Prompt Library

ChatGPT Prompts for Financial Analysts

30 copy-paste prompts

Thirty structured prompts to accelerate modeling, statement analysis, forecasting, valuation, reporting, and scenario work โ€” so you spend less time formatting and more time on judgment. Always verify every figure and treat output as analysis support, not financial advice.

In short: This page contains 30 copy-paste ready prompts, organized into 6 categories with a description and pro tip for each. The first 15 prompts are free instantly โ€” no signup needed. Hand-curated and tested by the AI Academy team.

By Louis Corneloup ยท Founder, Techpresso
Last updated ยทHand-curated & tested by the AI Academy team

Financial Modeling

5 prompts

Three-Statement Model Structure

1/30

<context> You are a senior FP&A analyst building a three-statement model for [COMPANY], a [INDUSTRY] business with [REVENUE] in annual revenue. Historical figures: [FIGURES]. I need a clean, auditable model skeleton in Excel/Google Sheets, not formula automation. </context> <task> 1. List the tab structure (Assumptions, Income Statement, Balance Sheet, Cash Flow, Supporting Schedules). 2. For each statement, lay out the line items in standard order and note which are driver-based vs. calculated. 3. Specify the key links between statements (e.g., net income to retained earnings, capex to PP&E, change in cash to balance sheet). 4. Flag the circular reference risk (interest on revolver) and how to handle it with a switch. 5. List 8 input assumptions I must source myself, with the unit for each. Label every output as requiring my verification against source filings before use. </task>

A tab-by-tab blueprint for an auditable three-statement model with statement links and assumption inputs.

๐Ÿ’ก

Pro tip: Paste your actual historical line items into the [FIGURES] block so ChatGPT mirrors your real chart of accounts instead of a generic template.

Driver-Based Revenue Build

2/30

<context> You are a financial analyst modeling revenue for [COMPANY]. Business model: [BUSINESS MODEL]. Known drivers: [FIGURES] (e.g., units, price, customers, churn, ARPU). I want a bottom-up revenue build I can replicate in a spreadsheet. </context> <task> 1. Decompose revenue into its core drivers for this business model (volume x price, or customers x ARPU x retention). 2. Show the calculation logic for each driver as a written formula, not a number. 3. Identify which drivers are most sensitive and worth isolating as separate assumptions. 4. Suggest a sanity-check ratio to validate the build against reported totals. 5. List the data sources I should pull each driver from. State that all driver values are placeholders I must replace and verify. </task>

A bottom-up revenue decomposition with driver formulas and validation checks.

๐Ÿ’ก

Pro tip: Ask ChatGPT to express formulas in spreadsheet syntax (=B2*B3) so you can paste them directly into your model.

Debt Schedule and Interest Build

3/30

<context> You are building a debt schedule for [COMPANY] inside a three-statement model. Facilities: [FIGURES] (principal, rate, term, amortization for each tranche). I need the schedule logic and how it feeds the other statements. </context> <task> 1. Lay out the debt schedule rows: beginning balance, draws, mandatory amortization, optional prepayment, ending balance. 2. Show how to calculate interest expense (average balance vs. beginning balance methods) and recommend one. 3. Explain how the revolver acts as the cash sweep / plug and the circularity it creates. 4. Detail the links to the income statement (interest), cash flow (financing), and balance sheet (debt balances). 5. Note covenant ratios I should track alongside the schedule. Flag every rate and balance as a placeholder requiring my confirmation against the credit agreement. </task>

A complete debt schedule structure with interest logic and statement linkages.

๐Ÿ’ก

Pro tip: Tell ChatGPT your exact amortization type (straight-line, bullet, sculpted) up front so the schedule rows match your facility terms.

Working Capital Schedule

4/30

<context> You are a financial analyst projecting working capital for [COMPANY]. Historical balances and ratios: [FIGURES] (DSO, DIO, DPO, accrued items). I want a working capital schedule that ties to the cash flow statement. </context> <task> 1. List the working capital line items to model (AR, inventory, AP, prepaid, accrued liabilities, deferred revenue). 2. For each, give the days-based driver formula (e.g., AR = DSO / 365 x revenue). 3. Show how the period-over-period change flows into cash flow from operations. 4. Recommend reasonable benchmark ranges to pressure-test my day assumptions for this industry. 5. Identify the one or two items that move cash the most for this business. Label benchmark ranges as starting points I must validate, not authoritative figures. </task>

A days-driven working capital schedule with cash flow linkage and benchmark checks.

