The Financial Back-Office is Stuck in 1995
Enterprises are magnificent in their complexity. A labyrinth of legal entities domiciled across jurisdictions, operating multiple companies with even more business units, all reporting to different regulators and authorities. Each entity within the enterprise is obligated to timely and compliantly execute transactions, manage performance, keep records, and report to management, shareholders, tax authorities and regulators.
This is no small task, and it falls on the back office to execute the transactional finance workflows that ensure compliance with these labyrinthian regulations. Yet this backbone of the enterprise remains stuck in 1995. There’s a lot of software and a lot of processes, but very little automation.
Here’s why: in 2025, the back office has become an ouroboros of ERP modules, consulting firms, and outsourcing companies — each feeding into the next in an endless cycle of complexity and billable hours.
This self-perpetuating complex costs organizations millions (or more) to maintain, and is holding the office of the CFO back from long overdue innovation. But that’s all about to change.
The Trillion-Dollar Triangle
Every enterprise back office sits at the center of what we call the “Unholy Trinity”: Enterprise Resource Planning (ERP) vendors, the Big Four accounting firms, and Business Process Outsourcing (BPO) providers. Each vertex of this triangle exists in perfect symbiosis with the others, creating a system so deeply entrenched that it makes the military-industrial complex look like a weekend hobby project.
Let’s break down how this works:
- ERPs create byzantine systems that require armies of developers fluent in arcane frameworks and languages like ABAP to implement and maintain. These implementations are so complex that they’ve spawned an entire economy of firms that specialize in them.
- The Big Four firms help write the regulations that make these systems necessary, then charge astronomical fees to interpret those same regulations. They’ll helpfully point out that you need both their services and an industrial-grade ERP to stay compliant.
- And when all of this inevitably creates more work than your team can handle? Enter the BPOs, ready to supply endless ranks of offshore workers to handle the manual processes that somehow still exist in your “digital” transformation. Their success metric? Headcount growth. In 2025. Let that sink in.
And at the center of this maze are the enterprise CIOs and CFOs, who are trapped in the maze. In an earnest effort to be transparent and accountable to their stakeholders, they find themselves hostage to the back office industrial complex.
The ERP Trap(s)
If this is such a bad deal, why don’t companies simply switch to better solutions?
If only it were that simple.
For a large company (over $1B in annual revenue), the average ERP implementation takes nearly three years and costs somewhere between 4% of your topline and the GDP of a small nation. But that’s just the beginning.
By the time you’re done, you’ve customized the system so heavily that upgrading it requires a team of archeologists to understand the decisions made by consultants who left the project before the implementation was even finished. Maintaining the systems that run your business turns into an exercise in painting the Golden Gate Bridge.
This creates what we call the Implementation Trap: your ERP becomes too expensive to replace, too critical to modify, and too complex to simplify.
“But wait,” the ERP vendors say, “we have a solution: Clean core! Move to our cloud! All your implementation problems will disappear like magic!”
Let’s translate what’s actually happening here: Your ERP vendor is pulling the enterprise software equivalent of a protection racket.
The “cloud migration” strategy is presented as a technical salvation. The promise of a “clean core” is painless upgrades. But what they don’t mention is that this clean core comes at the cost of moving all your customizations into their proprietary cloud platform, where they control both the technical boundaries and the pricing.
Want to integrate with a new fintech solution? Better hope it’s on the approved partner list. Want to determine when (or if) to upgrade your version? That’s no longer your decision.
It’s a masterclass in vendor lock-in. They’re not just selling you cloud hosting – they’re buying themselves a captive audience for every new “innovation” they want to push. Your clean core becomes their clean shot at your budget, forever.
You can see the writing on the wall. SAP has already extended their deadlines for cloud migration twice — but the second is only allowed if customers sign up and pay for the RISE with SAP migration package.
When your choice is between cloud migration, and a game of “will it be 2027, 2030, or 2033 ?” chicken with technical obsolescence, it’s not really a choice at all.
The Compliance Shield
Meanwhile, the Big Four firms have mastered a particularly clever trick: they help write the regulations that make financial operations complex, then sell you the expertise to navigate that complexity.
Every new regulation adds processes requiring manual oversight and verification. Each framework becomes a de facto standard, demanding more consultants and labor.
And to make matters better, The Big 4 can’t even wrap their minds around auditing a company that’s not running one of the “blessed” systems from the top ERP vendors.
The alliances between the Big Four and ERPs run very deep. Each firm maintains official partnership statuses (e.g. Diamond Partner or Global Strategic Services Partner) with SAP and Oracle. Frequently, firms and ERPs co-develop industry-specific templates for implementations, often branded by the consulting firm (such as the Deloitte Reimagine Platform on SAP or PwC Industry Cloud).
And every time the Big Four successfully lobbies for new regulations, these “accelerators” require new controls and compliance capabilities, furthering the complexity of the ERP implementation and the need for expert oversight.
New e-invoicing regulations? Hire the consultants to specify your requirements, obligations, and implementation plan. Then pay them for a half-working implementation of the ERP vendor’s new module.
Yes, regulations serve a vital purpose — but the conflict of interest here is staring the world in the face.
