← Index12 entries · By Task
Section IV · For the Task

Prompts for financial analysis

Prompts for variance analysis, unit economics, and turning financial data into decisions.

§ Overview

Financial analysis is where raw statements become decisions — and the slow part is rarely the math, it's the narrative. Why did expenses come in over plan? Is this variance a timing blip or a real overspend? AI helps by drafting that interpretation quickly: it can explain budget-versus-actual variances and propose root causes to chase, write a plain-English expense narrative that separates timing from true variance, or turn a month-end close into a memo leadership will actually read.

The prompts here span the analyst's range. Some compute and interpret — unit economics, the margin impact of a price change, runway and spend-cut trade-offs. Others critically review, like stress-testing a model's assumptions for aggressiveness and internal consistency. And several are about packaging the answer: an investor update with a candid cash section, a leadership KPI dashboard with exact metric definitions, or the financial slides of a board deck.

The rule that overrides everything: these outputs are analysis aids, and the figures must be verified. A model will narrate a variance confidently whether or not its arithmetic holds, and it can't see the journal entry that explains the anomaly unless you tell it. Let it draft the story and frame the questions; you confirm the numbers and the conclusions.

§ Field Notes

What makes a good prompt for financial analysis

A strong financial-analysis prompt supplies the real figures, the comparison baseline, and the audience for the output. "Explain this variance to the board" needs different framing than the same analysis for the FP&A team. Give the model the categorized numbers, the period, and what decision the analysis feeds — then ask it to distinguish what the data shows from what it's inferring.

The best prompts build in verification and skepticism. Have the model show its formulas and intermediate steps, separate timing differences from genuine over- or underspend, and flag assumptions that look aggressive. Treat the result as a well-structured draft narrative: you keep the framing and questions, then independently confirm every figure before it informs a real decision.

§ Pro Tips

Get sharper results

  • 01Give the model the categorized actuals and the baseline it's comparing against; a variance narrative is only as good as the numbers behind it.
  • 02Ask it to separate timing differences from true over- or underspend — conflating the two is the most common variance-analysis error.
  • 03Have it show formulas and intermediate steps for any metric so you can re-derive the result before quoting it.
  • 04Tell it the audience (board, investors, FP&A) up front; the right level of detail and candor depends entirely on who's reading.
§ FAQ

Common questions

Should I verify the figures in an AI financial analysis?

Always. The output is an analysis aid, not an audited result. Language models can make calculation errors and will narrate them just as confidently as correct ones. Re-derive key metrics yourself and reconcile to source statements before the analysis informs any decision.

Can AI tell me why a variance happened?

It can propose plausible root causes worth investigating, but it can't see your transactions. Use it to generate a prioritized list of hypotheses and the questions to ask, then confirm the actual driver against your ledger and operational context.

How do I make a financial summary land with executives?

Tell the model the audience and lead with takeaways. Ask for headline-first commentary that states the result versus plan, the few drivers that matter, and the open items — then supports each with data, rather than walking through every line.

§ The Prompts · 12
№ 130finance

Update a 12-Month Rolling Forecast with New Actuals

Refreshes a 12-month rolling forecast using the latest closed-period actuals and updated business assumptions.

For
chatgpt·claude
№ 131finance

Explain Budget vs. Actual Variances for the Period

Computes and prioritizes material budget-versus-actual variances and proposes plausible root causes to investigate.

For
chatgpt·claude
№ 132finance

Break Down Unit Economics: CAC, LTV, and Payback

Calculates and interprets CAC, LTV, LTV:CAC, and payback period from acquisition and retention inputs.

For
chatgpt·claude
№ 133finance

Draft a Monthly Investor Financial Update

Turns monthly metrics, wins, and challenges into a candid, investor-grade financial update with a clear cash and runway section.

For
chatgpt·claude
№ 134finance

Write a Plain-English Expense Variance Narrative

Converts categorized expense variances into a readable monthly commentary that separates timing from true over- or underspend.

For
chatgpt·claude
№ 135finance

Model Best, Base, and Worst Cash-Flow Scenarios

Builds best, base, and worst cash-flow scenarios with period-by-period balances and flags when cash could turn negative.

For
chatgpt·claude
№ 136finance

Analyze Cash Runway and Model Spend-Cut Options

Calculates current runway and zero-cash date, then models how specific spend-cut options extend runway with their trade-offs.

For
chatgpt·claude
№ 137finance

Analyze Pricing Changes and Their Margin Impact

Quantifies the margin and break-even impact of a price change and stress-tests it against several volume scenarios.

For
chatgpt·claude
№ 138finance

Stress-Test the Assumptions Behind a Financial Model

Critically reviews a financial model's assumptions for aggressiveness, sensitivity, and internal consistency.

For
chatgpt·claude
№ 139finance

Design a Finance KPI Dashboard for Leadership

Specifies a grouped, audience-appropriate finance KPI dashboard with exact metric definitions and a layout plan.

For
chatgpt·claude
№ 140finance

Summarize Month-End Close Results for Leadership

Produces a skimmable month-end close memo covering results versus plan, balance-sheet moves, adjustments, and open items.

For
chatgpt·claude
№ 141finance

Draft the Financial Section of a Board Deck

Outlines the financial slides of a board deck with takeaway headlines, supporting data points, and anticipated-question talking points.

For
chatgpt·claude