๐Ÿ’ก

Pro tip: Provide ChatGPT three years of historical days metrics so it can comment on whether your forward assumptions are trending realistically.

Model Audit and Error Check

5/30

<context> You are reviewing a financial model for [COMPANY] before it goes to the investment committee. I will describe the model structure and flag areas I am unsure about: [FIGURES] (describe formulas, links, or outputs in question). </context> <task> 1. Produce a model-review checklist covering balance sheet balancing, cash flow tie-outs, sign conventions, and hardcodes inside formulas. 2. List the most common three-statement linkage errors and how to spot each. 3. Suggest stress inputs (zero growth, extreme margin) that quickly expose broken links. 4. Recommend formatting conventions that make audit faster (input vs. formula color coding). 5. Give me 5 specific questions to ask myself about the areas I flagged. Remind me you cannot see my actual spreadsheet, so every check must be performed and verified by me. </task>

A structured model-review checklist for catching linkage and integrity errors before review.

๐Ÿ’ก

Pro tip: Describe suspicious outputs in plain language ("cash goes negative in year 3") and ChatGPT can suggest which links likely broke.

Prompts get you started. Tutorials level you up.

A growing library of 300+ hands-on AI tutorials. New tutorials added every week.

Start 7-Day Free Trial

Ratio & Statement Analysis

5 prompts

Full Ratio Analysis Workup

6/30

<context> You are a financial analyst reviewing [COMPANY]. I will provide income statement, balance sheet, and cash flow figures: [FIGURES]. I want a complete ratio analysis grouped by category. </context> <task> 1. Calculate liquidity ratios (current, quick, cash) and explain what each says for this business. 2. Calculate leverage ratios (debt/equity, debt/EBITDA, interest coverage). 3. Calculate profitability ratios (gross, operating, net margin, ROE, ROA, ROIC). 4. Calculate efficiency ratios (asset turnover, inventory turnover, receivables turnover). 5. Present results in a table with the formula, the value, and a one-line interpretation per ratio. Show your arithmetic for each ratio so I can verify it, and note that I must confirm all input figures against audited statements. </task>

A categorized ratio table with formulas, values, and plain-language interpretation.

๐Ÿ’ก

Pro tip: Ask ChatGPT to show the arithmetic step for each ratio so you can spot-check the math before trusting any number.

Common-Size Statement Analysis

7/30

<context> You are analyzing [COMPANY] using common-size statements. Multi-year figures: [FIGURES]. I want both vertical and horizontal common-size views with commentary. </context> <task> 1. Convert the income statement to vertical common-size (each line as a percent of revenue). 2. Convert the balance sheet to vertical common-size (as a percent of total assets). 3. Produce a horizontal year-over-year percent-change view. 4. Highlight the three line items with the largest shifts and propose hypotheses for each. 5. List the follow-up data I would need to confirm or reject each hypothesis. Frame all hypotheses as questions to investigate, and flag that percentages depend on figures I must verify. </task>

Vertical and horizontal common-size statements with flagged shifts and investigation hypotheses.

๐Ÿ’ก

Pro tip: Feed at least three years of data so ChatGPT can distinguish a one-off blip from a genuine structural trend.

Cash Flow Quality Assessment

8/30

<context> You are assessing earnings quality for [COMPANY] by comparing reported profit to cash generation. Figures: [FIGURES] (net income, operating cash flow, capex, working capital changes, non-cash items). </context> <task> 1. Calculate the cash conversion ratio (operating cash flow / net income) and interpret it. 2. Reconcile net income to operating cash flow and identify the largest reconciling items. 3. Flag any signs of low earnings quality (rising receivables outpacing revenue, capitalized costs, one-time gains). 4. Calculate free cash flow and free cash flow conversion. 5. Summarize whether reported profitability appears backed by real cash, with caveats. </task>

An earnings-quality review reconciling profit to cash with red-flag identification.

๐Ÿ’ก

Pro tip: Ask ChatGPT to rank the reconciling items by size so you immediately see what is driving the gap between profit and cash.

