AI mandate expectations: high.
Margin for error: zero.
A practical hub for Controllers and finance leaders who've been told to implement AI, but can't afford to get it wrong.
Introduction
Somewhere in the past 18 months, the question changed. It used to be "should we be using AI?" Now it's "why aren't we further along?"
The data backs up the urgency. Deloitte's Q4 2025 CFO Signals survey found 87% of CFOs expect AI to be extremely or very important to their finance department in 2026. But only about 11% are using AI meaningfully in finance, and roughly 35% are still running pilots.
The expectation is near-universal. The execution is hard. We built the AI Mandate Playbook to help you close that gap.
Phase 0
Setting a Foundation
Most teams receiving an AI mandate get stuck early because they jump straight to tool selection before assessing what they actually have to work with.
David Fuhriman, CFO of the Jewish Federation of San Diego, puts it simply: AI is much more a capability to build than a tool to acquire. If AI is a tool, you buy it, install it, and wait for results. If it's a capability, you build the organizational conditions that allow AI to be useful.
David Fuhriman's Five Foundations
| Foundation | What it means in practice |
|---|---|
| Data Readiness | If finding the answer to "How many active customers do we have right now?" requires pulling from multiple systems or debating definitions, the data isn't ready for AI. |
| Process Maturity | Could a new employee follow your workflows from documentation alone? If the answer involves pointing them to a specific person, the process exists as tribal knowledge — and tribal knowledge cannot be automated. |
| Human Capital | AI frees the team to finally get to the work they've been deferring. When that's the frame leadership communicates, the institutional knowledge holders stay engaged. |
| Governance | Good governance requires clear answers: What can AI access? What review applies? Who is accountable when something goes wrong? When do you disclose AI was involved? |
| Technology Infrastructure | This is where most organizations start — but it should actually be the last step. The right technology becomes obvious once the first four foundations are in place. |
Building the Right Team
Three roles tend to accelerate AI work in accounting. On lean teams, one person often covers more than one.
Translates documented workflows into running automations. The technical barrier has dropped; what's grown is the judgment about what to build and maintain.
Ensures deployed automations continue operating correctly over time, accounting for model drift and catching quiet failures. Most finance organizations are underinvested here.
Owns process improvement across functions and creates the operational infrastructure that makes automation stick. Steve Nolan at Public.com describes this as "the track the train runs on."
Another key role worth distinguishing: what Tom Alexander, Partner and Head of AI Innovation & Transformation at CrossCountry Consulting, calls the "transformation athlete." The qualities that make someone effective at AI transformation are largely the same qualities that made them effective at previous transformations — change management, cross-functional coordination, the ability to prioritize a funnel of competing ideas. If you have someone on your team who has demonstrated those qualities before, they are your most valuable asset in executing an AI mandate.
Where to Start Applying AI: Outcomes First
Start with: what does the CFO need to be able to do that they can't today? Then work backward to identify which process constraints are preventing it. Outcome first, process second, tool third.
- Repetitive, rules-based tasks with documented inputs and outputs
- Recurring reconciliations on stable, clean data sources
- Routine AP workflows: invoice coding, payment runs
- Reporting that runs on a fixed cadence with consistent structure
- Exception handling, edge cases, anything that breaks the pattern
- Decisions that require business context AI doesn't have access to
- Variance explanations that require understanding why, not just what
- Controls review and sign-off — the human remains in the loop
Phase 1
Get the Underlying Data Right
"AI doesn't fix bad data. It amplifies it."— David Fuhriman, CFO, Jewish Federation of San Diego
Your data needs to be accessible, clean, and consistently defined before AI can do anything useful with it. Most teams discover during this phase that their data is worse than they thought, and that fixing it is unglamorous work. It's also the work that makes everything else possible.
What Data Readiness Actually Requires
The gap between having data and having usable data comes down to three things:
| Requirement | What it means |
|---|---|
| Accessibility | Data behind a manual CSV export isn't accessible for automation. API access to source systems is the baseline. |
| Consistency | Single sources of truth per entity. Consistent definitions across systems. |
| Documentation | Definitions that exist only in people's heads are not definitions. AI can only work with what is explicitly stated. |
Phase 2
Calibrate Risk to Your Company Complexity
Not all accounting processes carry the same risk. Some can tolerate experimentation. Others require near-zero error tolerance. Phase 2 is about mapping which is which before you commit to anything.
The Readiness Scorecard
A useful tool for upfront calibration is the AI Mandate Readiness Scorecard. It covers seven categories:
| Category | Rating (1–7) | Comments |
|---|---|---|
| Data Readiness | ||
| Process Documentation | ||
| Tool Governance | ||
| Team Readiness | ||
| Leadership Alignment | ||
| Cross-Functional Buy-in | ||
| Iteration Cadence |
1 = strongly disagree · 7 = strongly agree
What Your Score Means
7–20: Stage 1 · Start with the basics 21–34: Stage 2 · Build your plan 35–49: Stage 2–3 · You have a foundation 50+: Stage 3–4 · Ready to scale
Two categories tend to be the most revealing.
Cross-functional buy-in: most finance teams underweight this. A low score doesn't mean the mandate is stuck, but it probably means the next priority is building the steering committee rather than evaluating tools.
