The Three Pillars of AI Transformation: A CFO’s Framework

By Eric Mersch

CFOs have a broad view of how AI is changing technology companies, from investment and returns to governance and product strategy. 

That perspective puts us in a position to help shape decisions and support implementation alongside product, technology, and operations leaders.

After a year of AI transformation work with technology company clients, I’ve found that the strongest approach rests on three connected pillars: AI strategy, AI governance, and workflow automation.

Here’s where CFOs can and should contribute across each of these pillars:

1. AI strategy: Make the product economics work

For tech companies, adding AI functionality to an existing platform raises a commercial question: What will customers pay for, and what will it cost to deliver?

An AI feature may strengthen the product, but greater usage can also increase inference costs. A pricing model that ignores those costs may drive adoption while pressuring gross margin. One that passes every cost through to customers may make the product harder to sell.

CFOs should be involved before pricing is set. Finance can work with product and go-to-market leaders to model different customer-usage patterns, estimate the cost to serve each segment, and test whether pricing holds up as adoption grows. The goal is to understand the economics of the feature, not just the cost of the model behind it.

This is where much of the AI revenue conversation happens. It is also where CFOs need a more consistent seat at the table.

2. AI governance: Give teams visibility and useful guardrails

As AI use spreads, costs become harder to predict. Spending can change with the number of users, the volume of requests, the size of those requests, and the models selected. Finance needs enough visibility to understand what is driving the bill and whether that spending supports valuable work.

Governance works best when finance and IT build it together. 

• IT understands the systems, security needs, and model choices.  
• Finance can help establish budget ownership, reporting, and value measures. 
• Together, they can set guidelines that support adoption while managing cost and risk.

This may include token budgets, inference-cost reporting, and guidance on when to use a more capable model. The goal is to match the right model to the right workload and know what each use case costs.

3. Workflow automation: Look at the work already happening

New AI applications attract attention, especially when employees can quickly create tools for their own teams. But building another internal app is not always the best use of time or money.

Some of the strongest opportunities I’ve seen start with an existing process: information arriving by email, being copied between systems, waiting for review, or requiring someone to prepare the same report each week. These workflows already have users and a clear purpose. That gives teams a practical starting point for measuring improvement.

Before approving a new AI project, ask:

• where work regularly stalls, 
• how much time the current process takes, 
• and what a better process would change. 

Useful measures might include faster turnaround, fewer manual steps, greater capacity, or more consistent output. Those outcomes make it easier to decide whether the automation is worth the investment.

The advantage is in connecting all three

Each pillar supports the others. 

Strategy defines where AI can create customer value and how the company will capture it. 

Governance makes the cost and risk of that work visible. 

Workflow automation turns AI into measurable operating improvements inside the business.

In my experience, many companies are making progress in one of these areas. Far fewer are managing all three together. 

For CFOs, the next step is to identify the weakest pillar and bring the right leaders together around it. That is how AI moves from a collection of projects to a business transformation with economics the company can understand and manage.

Which pillar is your organization furthest behind on? 

If your company is advancing AI initiatives, connect with FLG Partners to discuss how finance leadership can help align strategy, governance, and execution.

–

Frequently Asked Questions

What is the CFO’s role in AI transformation?

The CFO helps connect AI investment to business outcomes. That includes testing the economics of AI features, working with IT on cost visibility and controls, and helping teams assess whether automation improves an existing process. Product, technology, and operations leaders remain essential to those decisions and their implementation.

How should CFOs evaluate the ROI of an AI initiative?

Start with the result the initiative is meant to improve, then compare it with the full cost of delivering and maintaining it. For an AI product feature, that may mean revenue, adoption, inference costs, and gross margin. For an internal workflow, it may mean time saved, faster turnaround, or fewer manual steps. Measure the result after launch, since both usage and costs can change.

Where should a company start with AI workflow automation?

Look for a recurring process with clear steps, frequent delays, and a measurable outcome. Define how the process works today before changing it. That baseline lets the team judge whether automation actually improves turnaround, capacity, or consistency.