Peer Benchmarking Table

9/30

<context> You are benchmarking [COMPANY] against peers. I will supply the metrics for [COMPANY] and each peer: [FIGURES] (margins, growth, leverage, returns). I want a comparison that surfaces relative strengths and weaknesses. </context> <task> 1. Build a side-by-side table of the supplied metrics for the company and peers. 2. Add a column showing the peer median or mean for each metric. 3. Mark where the company is above, in line with, or below the peer set. 4. Summarize the company's relative positioning in 3 bullet points. 5. Note which metric differences most warrant a deeper look. Do not invent peer figures โ€” use only the values I provide, and remind me to verify them. </task>

A peer comparison table with relative positioning and deeper-dive flags.

๐Ÿ’ก

Pro tip: Explicitly instruct ChatGPT to use only the peer numbers you paste; otherwise it may fill gaps with outdated training data.

Trend and Anomaly Detection

10/30

<context> You are reviewing several years of financials for [COMPANY] to spot trends and anomalies before an earnings discussion. Figures: [FIGURES] (multi-period line items). </context> <task> 1. Calculate the compound annual growth rate for revenue, gross profit, and operating income. 2. Identify line items whose trend direction or rate of change shifts noticeably between periods. 3. Flag any item that moves inconsistently with related items (e.g., revenue up but margin collapsing). 4. Propose a likely operational or accounting explanation for each anomaly as a hypothesis. 5. List the questions to raise with management or in the 10-K notes. Present explanations as hypotheses to test, not conclusions, and flag that all figures need verification. </task>

A multi-year trend and anomaly scan with hypotheses and management questions.

๐Ÿ’ก

Pro tip: After the output, ask ChatGPT to re-rank the anomalies by how material each is to the investment thesis.

Forecasting & Budgeting

5 prompts

Annual Operating Budget Framework

11/30

<context> You are an FP&A analyst building next year's operating budget for [COMPANY], a [INDUSTRY] business. Prior-year actuals and known plans: [FIGURES]. I need a structured budget framework. </context> <task> 1. Outline the budget build order (revenue, COGS, opex by function, capex, headcount). 2. For each major cost category, recommend whether to budget as a percent of revenue, a fixed amount, or a per-headcount figure. 3. List the cross-functional inputs I should collect from each department. 4. Suggest a small set of guardrail metrics to keep the budget realistic (margin, opex growth vs. revenue growth). 5. Recommend a monthly phasing approach for seasonality. Mark every prior-year figure and assumption as requiring my verification. </task>

A structured annual operating budget framework with build order and guardrails.

๐Ÿ’ก

Pro tip: Give ChatGPT your fiscal calendar and seasonality pattern so the phasing recommendations match your real demand curve.

Revenue Forecast with Methods

12/30

<context> You are forecasting revenue for [COMPANY] for the next [PERIOD]. Historical revenue and drivers: [FIGURES]. I want multiple forecasting approaches compared, not a single guess. </context> <task> 1. Describe at least three forecasting methods relevant here (trend extrapolation, driver-based, run-rate annualization). 2. For each method, show the calculation logic using my figures as placeholders. 3. Compare the methods on accuracy assumptions and data requirements. 4. Recommend which method fits this business and stage, with reasoning. 5. Suggest how to triangulate a final number from the methods. Remind me the outputs are projections dependent on my inputs and must be sanity-checked against bookings and pipeline data. </task>

A side-by-side comparison of revenue forecasting methods with a recommended approach.

๐Ÿ’ก

Pro tip: Ask ChatGPT to state the implied growth rate of each method so you can sanity-check it against historical reality.

Expense Forecast and Cost Drivers

13/30

<context> You are forecasting operating expenses for [COMPANY]. Historical opex detail: [FIGURES] (by category and as a percent of revenue). I want a disciplined expense forecast. </context> <task> 1. Classify each cost category as fixed, variable, or semi-variable. 2. Recommend the right driver for each (revenue, headcount, units, fixed escalation). 3. Build the forecast logic per category as written formulas. 4. Flag categories where scaling assumptions are riskiest. 5. Show how total opex and operating margin evolve under the build. Flag all historical percentages as inputs I must confirm before relying on the forecast. </task>

A category-by-category expense forecast classified by cost behavior with driver logic.