Iteration cadence: teams that score low here are waiting for certainty before they move. The antidote is 30-90 day sprints with defined outcome targets.
| Old Approach | New Approach | Why It Matters |
|---|---|---|
| 18-month transformation roadmap | 30-90 day outcome-based sprints | The AI landscape shifts faster than long plans can accommodate |
| Requirements fixed upfront | Iterative requirements that sharpen over time | You don't know what AI can do in your environment until you run it |
| Pain point as the starting question | Business outcome as the starting question | Fixing a pain point optimizes the past |
| Finance-led, finance-owned | Cross-functional steering committee | Finance-only AI produces finance-only results |
Phase 3
Get Proof at Your Scale
By Phase 3, the foundations are in place. Now comes finding tools that can actually deliver in your environment — not just in a demo.
"Every demo you see is very impressive. You have to push on those demos. Initially, it demos great and someone gets excited and they drive the buy decision — but then once you get the software into your toolkit, it's not always delivering on the value prop."— Drew Armanino, COO, Armanino LLP
What Good Vendor Evaluation Looks Like
| Ask the Vendor | What the Answer Tells You |
|---|---|
| Can you show us value within 30-60 days? | Whether their delivery model is genuinely iterative |
| Can we speak with a reference at our scale, in our ERP? | Whether their success stories represent your environment |
| What broke during implementation for that reference? | Whether they're honest about friction |
| Is the team still using it 3 months after go-live? | Whether adoption held |
Phase 4
Build an Experimentation Practice
You have clean data, calibrated risk, and real proof points. Now you can move fast, but only in an environment where mistakes don't hit the general ledger.
"Experimentation" can sound unstructured. In practice, it's the opposite: small, sandboxed automation work that produces real output and builds the internal evidence you need before anything goes to production.
"The thing about coding with the purpose of finding efficiencies is that it's very contagious. Once you start, you cannot stop."— Francisco Meyo, Controller, Abridge
What to Experiment On
- CSV-to-journal-entry automation for a recurring workflow
- Data transformation between systems with known mapping logic
- Custom report built on a clean data export
- Automations replicating a manual process you understand end to end
- Anything that posts to the ledger without human review
- Automations touching data you'd need to explain to an auditor
- Full integrations requiring security or infrastructure oversight
- Workflows where edge cases aren't documented yet
Phase 5
Implement Carefully in the GL
Moving to the GL is a different discipline than experimenting outside it. Phase 5 is where you need rollout plans, fallback procedures, and a real change management strategy. Without all three, adoption won't stick.
"Auditability, observability, repeatability — when you're building agents, those three things are really, really important."— Tom Alexander, CrossCountry Consulting
Cross-Functional Governance
Once automations move to the GL, they stop being finance projects. They touch systems IT never reviewed, data legal hasn't approved, and controls that external auditors will have opinions on.
Cindy Cruz, AI Innovation Lab Leader at CrossCountry Consulting, is direct: "Finance leaders have a responsibility, if they're the ones getting the mandate first, to bring the C-suite together to help create a steering committee."
| Steering Committee Member | What They Need to Own |
|---|---|
| CFO | Strategic outcome definition, cross-functional authority |
| Controller | Implementation plan, phased rollout, surfacing blockers |
| CIO / IT | Vendor access approval, data security, infrastructure |
| Legal | Data handling, third-party AI contracts, regulatory exposure |
| Internal Audit | Control design review, audit trail requirements |
Build vs. Buy
Build vs. Buy — What to Consider
AI has made building cheaper and faster. But that doesn't simplify the decision. It just moves the hard question.
"Feasibility is no longer the main question. I'd say it's more like: how costly will this be to maintain?"— Ben Sheridan, Product Manager, Numeric (former Revenue Accountant, Snowflake)
The teams that will look smart in two years won't be the ones that built the most. They'll be the ones that built the right things and bought the rest from vendors who've already solved for security, audit, and infrastructure at a depth no accounting team should try to replicate.
- Data transformation scripts
- Process automations with known inputs and outputs
- Commissions calculators
- Internal workflow automations
- Tools that enhance data quality
- Payroll tooling
- AI-powered reporting & flux analysis
- Anything with severe legal exposure
Before You Build: Three Questions
1. Who is going to maintain this in a year? If you can't name the person and describe what their maintenance responsibilities would look like, the build decision needs more scrutiny.
2. Ask your vendors what's on their roadmap. Teams routinely build something their existing product vendor is three months away from shipping. A quick conversation before you start building saves weeks of work and avoids the awkward discovery that you've replicated something you're already paying for.
3. Ask whether your process is actually unique. If the answer is yes — custom business logic, competitive workflows you don't want commoditized, data models specific to your organization — build it. If the answer is "we just need this standard thing done," find the vendor who's already built it well.
Closing
What to Bring Back to Your CFO
At some point, the CFO or board that issued the mandate will ask for an update. What you bring to that conversation matters.
The instinct is to show activity: a vendor shortlist, a pilot timeline, a demo that landed well. What actually works is showing judgment. What you evaluated. What you ruled out, and why. What the first deployment will prove, and what it won't.
A Controller who walks in with a phased plan, a risk framework, and a 90-day proof point isn't just responding to a mandate. They're making the case that finance should lead the company's defining initiative of the year.
"Your organization can build these foundations. The real question is whether you will."— David Fuhriman, CFO, Jewish Federation of San Diego
AI made building cheap.
It didn't make the decision easy.
Contributors from Numeric, Abridge, Armanino, and more on how they think about the build vs. buy decision.
