. And they can make a lot of margin while they wait. But eventually it will show up. Just ask Blockbuster or Kodak.

And the trigger is always the same: a breakout technology that collapses the cost of entry for new disruptive challengers…

AI has entered the chat.

"Yeah, sure, Secret. It's different this time."

I admire the skepticism. Hailing AI as the savior of finance technology might seem lazy. A little LinkedIn. And honestly, I’m not sure even I believed what I'm about to say three months ago.

But let me zoom out for a moment.

Most of us are looking at AI through the direct application lens. Playing with the foundational models. Getting braver. Asking how we can use this thing. Some claim it's going to change everything overnight (it won’t). Others can't get off the starting blocks because of privacy and security concerns. Next month's Playbooks will be a full deep dive into AI for finance. We'll tackle all of it.

But now I want to talk about something less discussed. The second and third-order effects of AI on finance technology. Because that's where the real story for CFO’s is:

Ramp is the most talked about example, but there are others coming right behind them across different parts of the finance tech stack, building in exactly the same way.

Am I saying these businesses are going to eat the legacy giants overnight? No. The biggest names are big enough to pivot or acquire their way out of trouble if they have to. And the switching costs for the biggest enterprises are still high… so their top line will be safe long enough to figure it out.

But at a minimum, we are going to see a flood of high-quality new entrants driving product standards up and switching costs down. That's just supply and demand. 

So here is my ask.

Put your historical skepticism and your entirely legitimate systems PTSD to one side. It is time to look at your finance tech stack with fresh eyes. Understand your true CFO tech debt, and then figure out how to start paying it down. I hope this series will help you do that.

But first, a little group therapy. I'll be sharing more of my own scars through this series. Now I want to hear yours. I’ll share my favorites (and won’t name names).

Don’t hold back … give me a full trauma dump:

OK … with that load lightened, let’s get ready to dig in with fresh energy, and look at your finance tech stack with clean glasses.

After decades of false promises, and legit excuses, the Office of the CFO lags further behind on technology than any other function. It's time to catch up.

The conditions are finally right. The economics are shifting, the builders are here, and the excuses are running out.

Your finance tech stack is probably a mess. That's not entirely your fault, but it is your problem. And more importantly, it is now your opportunity.

Before we get to the solution, we need to go deeper into the diagnosis. How did we actually get here? Where precisely are the root issues in most finance tech stacks?

That's where we go next week.

:::::::::::::::::::::::::::::::::::::::::::::::::::
:: Thank you to our sponsor ::
:: Maximor ::
:::::::::::::::::::::::::::::::::::::::::::::::::::

If you enjoyed today’s content, don’t forget to subscribe.

Disclaimer: I am not your accountant, tax advisor, lawyer, CFO, director, or friend. Well, maybe I’m your friend, but I am not any of those other things. Everything I publish represents my opinions only, not advice. Running the finances for a company is serious business, and you should take the proper advice you need.