The Human Labor Cycle
Then there’s BPO — an entire industry predicated on labor cost arbitrage with incentive to ensure humans are the atomic unit of value.
As Kimberly Tan from a16z points out, “their business model depends on employing people and selling the output of that labor to customers.”
Think about it: In an era where AI can generate photorealistic images from text, your accounts payable process still requires human beings to follow desktop procedures that involve manually keying in invoice data. Something’s not right here.
BPO Industry Growth Analysis (2015-2024)
Loading revenue data...
So how did companies like Wipro and CapGemini build massive, durable businesses around labor arbitrage? They manufacture stickiness by slurping up their customers’ institutional knowledge and encoding it in complex desktop procedures that customers have to beg to even get a look at.
Furthermore, they build and operate custom systems to support the increasingly complex processes they orchestrate for their largest customers, ensuring maximum pain for any customer that tries to escape their grip.
Baumol’s Back Office
There’s a fun economic concept that explains why your back office feels like it’s always stuck in 1995: the Baumol effect, or as it’s sometimes called, “cost disease.” The basic idea is that sectors that don’t see productivity improvements still have to raise wages to compete with sectors that do.
The back office has been the perfect example of this effect in action. Despite decades of “digital transformation,” the core work still requires roughly the same number of human beings pushing roughly the same number of buttons. Sure, they’re pushing them on SAP instead of paper ledgers (or— as it were, clicking around in Fiori UIs on S/4HANA instead of punching TCodes into ECC), but the fundamental productivity hasn’t changed much. Meanwhile, these workers need to be paid more because, well, productivity has skyrocketed everywhere else.
This creates a particularly nasty feedback loop in our Unholy Trinity:
- Rising wages force BPOs to constantly seek cheaper labor markets
- This geographical arbitrage keeps the focus on human labor rather than automation
- ERPs and consultants optimize for these human workflows, further cementing them
- Rinse and repeat until you’re explaining to your board why AP costs keep rising
So…what about AI?
Why Their “AI Initiatives” Won’t Save Us
Recently, every player in this trinity has announced their own AI initiatives. Wipro claims a 140% increase in AI adoption. Infosys has “100+ gen AI agents.” Accenture boasts “$1.2B in gen AI bookings.”
But here’s the thing: These initiatives are like being promised a faster horse in the age of the automobile. The fundamental business model remains unchanged. They’re incentivized to maintain the complexity that makes their services “necessary.”
AI promises to provide the “automation” ERPs sold to you 10 years ago, the guidance the Big 4 extorts you for, and the labor BPOs upcharge you for– all at a cost that will go to zero faster than most people can imagine.
Real transformation can’t come from within this system. It’s like expecting traditional stockbrokers to have invented Robinhood. The incentives are fundamentally misaligned.
The AI Inflection Point
Here’s where it gets interesting: AI completely breaks this model. When you can automate not just the button-pushing, but the actual decision-making, you’re no longer bound by the cross-elasticity of human labor markets. The very concept of labor productivity starts to mean something different.
We’re entering an era where the marginal cost of an additional unit of customized software is approaching zero. Software will become cheaper and more abundant. It will be deflationary. AI is fantastic at performing well documented human tasks like audit and common BPO tasks. Basically: AI, correctly and scrupulously applied, breaks this entire industry wide open.
We’ve seen moments like this before. Remember when you needed a human being to execute a stock trade? Neither do I. That $200 commission and phone call to your broker has been replaced by instant, zero-commission electronic execution.
The same transformation is coming to the back office. AI isn’t just another tool to add to the existing stack — it’s the catalyst that will expose how much of this “necessary” manual work isn’t necessary at all.
Think about it: What happens when you can automate not just the data entry, but the decision-making? When AI can understand context across languages and documents? When it can learn your business rules and apply them consistently? When it can provide a full audit trail for every single ledger entry and seamlessly map a set of controls and policies? When it can transform static records into autonomous, goal-seeking entities?
The entire premise of the current system — that complexity requires human intervention at every step — begins to crumble.
What Happens Next
The back-office industrial complex won’t disappear overnight. Like all seemingly permanent power structures, it will resist change with impressive vigor. But the cracks are already showing.
Companies are starting to ask dangerous questions: Why does our ERP need six months of configuration to handle a new subsidiary? Why are we paying consultants to interpret regulations that could be encoded as rules? Why do we have hundreds of people manually processing documents in 2025?
The answers to these questions all point in the same direction: The system is operating exactly as designed: to maintain its own existence. But like the traditional stockbroker, its time is running out.
“AI is going to change how work gets done across every facet of every company,” says Jordan Angelos of Ribbit Capital. “The businesses that fully embrace AI—inside and out—will be the ones that win over the coming decades. You can’t expect to thrive in the AI era if your own company isn’t living and breathing the technology at every level, including the back office.”
The revolution in back-office operations won’t come from within. It will come from companies building on top of AI to create systems that are inherently simple, adaptable, and intelligent. Systems that work the way modern businesses do — not the way they worked in 1995.
The only question is: How long will you wait to join it?
CoPlane is building the intelligent financial operations platform that makes this transformation possible. Want to learn more? Drop us a line.