๐Ÿ’ก

Pro tip: Tell ChatGPT your hiring plan separately so people-related costs flow from headcount rather than a blunt revenue percentage.

Budget vs. Actual Variance Analysis

14/30

<context> You are preparing a budget-vs-actual variance report for [COMPANY] for [PERIOD]. Budget and actual figures by line: [FIGURES]. I need a clean variance analysis with commentary. </context> <task> 1. Calculate the dollar and percent variance for each line item. 2. Classify each variance as favorable or unfavorable from a profit standpoint. 3. Decompose key variances into volume vs. rate/price components where applicable. 4. Highlight the three variances most material to the period result. 5. Draft neutral, factual commentary for each material variance for management review. Show the variance arithmetic and remind me to confirm both budget and actual figures. </task>

A line-item variance report with favorable/unfavorable classification and decomposition.

๐Ÿ’ก

Pro tip: Ask for a materiality threshold (e.g., flag variances over 5% or $50k) so the report focuses on what actually matters.

Rolling Forecast Update

15/30

<context> You are maintaining a rolling forecast for [COMPANY]. Current forecast, latest actuals, and new information: [FIGURES]. I want to reforecast the remaining periods systematically. </context> <task> 1. Outline a process to roll actuals into the forecast and extend the horizon by one period. 2. Identify which forward assumptions should change given the latest actuals and new information. 3. Quantify the impact of each adjustment on full-year revenue and operating income. 4. Flag assumptions that should stay fixed to avoid over-fitting to one period. 5. Summarize the reforecast story in three bullets for leadership. Label all impact estimates as dependent on figures I must verify. </task>

A systematic rolling-forecast reforecast process with quantified assumption changes.

๐Ÿ’ก

Pro tip: Provide both the prior forecast and the newest actuals so ChatGPT can isolate exactly what changed and why.

Valuation

5 prompts

DCF Valuation Framework

16/30

<context> You are building a discounted cash flow valuation for [COMPANY]. Inputs available: [FIGURES] (projected free cash flows, growth, margins, tax rate, capital structure). I want the full DCF framework and logic. </context> <task> 1. Lay out the DCF steps: project unlevered free cash flow, choose discount rate, discount flows, add terminal value, bridge to equity value. 2. Show the unlevered FCF formula and the discounting formula explicitly. 3. Explain both terminal value methods (Gordon growth and exit multiple) and when each fits. 4. Detail the enterprise-to-equity bridge (net debt, minority interest, etc.). 5. List the 6 assumptions that drive the output most. Show all formulas and remind me the result is a model output requiring verification, not financial advice or a price target. </task>

A complete DCF framework with FCF, discounting, terminal value, and equity bridge logic.

๐Ÿ’ก

Pro tip: Have ChatGPT walk through the discounting formula on one sample year so you can confirm period conventions (mid-year vs. year-end).

WACC Calculation Walkthrough

17/30

<context> You are calculating the weighted average cost of capital for [COMPANY]. Inputs: [FIGURES] (risk-free rate, beta, equity risk premium, cost of debt, tax rate, capital weights). </context> <task> 1. Show the cost of equity calculation via CAPM with my inputs as placeholders. 2. Show the after-tax cost of debt calculation. 3. Show the WACC weighting formula and combine the components. 4. Explain how to choose market-value vs. book-value weights and which is correct. 5. Note the sensitivity of WACC to beta and the equity risk premium. Show each step of arithmetic and flag every input as requiring my own sourcing and verification. </task>

A step-by-step WACC calculation via CAPM with weighting and sensitivity notes.

๐Ÿ’ก

Pro tip: Ask ChatGPT to express the final WACC to one decimal and remind you it is only as good as the beta and ERP you sourced.

Comparable Company Analysis

18/30

<context> You are running a comparable company (trading multiples) analysis for [COMPANY]. Company and peer data: [FIGURES] (EV, equity value, revenue, EBITDA, earnings for each). I want a clean comps framework. </context> <task> 1. Calculate the relevant multiples for each peer (EV/Revenue, EV/EBITDA, P/E). 2. Compute the peer set median and mean for each multiple. 3. Apply the peer multiples to the company's metrics to derive an implied valuation range. 4. Discuss which multiple is most appropriate for this industry and why. 5. Note adjustments needed for comparability (growth, margin, size differences). Use only the figures I provide, show the math, and label the implied range as indicative, not advice. </task>

A trading-comps framework deriving an implied valuation range from peer multiples.

