AI agents that run your EPM models — not chat over them.
StrategicFinance.ai™ puts agents built on Anthropic’s AI models to work inside your cost, profitability and planning models. They run allocations, explain variances and draft the management narrative, and every number traces back to the model that produced it.
We start on Oracle EPM, in banking and healthcare.
AI rarely fails because of the model. It fails between the model and the P&L.
of AI projects fail, by some estimates, roughly twice the rate of IT projects without AI.
RAND Corporation, The Root Causes of Failure for Artificial Intelligence Projects and How They Can Succeed, 2024
of companies have built the capabilities to move beyond proofs of concept and generate tangible value from AI.
Boston Consulting Group, Where’s the Value in AI?, 2024
Capabilities don’t create value on their own. The decision they enable does. So we start from the decision and trace it to the P&L, then put the agents to work.
Sáasil™ AI Management Accounting Framework
Decision-to-Value™ unifying people, strategy, data, processes, applications and governance to drive measurable performance and business value.
Enterprise strategy & value
Decision and performance assessment · roadmap · business case
Talent
Training · organizational design · workforce
Process D2V™
End-to-end workflows · activity analysis · information needs · ROIC optimization
EPM agentic orchestration
Intelligent agents · workflow orchestration · embedded analytics · natural-language interaction · proactive insights and actions
Domain models
Revenue · costing · investment
AI model governance
Model choice · controls · traceability of every output
EPM applications
Core · profitability · planning · narrative reporting · embedded AI
Data & analytics
Integration · data platform · knowledge base
Infrastructure & security
Cloud · security · scalability · APIs · extensibility
The three columns follow the capability logic McKinsey sets out in Rewired (Lamarre et al., Wiley, 2023; 2nd ed. 2026). Decision-to-Value™ adds the trail from each capability to the decision it enables and the P&L it moves.
Oracle’s agents run the model. We make sure it’s the right one.
Oracle EPM now includes its own Planning and Profitability and Cost Management agents. They build and run allocations well, and they’re built for every industry. Your model isn’t. StrategicFinance.ai™ works on top of them, through your platform’s own APIs and security, and adds what a general-purpose agent can’t know.
Banking models with funds transfer pricing, expected loss and capital charge. Healthcare models with cost of care by service line.
Every driver tied to the activity and the information it depends on, not just to a cell in the model.
Agents that question the driver before they run it, and test the decision against ROIC or RAROC.
Who owns each rule, how it changes, who reviews it, and what you show your board or regulator.
Four agents, each tied to a decision finance already owns
Profitability Agent
Works alongside Oracle’s PCM Agent. Checks each allocation cycle against your industry’s drivers, flags anomalies before close, and explains margin by product, customer, channel or service line.
Planning Agent
Works on top of Oracle’s Planning Agent. Ties each driver-based scenario to the process and the P&L line it moves, so a forecast change comes with its cause.
Narrative Agent
Drafts management and board commentary straight from the model, with every statement linked to the figure it describes. Your team edits; it doesn’t start from a blank page.
Strategic Agent
Simulates price, exit and investment decisions against ROIC or RAROC, using the same cost and capital logic the model already applies.
Every agent runs on a governed model. If the model isn’t ready, we say so first, and the Model Governance Review is where we start.
Platform scope: Oracle EPM (Planning, EPCM, Narrative Reporting) first. OneStream, SAP and CostPerform are on the roadmap.
Finance work is long, multi-step and has to be explained
We build on Anthropic because the work looks less like answering a question and more like running a close: many steps, each checked against the last, ending in something a controller can follow.
Two industries where the number gets questioned
Customer and product profitability
Funds transfer pricing, activity-based cost, expected loss and capital charge, brought together into economic profit and RAROC by customer, product and branch.
Cost of care
Direct, semi-direct and indirect cost traced through clinical drivers to service lines and medical units, so funding and consolidation decisions rest on what care actually costs.
Start with the model nobody fully trusts
A working conversation
With the principal who would design your model, not a sales team. Which decision, which model, which platform. If an agent isn’t the right answer, we say so.
Model review
We review your cost model, drivers and EPM setup to find where an agent can operate safely and where the model needs fixing first.
First agent
One agent on one decision, running against your model, before anything scales.
Deploying Anthropic resources in a finance team? We bring the model layer.
AI projects in finance stall when nobody owns the cost and profitability logic underneath. We work behind your team on that layer: driver design, allocation rules and the EPM model the agents operate.
Talk to the principal who will design your model.
No forms and no sales team. Write to us about the decision you need to support. You’ll get a reply from Pedro San Martín within two business days.
Price, Profitability & Cost Management
30+ years in cost and profitability models
Naples, Florida
Contact Leonidas Rey, PwC Interamericas, PwC Strategic Finance Center of Excellence
leonidas.rey@pwc.com