Carlos Vega | CEO and Cofounder of Tesorio | Connected Financial Operations
In 1990, Michael Hammer wrote a now-famous piece for Harvard Business Review subtitled “Don’t Automate, Obliterate.” His message was radical for the time: Instead of grafting technology onto outdated workflows, companies should reengineer those workflows entirely. The point wasn’t efficiency—it was reinvention.
Twenty-five years later, Fred Wilson revisited the phrase in a blog post identifying industries overdue for disruption. His takeaway, while falling short of an investment thesis, was that the real opportunity comes when you stop preserving the old model and build something entirely new.
Now, it’s 2025. AI has become the most significant platform shift since the internet. But just like with the internet in its earliest days, the temptation is to make AI conform to existing systems. We see companies automating here and there: speeding up manual tasks, plugging large language models (LLMs) into workflows and generating better forecasts and more polished emails. That’s great, but not the revolution we were promised.
But let’s not forget the lesson: Transformative tech doesn’t just accelerate what exists. It reshapes what’s possible. And in enterprise finance, that means it’s time to stop automating and start obliterating.
Why Automation Falls Short In Finance
Most enterprise finance teams still run on a precarious stack of legacy spreadsheets, brittle robotic process automation (RPA) scripts and semi-integrated systems. Workflows such as invoice matching and collections are held together by group knowledge and human logic. It’s a house of cards built in the 1990s.
When you apply AI to this legacy infrastructure, the best you can hope for is marginal gains: faster reconciliation, fewer manual errors, better forecasts. That’s not nothing—but it’s not transformation, either, especially not when macro volatility demands real-time agility and boards expect live financial visibility.
According to McKinsey, while 92% of companies plan on increasing their AI investment in the next three years, only 1% would consider their AI deployment “mature.” Far too few companies have made the leap toward meaningfully restructuring their systems around what AI does best: pattern recognition, probabilistic inference and autonomous decision making. That’s a massive opportunity cost—unless you’re part of that innovative 1%. For them, it’s a competitive advantage that compounds at the speed of AI advancement (and if you’re paying attention, you’d know that’s fast).
From Linear Workflows To AI Agents
This isn’t an appeal to CEOs to bolt on piecemeal “AI solutions”; it’s an invitation to reimagine workflows entirely in the face of this technological inflection point. Agentic AI is a great step in this transformative direction: Instead of making humans more efficient at step-by-step processes, AI agents can replace entire categories of coordination work. Think of what happened in DevOps and customer support: Agents can now resolve tickets, manage deployments and trigger follow-up actions without human intervention. Finance is next.
AI agents can now:
• Match payments to invoices using structured and unstructured data
• Predict which customers will delay payments, and tailor outreach accordingly
• Forecast cash flow in real time across banks and enterprise resource planning (ERP) and customer relationship management (CRM) systems
• Surface anomalies or liquidity constraints before they become crises
This is less about finding and fixing inefficiencies than it is about reducing the burden of human work for repetitive tasks entirely. The future will feature finance teams who can focus on scenario modeling, capital planning and strategic reinvestment.
And let’s be clear: Cash flow isn’t just a finance problem. It’s a data problem. AI gives us a fundamentally new way to act on that data: instantly, continuously and at scale.
The Technical Pillars Of Post-Automation Finance
To support this leap, enterprises need to rethink their financial infrastructure. Three areas stand out:
1. Unified Data Fabric
AI agents require complete, normalized and real-time data. That means connecting ERP and CRM systems, bank feeds, treasury systems and collections tools into a coherent, queryable layer. No more data silos. No more nightly batch jobs.
2. Embedded LLMs And Predictive Models
Foundational models trained on enterprise finance data can classify remittances, infer payment intent and generate rolling forecasts. They typically can improve with use and scale across regions and workflows.
3. Agentic Orchestration Layers
This is where the “obliteration” happens. Instead of scripts and approval loops, agentic layers autonomously execute tasks based on outcomes—e.g., resolving discrepancies, initiating follow-ups, reallocating working capital—all without manual triage.
Why This Isn’t ‘Automation On Steroids’
Automation says: “Here’s a task a human does. Let’s do it faster.” AI obliteration asks: “What’s the result we want? What’s the shortest path—and does it even require a human?”
In the case of invoicing, the goal isn’t to process them 30% faster. It’s to eliminate them through real-time data sync and payment agreements embedded at the contract level.
This shift demands courage—and vision. But it’s already underway.
How To Stop Making AI About Headcount
I started my career in finance. I know how much time is spent cleaning data, reconciling systems and chasing payments. Finance teams were never meant to be call centers or human ETLs, extracting, transforming and loading data.
Imagine if your most strategic thinkers weren’t tied up reporting what already happened but could focus on shaping what will happen. That’s the opportunity AI creates, but only if we’re bold enough to take it.
If you’re thinking in terms of cost-cutting exclusively, you’re not thinking big enough. You can only cut the bottom line so much, whereas the top line? Theoretically infinite.
The information provided here is not investment, tax or financial advice. You should consult with a licensed professional for advice concerning your specific situation.
Forbes Finance Council is an invitation-only organization for executives in successful accounting, financial planning and wealth management firms. Do I qualify?