Forecasting & FP&A

Reforecast the year: actuals plus remaining forecast

Combine year-to-date actuals with a reasoned forecast for the remaining months, comparing budget, run-rate and adjusted methods line by line.

Time
About 45 minutes
Level
Intermediate
Works with
Claude, ChatGPT, Microsoft Copilot, Gemini

What you need

  • Year-to-date actuals by income statement line, by month
  • The full-year budget by month
  • Known changes: new hires, price increases, lost or won contracts

Steps

  1. Export actuals and budget by month to one table.
  2. List known changes with dates and amounts.
  3. Paste the prompt and review the method chosen for each line.
  4. Enter the final remaining-month forecast in your model. The free workbook's FY Outlook tab does the actuals-plus-forecast math for you.

The prompt

Fill in the highlighted brackets
Help me reforecast full-year results. Actuals are through [month]; we need a forecast for the remaining [number] months.

Data (actuals by month, budget by month, by income statement line):
[paste]

Known changes:
[paste, with dates and amounts]

For each line, show three forecasts for the remaining months side by side:
A. Budget for the remaining months
B. Trailing three-month run rate, excluding one-time items I identify
C. Budget adjusted for the known changes

Recommend A, B or C for each line with a one-sentence reason, and account for seasonality where the budget shows it. Then show the full-year projection versus budget by line, and list the top five risks and opportunities with dollar ranges. Mark every judgment call so I can review it.

Review before you use it

  • One-time items are excluded from run rates.
  • Seasonality in the remaining months is respected.
  • Headcount changes match the hiring plan.
  • The full-year projection is reasonable next to the prior year.

Never paste into an AI tool

  • Employee-level salary data (use department totals)
  • Use a business or enterprise AI plan that does not train on your data, and follow your company's AI policy.

Educational content, not professional advice. AI output can be wrong; review it before relying on it. Disclaimer