← All use cases

Use case #1 · CPA

Bookkeeper uses Claude for a loan amortization schedule and employee-level payroll forecast

Erica Goode, Aligned Accountants ·

The short version

What they did

Clean-up loan amortization schedule; payroll forecast by employee from last year’s QB payroll reports (no sensitive data), fed into a Fathom cash flow forecast.

Result (as claimed)

Built the payroll forecast “while I worked on something else”; output “reviewed by human-me”

Prompt

I’m a bookkeeper preparing a payroll forecast for [forecast year]. I’ll attach [last year’s QuickBooks payroll report with employees labeled A, B, C] and [known changes to pay, staffing and benefits]. Employer tax rates and wage limits to use: [paste confirmed rates and limits, or say not yet available].

1. Check that last year’s employee totals agree with the report.
2. Build a 12-month spreadsheet forecast by employee, showing wages, employer taxes and benefits separately.
3. Put assumptions on a separate tab and link calculations to them with formulas. Include monthly totals and totals by employee.

Add a short plain-English note on changes from last year. Ask for missing rates or pay details rather than guessing. Keep unconfirmed assumptions clearly marked so I can review them before using the forecast in cash flow planning.

Do it yourself

These steps are my suggested way in, not Erica Goode’s setup.

What you need

  • A chat assistant that can build spreadsheets, such as Claude or ChatGPT.
  • Last year’s payroll register, with names swapped for Employee A, B, C.
  • The loan’s terms: amount, rate, start date, payment and term.

Start simple

  1. Ask for a loan amortization schedule from the terms. Check the first two payments and the final balance against the lender’s statement.
  2. Paste the Prompt with the de-identified payroll report.
  3. Read the assumptions tab before anything else. Change any rate or raise that is wrong.
  4. Ask it to total the forecast by month and by employee so you can compare with last year.

Take it further

  • Feed the payroll forecast into your cash flow tool, as Goode did with Fathom.
  • Ask for a second version with a new hire or a raise, side by side.
  • Save the prompt and assumptions so next year’s forecast takes minutes.

Check your work

  • Check employer tax rates and wage caps against this year’s official figures. Models often use an old year.
  • Tie last year’s actual totals to the payroll report before you trust the forecast.
  • Keep names, SSNs and pay rates you can match to a person out of it unless you use a business plan of the AI tool.

Before you rely on it

  • This is general information about how accountants use AI, not professional guidance for your engagements.
  • You stay responsible for the work under your professional standards, including review and sign-off. An AI tool’s output is a draft, not evidence.
  • Before you send client data to an AI tool, read the tool’s data and training terms, and get any client consent you need. For tax return information, that includes IRC section 7216.

Source checked. The link, author, date and quote were confirmed against the source. The result itself has not been reproduced. Prompt adapted for this library, inspired by the case. Business size: Small practice. Results are quoted as the source claimed them and never improved. Spot an error or a broken link? Email charlie@cbarmorecpa.com.

More in Advisory/CAS

New cases go out every Saturday in Inside AI.