The main reason finance pros work with me is because they do not want to feel like they are falling behind.

This feels like a big problem, especially when you see other people in other companies that are using ChatGPT 'better' than you.

But this is the wrong way for you to look at things.

Since 3 years, I've trained 150+ finance teams on AI implementation. Something I’ve noticed is, the top 10% of teams (that are drastically changing the way they work) are doing something different.

They are no longer thinking about AI as a tool, they are thinking about it as a way to redesign how work gets done.

This way of thinking changes finance from task executors, to systems designers.

And if you don't make this change in 2026, you'll spend the next five years watching your team work harder while your competitors work smarter.

Today, I'll show you the 3 specific changes you need to make, and how to implement them.

The Execution Trap

Using AI to create an Excel formula, or using it to help you write a report commentary, or even to help you generate a board presentation written for the right audience?

These are all fine. You are using AI, and you are becoming more efficient.

But nothing has changed in actuality.

This is the execution trap. You are still doing the same tasks. You are just using AI as a slightly better calculator, commentator, or writing assistant.

When you continue to think like a task executor, your time still goes to tasks that don't require judgment: data entry, reconciliation, downloading reports, formatting spreadsheets, writing first-draft commentary etc.

AI won't fix this, because AI doesn't replace bad processes. It will just help you execute a bad process faster.

Meanwhile, CFOs like Saul (who I recently interviewed) have rebuilt their process entirely. He opens Claude Desktop at 7:30am, runs a custom project that searches his calendar and emails, and gets a complete briefing on every meeting, including context on who he's meeting and what they'll discuss. The preparation that used to take 45 minutes now takes 5.

You can watch the full video ‘How This CFO Uses AI to Save 10+ Hours Per Week’ here.

CFO’s like Saul have completely changed the way they think about work.

• Old thinking: "How can AI help me do this variance analysis faster?"
• New thinking: "How do I build a system where variance analysis happens automatically, and I only review exceptions?"

From Execution to Orchestration: Thinking in Systems

When you and your team start thinking in systems, your team stops doing the work. They start designing systems that do the work.

Instead of having AI help YOU execute tasks, the AI executes the tasks FOR you, based on the systems you design.

Take that variance analysis example.

• Old way: analyst spends 4 hours building it every month.
• New way: analyst spends 4 hours once defining the rules. What's material? What needs investigation? When do we escalate? What commentary structure do we use?

Then AI executes those rules every month. The analyst reviews the output in 30 minutes. They focus on exceptions and judgment calls. The system handles everything else.

Same task. Different role. Instead of doing variance analysis, they're designing and validating the variance analysis system.

This is what I mean by orchestration. And here are the 3 changes you need to make to build it:

Change 1: From Task Owners to Process Architects

Your analyst's job isn't to DO variance analysis. It's to DEFINE what variance analysis should look like, then provide insight on what's going on.

Change 2: From "What Tool?" to "What System?"

Before you touch ChatGPT or Copilot, you design the system. What's the input, the logic, the output, the validation? Tools come last, not first.

Change 3: From Individual Expertise to Shared Systems

Right now, your best analyst has all the knowledge in their head, and it disappears when they leave. You need to capture that expertise in systems that anyone can run.

How to Implement in 4-Weeks

The good news is, you can use AI as your consultant to help you design these systems.

So let me give you the roadmap successful finance teams use.

​
Week 1: Use AI to Audit Your Execution Trap

Instead of manually mapping every process, use AI to help you identify where you're stuck.

Open a secure LLM (Copilot, Gemini, ChatGPT Business, Claude Enterprise etc).

Enter this prompt:
​

​

Let AI interview you. Answer its questions and in 30 minutes you should have a clear view of your execution traps and which one to start with.

​

​

Week 2-3: Use AI to Design Your First System

Take the task AI recommended, and use this approach (this is what Saul does):

First, gather your context. As Saul explains: "I have my projects... what I've done here is I've added as files some of our most important strategic documents"

(16:38-17:38 in the YouTube video).

Upload your current process docs, templates, or examples to give AI the context it needs.

Then ask AI:

Now build it (or give it as a project to someone else to build).

Saul took this approach when building his portfolio dashboard:

"A month ago I said to myself you know what I have the data set let me actually do it myself"

Sometimes the quickest win is to build a system when you already have the data and know what you need.

Personal tip: if you don't have enough documentation about a process, go in a call to interview the process owner, activate the transcript and use the transcript to document the process.
​

​

Week 4: Implement Team Learning Sessions

Every week, run a one-hour session.

The format:

1. Someone presents a process they redesigned
2. They show the before/after (what used to take 4 hours now takes 30 minutes)
3. They explain the system they built (the rules, the logic, the validation)
4. Team discusses: What else could we apply this to?

Two things happen in these sessions:

First, your team learns by doing. They see their colleagues thinking like process architects instead of task executors.

Second, knowledge spreads fast. That variance analysis system someone built? Now three other people can adapt it for their own use cases.

This is how you build a culture of orchestration.

​

​

The Bottom Line

Here's what I've learned training 150+ finance teams on AI over the past year.

The ones who struggle keep asking: "Which AI tool should I use for variance analysis? Which one is for forecasting?"

The ones who succeed ask different questions: "What system do I need? What rules should it follow? How do I validate it?"

They’re the ones thinking about using AI as a way to redesign how finance operates.

The finance teams leading in 2026 won't be the ones with the most AI tools. They'll be the ones who redesigned their processes to run systematically instead of manually.

Your Move

So take 30 minutes this week.

Pick one recurring task your team does manually every month. Use the Week 1 prompt from above, and let AI help you see where you're stuck in execution mode.

Then ask yourself: Do I want to keep executing this task faster, or do I want to build a system that runs it automatically?

One approach saves you 20 minutes. The other saves you 5 hours every month.

I know which one the top 10% are choosing.

Best,

Your AI Finance Expert,

Nicolas

Behind The Scenes

I’m working on my next video for you. Stay tuned for what the Lego bricks mean next week ;)

P.S. - What did you think of this edition? Hit reply and let me know (I read all replies).

P.P.S. - The AI Finance Accelerator launches January 2026 - 6 weeks where you will deploy real AI reporting, finance agents, and ML forecasting (not just learn about them). After training 10,000+ finance professionals, we will help you implement AI that you can run the day the cohort ends. 6-8 hours/week. Early-bird pricing for the first 100 only. If you're serious about leading finance in 2026 - Join the waitlist here.