A policy surface is the place an operator can see and change the rules the agent has to survive, without opening a prompt or a repo. Amounts. Propose vs post. Who owns the miss. Most agentic finance products skip that object. The rules live in instructions. The demo still clears a clean invoice. If the dual-review
The Policy Has to Live on the Glass Read More »
This site is a positioning house for agentic tools sold into the Office of the CFO. Read it in this order if you build or market those products. Do not start with autonomy. Start with who owns the miss. 1. The four arguments Trust before autonomyFinance buyers reject the agent-as-hero story. Lead with control, then capability.https://www.timpratt.me/trust-before-autonomy/ The
Most agentic AI deals do not die because a rival agent told a better story. They die because the buyer already has a default. Hire another clerk. Keep the RPA bot. Use the ERP screen. Send the pile to shared services. Stand up a chatbot and call it a pilot. If the first conversation never names that
Agentic AI in Finance: What Buyers Compare It To Read More »
Most agentic close products still lead with autonomy. The homepage says the agent can post, reconcile, and finish the close. The demo shows journals moving. The room nods. Then a Controller asks who owns a bad entry, whether the posting can be explained, and whether “faster close” means fewer controls. That is where the story
Agentic Close Positioning: Before and After Read More »
Most agentic AI business cases still treat the same line as the win. We will automate this workflow and reduce headcount. That line is easy to put on a slide. It is easy to defend in a budget meeting. It is also how a lot of finance transformations stall. The hours come out. The people
Budget Savings Are Not Value Creation Read More »
Most agentic AI pitches fail in the same place: not in the demo, but in the handling of CFO agentic AI objections. Finance buyers rarely say “this is unimpressive.” They say something more specific, more cautious, and more revealing. If the response is generic, the deal slows down. If the response shows you understand the
What CFOs Actually Push Back On When Agentic AI Is Pitched Read More »
Most CFOs start agentic AI where visibility is highest. Forecasting. Board reporting. Scenario narratives. The work already gets executive airtime, so it feels like the obvious place to prove that AI belongs in finance. Then the pilot hits messy data, unclear ownership, and a public tolerance for error that is close to zero. Progress stalls. The conclusion
The Back Office is the Beachhead Read More »
Illustrative positioning case based on common patterns in agentic finance products. Not a client case study. Most agentic finance products still lead with autonomy. This before-and-after case on agentic AP positioning shows why that story fails with finance buyers, and what a control-first rewrite looks like. The demo shows invoices moving end-to-end. The narrative promises
Before → After: Positioning an Agentic AP Exception Product for the Office of the CFO Read More »
Agentic AI governance finance questions dominated the audience chat during OpenAI’s recent webinar on how its own finance team uses AI. The demos focused on faster forecasting, live scenario collaboration, automated workflows, and interactive dashboards. One presenter even joked about making forecasting “great again.” The audience asked different questions. According to reporting by Adam Zaki on CFO.com
OpenAI’s Finance AI Demo Missed the Real Question Read More »
A Controller once told me, after a polished demo of an autonomous invoice-processing agent: “This is impressive. But if it posts the wrong entry, whose job is on the line, yours or mine?” The conversation ended shortly after. That question sits at the center of almost every serious evaluation of agentic AI in the Office
Trust Before Autonomy: Why Finance Buyers Reject Agentic AI Read More »