Six months into a major AI transformation push, a Fortune 500 company had what looked like early traction: a centralized platform, a few promising pilots live, and a team of engineers and data scientists driving progress.
But the pace of delivering actual business value was lagging. Functional teams had ideas for where automation could create value, but no real way to act on them. Every request had to route through the platform team, and every iteration required weeks of coordination, resourcing, and approval across layers of governance.
Meanwhile, leadership was hearing about multi-million dollar wins from their peers and in the news. And every board conversation started with questions about the AI strategy and ended with an interrogation about lackluster results. The pressure was mounting.
Their initial thought was to increase investment into the central platform and supporting team. But they’ve seen this cycle play out time and time again, and are still nursing wounds from a years-behind-schedule ERP upgrade.
So they chose a different tack. To create real value with AI across the business, they realized that functional teams would need more than input. They would need agency.
The Implementation Mindset Shift
Most companies are approaching AI like they approached cloud migrations or ERP rollouts: centralized ownership, long timelines, and heavy IT involvement. This approach makes sense when dealing with rigid, fully deterministic systems that require every case and edge case be perfectly mapped and hard coded. But AI is different.
AI systems are malleable. They're designed to learn, adjust, and evolve. They improve with iteration cycles that provide more feedback and context. Sure, AI still needs to be integrated and governed by IT, but the functional business users are the ones who will create real business value with AI.
Functional teams already know where the friction is. They know which processes stall, which decisions are repetitive, and where automation could have the biggest impact. But they're stuck in a delivery model that treats every improvement like a multi-quarter project.
- A simple workflow change means creating yet another ticket for IT’s backlog.
- A new business policy requires months of cross-team coordination to update a processing rule.
- A novel idea to optimize a process gets shot down due to lack of resources, or stuck in low-priority purgatory
AI promises to augment and up-level an organization’s most valuable human resources, but unlocking it requires rethinking the roles, responsibilities, and rules-of-engagement that comprise organizational inertia.
Ultimately, this isn’t a tooling or infrastructure problem. It’s an ownership and agency problem. Functional business teams need to be empowered to take control of their processes. They need to be empowered to experiment, iterate, and execute rapidly.
AI enables this, but it requires a mindset shift and a new operating model.
The New Operating Model: Iterative Transformation
Historically, software has been expensive to build and cheap to operate. So, taking months to gather requirements, scope solutions, evaluate options, and align the organization before actually starting development made sense. “Slow down to speed up” as they say.
AI changes the math, though. Software can now be rapidly and cheaply developed, with costs shifting towards maintenance and inference. That means the economic risk of automating a process or optimizing a workflow goes down substantially.
Most importantly, AI lowers the barrier for experimentation and iteration.
We’ve seen this shift before with self-service BI.
Not long ago, getting a new dimension added to a filter for a critical report meant filing a request with the central data team. It could take weeks. Then came tools like Power BI and Tableau deployed on flexible, scalable data lakehouses like Snowflake and Databricks.
Seemingly overnight, business analysts started learning SQL to explore the data, build dashboards, and answer questions on their own. And central data teams shifted their focus upstream, integrating incremental datasets, optimizing infrastructure, and ensuring data quality.
AI is going to have the same effect on business process automation.
Business operators gain agency. They shift from gathering business requirements, supervising KPIs, and manually handling endless edge cases to directly designing and managing autonomous workflows. AI helps generate logic, surface exceptions, and adapt workflows based on new inputs.
This unlocks an entirely new operating model:
- Finance SMEs become builders, not just requestors
- Automation happens incrementally, not in multi-year phases
- Processes evolve rapidly, based on real-world feedback and outcomes
In this paradigm, business teams take full ownership of their processes, inclusive of the systems that automate them. They’re able to make more frequent, smaller bets — some that create immediate value and some that don’t, and get shut down instantly.
That doesn't mean IT disappears. Rather, they play a critical role in enabling this shift — governing data and system access, ensuring security compliance, managing integration, and providing reliable infrastructure for scale. But the loop between insight and action no longer has to route through them.
The organizations that empower their business teams to take full control of their operations are the ones that will create compounding value with AI. And it will happen quicker than you think.
