When I started to explore AI tools, the outputs were hit-or-miss, workflows felt clunky, and adoption was something you had to pitch. Today, everything’s different: tools are smarter, memory features work, and AI is showing up inside the platforms finance teams already use.

But despite all this momentum, I still meet finance leaders who know AI is important but don’t know where to begin.

If that sounds familiar—or if you're helping someone else get started—this is the exact roadmap I’d follow today.

Step 1: Get Access to the Right AI Tools

You can’t experiment with AI workflows in finance using limited, unsecured tools. Start with the right setup.

🔹 If you're inside a company:

Push for enterprise access to ChatGPT, Claude, or Microsoft Copilot Use tools aligned with your stack:

Microsoft? → Use Copilot

Google Workspace? → Try Gemini

This gives you secure access and tight integration with your existing workflows.

🔹 If you’re solo or fractional:

Start with ChatGPT Plus and/or Claude Pro.

Bonus tip: Use the desktop app for speed and shortcuts, but the browser version works great, too, when you’re just getting started.

💡 Your organization likely already has some kind of AI access—check with IT or your manager before spinning up anything solo.

Step 2: Learn Your Tool First

Not all tools work the same. ChatGPT, Claude, and Copilot each have their own quirks.

• Use official onboarding and tutorials from the vendor
• Search YouTube for 10-minute demos tied to your actual tools
• Apply what you learn immediately to a real task

Don’t waste time watching theory. Practice!

💡 Search for videos like "best tips and tricks for productivity with ChatGPT" on YouTube. Watch one or two per week, subscribe to creators you trust, and test features they explore—in your actual work.

Step 3: Set Your Usage Policy

Even if your company doesn’t have AI guidelines, you should. A future AI champion should lead by example.

• Never upload sensitive data into public AI tools
• Turn off any data-sharing options in ChatGPT or Claude
• Treat AI outputs as first drafts, not final answers
• Track time saved and quality improved—you’ll need it to make your case for expansion

Some think not having a policy gives more flexibility. In reality, it’s usually the opposite. Responsible leaders self-impose boundaries.

💡Ask AI to help you draft a basic AI use policy for your team. Sometimes, leadership pushback comes from the perception that creating a policy is a big lift. AI can reduce the lift significantly

Step 4: Use AI in the Work You Already Do

Don’t experiment with fake use cases. Start with tasks you already know inside and out.

💡Block 30 minutes each week and title it: “Test 1 task with AI.” Use it on a real deliverable. Track how long it takes and what you’d change.

Step 5: Build Your Personal “AI Stack,” Knowledge Base, and Community

As you gain experience, build a system that works for you.

• Save your best prompts
• Document the outputs that feel “like you”
• List repeatable workflows you want to scale
• Follow AI experts who test new tools so you don’t have to

💡 Create a “Finance AI Knowledge Base” in Google Docs or Notion. Include prompts you like, use cases that worked, links to people, videos, or tools you want to try

AI is a gateway—not just to speed, but to leadership fluency.

Once you understand how AI reasons, what it gets wrong, and where it shines, you’ll ask better questions, make better decisions, and lead your organization through the AI transformation.

When I look back on how I started with AI, I didn’t have a master plan. I just tried one thing, then another. Some things worked, some didn’t, but over time, the dots connected.

So don’t worry about doing it perfectly. Just start learning out loud. The clarity comes as you go.

‍