๐Ÿ’ก

Pro tip: Tell ChatGPT to flag any peer that is a poor comp (very different growth or margin) so you can exclude it from the median.

Precedent Transactions Framework

19/30

<context> You are analyzing precedent M&A transactions to value [COMPANY]. Deal data I will supply: [FIGURES] (target metrics and deal values for each transaction). I want a transaction-multiples framework. </context> <task> 1. Calculate transaction multiples for each deal (EV/Revenue, EV/EBITDA) from my data. 2. Note the control premium embedded in transaction multiples vs. trading comps. 3. Compute the median/mean transaction multiple and apply to the company. 4. Discuss how deal vintage and market conditions affect comparability. 5. Derive an implied valuation range and list its key caveats. Use only the deal figures I provide and frame the range as indicative analysis, not a recommendation. </task>

A precedent-transaction framework with control-premium context and an implied range.

๐Ÿ’ก

Pro tip: Ask ChatGPT to call out stale deals; transactions older than a few years often reflect a very different valuation environment.

Valuation Football Field Summary

20/30

<context> You are summarizing multiple valuation methods for [COMPANY] into one view. Outputs from each method: [FIGURES] (DCF range, comps range, precedents range, 52-week range). I want a football-field synthesis. </context> <task> 1. Present each method's valuation range in a structured list (low, midpoint, high). 2. Explain why the ranges differ and what each method emphasizes. 3. Identify where the ranges overlap to suggest a defensible central value. 4. Note which method deserves the most weight for this company and why. 5. List the biggest risks that could move the overall valuation. Label the synthesis as analysis support requiring verification, explicitly not a price target or investment advice. </task>

A multi-method valuation synthesis identifying an overlap range and method weighting.

๐Ÿ’ก

Pro tip: Provide each method's range as low/mid/high and ChatGPT can describe the football field even though it cannot draw the chart.

Reporting & Decks

5 prompts

Monthly Financial Report Narrative

21/30

<context> You are writing the narrative for [COMPANY]'s monthly financial report to leadership. Period results: [FIGURES] (revenue, margin, opex, cash, vs. budget and prior period). Tone: concise, factual, executive. </context> <task> 1. Draft a 3-sentence executive summary of the period. 2. Write a revenue paragraph covering the result, the driver, and the variance to budget. 3. Write a profitability and cost paragraph with the key movements. 4. Write a cash and liquidity paragraph. 5. Close with a forward-looking watch-items section. Keep it factual and neutral, attribute drivers only as supported by the figures, and flag that I must verify all numbers before distribution. </task>

A structured executive narrative for a monthly financial report with watch items.

๐Ÿ’ก

Pro tip: Give ChatGPT both vs-budget and vs-prior-period figures so it frames each movement against the most relevant baseline.

Board Deck Financial Slides

22/30

<context> You are preparing the financial section of a board deck for [COMPANY]. Quarter results and full-year context: [FIGURES]. I want slide-by-slide content, not design. </context> <task> 1. Propose the slide sequence for the financial section (KPI summary, P&L, cash/runway, forecast, asks). 2. For each slide, list the 3-5 data points to show and the one headline takeaway. 3. Write speaker-note talking points for the two most important slides. 4. Anticipate three tough board questions and draft factual answers. 5. Flag any data point that, if weak, the board will probe hardest. Keep all content tied to my figures and remind me to verify every number before the meeting. </task>

Slide-by-slide board deck financial content with takeaways and anticipated questions.

๐Ÿ’ก

Pro tip: Tell ChatGPT your audience (VC board vs. PE board) so it tailors which metrics to lead with, like growth vs. cash efficiency.

KPI Dashboard Definition

23/30

<context> You are defining the financial KPI dashboard for [COMPANY], a [INDUSTRY] business. Available data: [FIGURES]. I want a focused set of metrics with clear definitions. </context> <task> 1. Recommend 8-10 financial KPIs that matter most for this business model. 2. For each KPI, give a precise definition and the exact formula. 3. Note the cadence (daily, monthly, quarterly) and the owner for each. 4. Suggest a sensible target or benchmark range to compare against, flagged as a starting point. 5. Identify the two or three KPIs that should sit at the top as headline metrics. Flag all benchmark ranges as illustrative and dependent on figures I must verify. </task>

A focused financial KPI set with precise definitions, formulas, and cadence.

