How to Use AI at Work: Real Examples for Finance and Accounting
The real ways I use AI to do my finance job better: building flow charts, drafting decks, debugging formulas, and managing projects.
When I was a finance program manager at Google, the people who got ahead were not the ones who worked the longest hours. They were the ones who got their work done faster and freed up time to think. AI is the biggest lever I have found for that, and I use it every single day. This post is not about using AI to find a job. It is about using AI to do your finance or accounting job better, faster, and with fewer errors. Here are the real applications I rely on and exactly how I use them.
Turn messy processes into clear flow charts
Finance runs on processes, and most of them live only in someone's head. When I need to document a close process, an approval workflow, or a reconciliation procedure, I describe the steps to an AI tool in plain English and ask it to generate a flow chart in a diagram format like Mermaid or to lay out the boxes and arrows for me. I describe who does what, where the handoffs are, and where the controls sit, and in seconds I have a draft I can refine. This is invaluable for SOX documentation, onboarding a new analyst, or showing an auditor how a control actually operates. A few prompts I use:
- Map this month-end close into a swimlane flow chart with columns for staff accountant, manager, and controller.
- Show the three-way match for purchase order, receipt, and invoice as a decision flow with the rejection paths included.
- Convert this written reconciliation procedure into numbered steps with the control point at each stage flagged.
Draft decks and turn analysis into a story
Building a board deck or a monthly business review used to eat an entire afternoon. Now I give the AI my key numbers and the message I want to land, and ask it to draft the slide structure and the talking points. For example, I will paste in the variance summary and say, the takeaway is that we beat revenue but missed on margin because of a supplier price increase, draft six slides that tell that story to a board audience. It gives me a skeleton I can fill with the real charts. The judgment stays mine, but the blank-page problem disappears. I always rewrite the narrative in my own voice and double-check every figure before it goes anywhere.
Use AI as a second set of eyes on the numbers
AI is genuinely useful as a reviewer, as long as you stay in control of the data. I paste a formula and ask why my SUMIFS is returning a value error, or I describe a reconciliation that is off and ask what categories of difference I should check first. It will not balance your books for you, but it is excellent at catching the thing you are too close to see. Some ways I put it to work:
- Explain and debug a complex Excel or Google Sheets formula, or rewrite it to be more readable.
- Draft the logic for a nested formula when I describe the result I want in words.
- List the likely causes when a bank reconciliation or intercompany balance does not tie out.
- Sanity-check the assumptions in a forecast and point out where the model is most sensitive.
Manage projects and your own week
Program management is mostly about keeping many threads moving, and AI helps me stay on top of it. I paste in a wall of meeting notes and ask for a clean list of action items with owners and due dates. I ask it to turn a vague initiative, like improve the close timeline, into a phased project plan with milestones. Before a big stakeholder meeting, I have it generate the likely questions a CFO would ask so I walk in prepared. It is like having a tireless chief of staff who never forgets a follow-up. The structure it provides is often better than what I would build under time pressure.
Write the communication you keep putting off
Finance professionals write constantly, and a lot of it is the kind of writing we avoid. AI clears the backlog. I use it to draft the polite but firm email to a budget owner who is overspending, to rewrite a dense technical accounting memo so a non-finance leader can follow it, and to summarize a forty-page contract into the revenue recognition implications I actually care about. The first draft is never the final draft, but starting from something is far faster than starting from nothing.
The rules I never break
AI is a tool, and using it well in finance means using it carefully. These are the guardrails I hold to without exception:
- Never paste confidential, customer, or material non-public data into a public AI tool. Use your company's approved enterprise version or anonymize first.
- Verify every number and every factual claim it produces, because it will state wrong figures with total confidence.
- Keep the judgment yours. AI drafts and organizes, but the accountability for the work stays with you.
- Check your company policy before you start, since many firms have clear rules about which tools are allowed.
I teach this same material live and for free, walking through these exact use cases with real examples in each session. You can see the upcoming schedule and reserve a spot at summitresume.com/resources.
Want the complete roadmap? Read The Complete Guide to Breaking Into Big Tech Finance.
I'm a former Google finance program manager and the founder of Summit Resume. I have helped 1,400+ finance and accounting professionals land roles at the Big 4 and Big Tech.
Connect on LinkedInWant to turn these skills into offers?
I help finance and accounting pros package their strengths and land competitive roles. Book a free Career Strategy Call.