Start Identifying "Complexity Sinks" in Finance Operations
With this new operating model in mind, the best places to start are what we call "complexity sinks". These processes require domain experts to babysit workflows and manage endless exception cases, following SOPs that require navigating multiple systems to make sure the business doesn’t grind to a halt.
Consider a typical invoice exception process: when an invoice doesn't match a PO or receipt, it triggers a complex workflow involving multiple people, systems, and decisions. Eventually, it might get escalated to the AP manager who needs to spend hours toggling between ERP screens, email threads, historical documents, and multiple spreadsheets to resolve a single exception.
This "complexity sink" is perfect for surgical AI implementation. Rather than waiting for a complete system overhaul, the AP team can implement a targeted, intelligent solution that synthesizes information from multiple systems, provides historical context, and suggests the ideal resolution path.
The result: median exception handling time drops from hours to minutes, and the team builds valuable implementation expertise for future AI projects.
These "islands of autonomy" — focused areas where finance teams can apply AI safely and independently — become the foundation for broader transformation.
What It Looks Like When Finance Leads
In this new model, finance and operations teams embrace AI to drive business transformation.
They start by fixing the pain points they know best. A controller sets up an agent that provides immediate insights to ad hoc inquiries about internal policies, process SOPs, or what-if scenarios. An AR analyst improves DSO with a new risk-based prioritization system. A procurement manager streamlines the new vendor onboarding process with an agent that automatically validates provided documents and information.
These aren't massive projects. They're tactical, high-leverage starting points that provide immediate productivity improvements and compound value as they take on more scope.
And it all happens without waiting on a quarterly sprint plan or rearchitecting core systems. Finance SMEs design and iterate. IT ensures the right guardrails are in place. Everyone moves faster and more confidently.
Workflows become intelligent and self-improving. AI agents surface edge cases, learn from human input, and refine themselves over time. Rules are transparent, and changes can be tested and deployed without a 30-day dev cycle.
What starts as a tactical proof of concept becomes a system that quietly saves the company millions.
That's the power of giving the business real agency.
How CoPlane Makes This Possible
After witnessing finance teams struggle with the limitations of traditional IT-led approaches, we built CoPlane specifically to enable this new paradigm — where finance teams own their AI transformation journey.
Unlike general-purpose AI platforms that require specialized technical skills, CoPlane was designed from the ground up to empower finance domain experts to implement, manage, and scale intelligent workflows that integrate with their existing systems.
Our platform provides finance teams with:
- Pre-built, finance-specific AI workflows that don't require coding
- Guardrails that maintain compliance while enabling finance team autonomy
- Integration capabilities that work alongside existing ERPs without disruption
- The ability to start with targeted workflows and expand incrementally
Unlike traditional enterprise software deployments, CoPlane can be implemented in weeks, not months, with minimal IT resources. And because our solutions are designed specifically for finance processes, they deliver immediate value while learning from your team's expertise.
We understand that governance and control are critical in finance operations. That's why CoPlane's approach maintains proper controls with built-in approval workflows, audit trails, and permission structures that ensure compliance requirements are met — even as finance teams move faster. Our platform works within your existing control framework while enabling the agility finance teams need.
A global consumer goods company using CoPlane automated 83% of their vendor master data maintenance workflows in just 6 weeks, while strengthening their control environment by eliminating manual handoffs and inconsistent approvals. Their finance team led the entire implementation while IT provided oversight rather than hands-on development.
This is an example of the Finance AI Implementation Framework in action**:**
- Start Small: Automate a single manual, tedious task.
- Leverage Existing Systems: Target opportunities that augment existing systems without requiring massive changes
- Expand Strategically: As value proves out, extend to adjacent tasks in the end-to-end process, then to adjacent processes.
Where Will You Start?
AI is already reshaping how finance work gets done. The question isn't whether your company will adopt it — it's who will lead the change and how quickly you'll see results.
You don't need to reinvent your finance tech stack. Start with one high-friction process. One team. One workflow.
Find the exceptions your ERP can't handle efficiently. The approvals that bottleneck operations. The reconciliations that consume days each month. Then build something better — and let it grow.
With CoPlane, you already have everything you need to start your finance transformation today: targeted improvements that deliver immediate value while building toward your long-term vision.
You already know where the opportunities are. Now you can act on them.