๐Ÿ’ก

Pro tip: Ask ChatGPT to avoid vanity metrics and justify each KPI by the decision it informs, keeping the dashboard tight.

Earnings Call Talking Points

24/30

<context> You are preparing finance talking points for [COMPANY]'s results review. Period figures and key changes: [FIGURES]. I want clear, factual messaging for an internal or investor audience. </context> <task> 1. Draft a headline result statement covering revenue, margin, and cash. 2. Provide three supporting talking points that explain the drivers. 3. Prepare neutral framing for any negative or below-plan items. 4. Draft answers to the three most likely questions about the results. 5. List metrics or claims I should double-check before saying them aloud. Keep all claims tied strictly to my figures and remind me to verify them; this is messaging support, not guidance. </task>

Factual earnings talking points with driver explanations and Q&A prep.

๐Ÿ’ก

Pro tip: Have ChatGPT flag any statement that could be read as forward-looking guidance so you can soften it before it goes external.

Variance Commentary Writer

25/30

<context> You are writing variance commentary for [COMPANY]'s management report. Line-level variances: [FIGURES] (line, budget, actual, variance, and the cause where I know it). I want concise, consistent commentary. </context> <task> 1. For each material variance, write a one-to-two sentence explanation in a consistent format. 2. Lead each with whether it is favorable or unfavorable and the dollar size. 3. Separate volume-driven from rate/price-driven causes where I supply them. 4. Keep tone neutral and avoid speculation beyond the causes I provide. 5. Group the commentary by statement section for easy reading. Do not invent causes I have not supplied, and flag that all figures need my verification. </task>

Consistent, neutral variance commentary grouped by statement section.

๐Ÿ’ก

Pro tip: Set the format once (e.g., "[Favorable/Unfavorable] $X โ€” cause") and ask ChatGPT to apply it identically to every line.

Go from copy-pasting to actually mastering AI.

AI Academy: 300+ hands-on tutorials on ChatGPT, Claude, Midjourney, and 50+ other tools. New tutorials added every week.

Start Your Free Trial

Scenario & Sensitivity Analysis

5 prompts

Base, Bull, Bear Scenario Build

26/30

<context> You are building three scenarios for [COMPANY]'s plan. Base-case assumptions: [FIGURES]. I want disciplined bull and bear cases derived from the base, not random numbers. </context> <task> 1. List the key assumptions that should differ across scenarios (growth, margin, churn, pricing). 2. For each, propose a directional and magnitude change for bull and bear, with reasoning. 3. Describe the resulting impact on revenue, operating income, and cash for each scenario. 4. Define the trigger events that would push reality toward bull or bear. 5. Recommend which scenario to use for planning vs. stress purposes. Frame all magnitudes as illustrative and dependent on base figures I must verify. </task>

Three coherent scenarios derived from the base case with triggers and impacts.

๐Ÿ’ก

Pro tip: Ask ChatGPT to keep scenarios internally consistent (high growth usually pressures margin early) rather than just toggling one variable.

Sensitivity Table Design

27/30

<context> You are designing a sensitivity analysis for [COMPANY]'s valuation or plan. Output metric and key variables: [FIGURES] (e.g., output = equity value; variables = WACC and terminal growth). </context> <task> 1. Recommend the two variables most worth sensitizing for this output and why. 2. Propose a sensible range and step size for each variable. 3. Describe how to structure the two-variable data table (rows, columns, output cell). 4. Explain how to read the table and which cells represent the key cases. 5. Suggest a second one-variable sensitivity for a third important driver. Note that the table structure is the deliverable and all output values depend on figures I must verify. </task>

A two-variable sensitivity table design with ranges, steps, and reading guidance.

๐Ÿ’ก

Pro tip: Ask ChatGPT for the exact Excel Data Table setup steps so you can build the grid without manual recalculation.

Breakeven and Threshold Analysis

28/30

<context> You are running a breakeven analysis for [COMPANY] or a specific decision. Cost and revenue structure: [FIGURES] (fixed costs, variable cost per unit, price, current volume). </context> <task> 1. Calculate the contribution margin per unit and the contribution margin ratio. 2. Calculate the breakeven volume and breakeven revenue, showing the formulas. 3. Determine the margin of safety at the current volume. 4. Show how breakeven shifts if price or variable cost changes by a given percent. 5. Identify the single lever that most improves breakeven for this structure. Show all arithmetic and flag every cost and price input as requiring my verification. </task>

A breakeven and margin-of-safety analysis with sensitivity to price and cost.

๐Ÿ’ก

Pro tip: Provide your real fixed/variable cost split; ChatGPT cannot guess it, and a wrong split makes the breakeven meaningless.

Downside Stress Test

29/30

<context> You are stress-testing [COMPANY] for a downside scenario for liquidity planning. Base figures: [FIGURES] (revenue, cost structure, cash, debt, covenants). I want a severe but plausible stress. </context> <task> 1. Define a severe downside shock (revenue decline, margin compression) and justify the magnitude. 2. Walk through the impact on operating cash flow and ending cash by period. 3. Identify the period, if any, where cash or a covenant comes under pressure. 4. List the mitigating levers available (cost cuts, capex deferral, financing) and their rough timing. 5. State the single metric to monitor most closely in this scenario. Label the shock magnitudes as illustrative and all outputs as dependent on figures I must verify. </task>

A downside liquidity stress test identifying cash pressure points and mitigations.

๐Ÿ’ก

Pro tip: Ask ChatGPT to time the mitigations (which levers act in 30 vs. 90 days) so the stress test reflects realistic response speed.

Monte Carlo Setup Guidance

30/30

<context> You are setting up a Monte Carlo simulation to understand the distribution of outcomes for [COMPANY]'s key metric. Variables and assumed ranges: [FIGURES] (each uncertain input and its plausible distribution). </context> <task> 1. Recommend which inputs to model as distributions vs. hold fixed, and why. 2. Suggest an appropriate distribution type for each uncertain input (normal, triangular, uniform). 3. Describe the simulation setup conceptually (sampling, iterations, output collection). 4. Explain how to interpret the output distribution (mean, P10/P50/P90, probability of a threshold). 5. Caution where correlation between inputs would distort naive independent sampling. Frame this as setup guidance; all distributions reflect my assumptions and the results require my verification. </task>

Conceptual Monte Carlo setup guidance with distribution choices and output interpretation.

๐Ÿ’ก

Pro tip: Ask ChatGPT to flag which of your inputs are likely correlated so you do not overstate confidence by sampling them independently.

Frequently Asked Questions

ChatGPT can structure models, suggest formulas, calculate ratios from figures you provide, and draft commentary, but it is analysis support, not a substitute for your judgment. It cannot see your live spreadsheet or pull current market data on its own, and it can make arithmetic or logic errors. Always re-perform calculations in your model and verify every number against source filings before relying on it.
No. Nothing ChatGPT produces is financial, investment, or trading advice. These prompts help you build frameworks, run analysis faster, and draft reporting. Valuation ranges and scenarios are indicative model outputs based on the assumptions you supply, not price targets or recommendations. Treat all results as a starting point for your own professional analysis.
The placeholders force you to supply your own verified inputs rather than letting ChatGPT guess. Replace [COMPANY], [FIGURES], [INDUSTRY], and similar brackets with your actual data before sending. This keeps the output grounded in real numbers you control and prevents the model from filling gaps with outdated or invented figures from its training data.
It can apply the correct formulas, but it can still make arithmetic mistakes, especially with long number chains. That is why these prompts ask ChatGPT to show its arithmetic step by step. Use those steps to spot-check the logic, then redo the actual computation in a spreadsheet. Never paste a ChatGPT-calculated figure into a deliverable without independently verifying it.
Check your organization's policy before entering any non-public or sensitive financial data. Consider using an enterprise plan with data controls, anonymizing figures, or working with rounded or disguised numbers when testing prompts. When in doubt, run the analysis structure with placeholder values first, then apply the real numbers inside your own secured tools.

Prompts are the starting line. Tutorials are the finish.

A growing library of 300+ hands-on tutorials on ChatGPT, Claude, Midjourney, and 50+ AI tools. New tutorials added every week.

7-day free trial. Cancel anytime.