23+ Best AI Prompts for Entrepreneurs

Most AI prompts written for entrepreneurs are motivational filler — "act as a business coach and help me scale." They produce advice you could have got from a LinkedIn post, because they give the model nothing to work with.

The prompts below do the opposite. Every one of them requires you to bring real inputs: your actual numbers, your actual costs, your actual retention curve. That is the entire difference. An LLM reasoning over your P&L is genuinely useful; an LLM guessing what a business like yours might look like is worse than useless, because the output is confident and wrong.

There are 23 prompts here across six domains — planning, cash management, financial analysis, KPIs, investment decisions, and reporting. They are built for the work a founder actually does between raising and reporting.

Before You Use These

These are analysis and thinking tools, not financial, legal, or investment advice. An AI model cannot audit your books, does not know your jurisdiction's tax treatment, and will produce a plausible number when it does not have a real one. Use these to structure your thinking and pressure-test your reasoning — then verify anything material with your accountant, and treat investment decisions as yours alone.

That is not a disclaimer for its own sake. It shapes how the prompts are written: nearly every one below instructs the model to separate what it calculated from what it assumed, and to say plainly when your data is insufficient to answer.

Five Rules That Make These Work

1. Paste real data or do not bother. Every prompt has a data block. Fill it with actual figures, even messy ones. A model working from your real revenue split gives you something; a model inventing a plausible SaaS business gives you nothing.
2. Force the assumption list. The dangerous output is the one where an assumption is buried inside a conclusion. Every prompt below demands assumptions be stated separately and marked as assumptions.
3. Ask what would change the answer. A number without a sensitivity is a false comfort. "Which input, if wrong by 20%, breaks this conclusion?" is the most valuable question in financial analysis.
4. Demand the case against. Models are agreeable by default. If you do not explicitly ask for the strongest argument that you are wrong, you will not get it.
5. Never let it silently fill a gap. Instruct the model to say "I do not have this" rather than estimate. This single instruction removes most of the risk in financial prompting.

Business Planning and Strategy (Prompts 1–4)

1. The Business Model Stress Test

Finds the assumption your whole model rests on before the market finds it for you.

Act as a skeptical operating partner reviewing my business model. Your job is to find where it breaks, not to encourage me.

BUSINESS:
- What we sell: [PRODUCT/SERVICE]
- Who buys it: [CUSTOMER SEGMENT AND HOW THEY BUY]
- Price and model: [PRICING AND WHETHER ONE-OFF, SUBSCRIPTION, USAGE]
- How we acquire customers: [CHANNELS AND ROUGH COST]
- Current scale: [REVENUE, CUSTOMER COUNT, GROWTH RATE, TEAM SIZE]
- What we believe makes us defensible: [YOUR ANSWER]

Deliver:
1. The three to five load-bearing assumptions this model depends on — the ones where, if wrong, the business does not work
2. For each: how confident I should be, what evidence would confirm or kill it, and the cheapest test to run in the next 30 days
3. The single assumption most likely to be wrong, and why
4. The failure mode I am least likely to see coming, given how I described the business
5. Two structurally different models that could serve the same customer, and what each would require

Rules: separate what I told you from what you inferred. Where you lack information, say so rather than estimating. Do not give me encouragement or a summary of my strengths.

2. The Unit Economics Teardown

The maths that determines whether growth makes you richer or poorer.

Analyse my unit economics and tell me whether this business gets healthier or sicker as it grows.

DATA (last [PERIOD]):
- Revenue per customer: [FIGURE, and whether monthly, annual, or one-off]
- Direct cost to serve one customer: [FIGURE — list what is included]
- Total sales and marketing spend: [FIGURE]
- New customers acquired: [NUMBER]
- Customer churn: [RATE AND OVER WHAT PERIOD]
- Gross margin: [FIGURE OR "calculate it"]
- Fixed overhead per month: [FIGURE]

Calculate and show your working for:
1. Contribution margin per customer
2. Customer acquisition cost, and what it excludes that it probably should include
3. Lifetime value, using a method you name and justify — and state its weaknesses
4. Payback period in months
5. The LTV/CAC ratio, with an honest note on how much weight it deserves at my stage

Then tell me:
- Whether each additional customer currently makes the business better or worse off, and at what volume that changes
- Which single input is most fragile, and what happens to the conclusion if it is 20% worse than stated
- The two levers that would most improve these economics, ranked by effort against impact

Label every figure as either CALCULATED from my data or ASSUMED by you. Do not fill missing inputs with industry benchmarks unless I ask — flag the gap instead.

3. The Strategic Options Analysis

For decisions where the alternatives are genuinely different rather than variations of one plan.

I am choosing between strategic directions. Analyse them as an investment committee would.

SITUATION: [WHERE THE BUSINESS IS AND WHAT FORCED THIS DECISION]
CONSTRAINTS: [CAPITAL AVAILABLE, RUNWAY, TEAM CAPACITY, ANY HARD DEADLINES]

OPTIONS:
A: [DESCRIBE]
B: [DESCRIBE]
C: [DESCRIBE — or "identify a third option I have not considered"]

For each option, give me:
1. What has to be true for it to work, stated as testable claims
2. Capital and time required before it produces evidence of working
3. What it costs me in options forgone — what doors it closes
4. The realistic downside case, not the worst imaginable one
5. How reversible it is, and at what point it stops being reversible

Then:
- Score each on expected value, downside protection, and speed to evidence, and show the scoring
- State which you would choose and why, in one paragraph
- Give the strongest argument against your own recommendation
- Name the one piece of information that would most change your answer, and how I could get it

Do not hedge into "it depends". Commit to a recommendation and expose your reasoning.

4. The Competitive Position Audit

An honest read on whether your advantage is real or a story you tell investors.

Audit my competitive position honestly.

MY BUSINESS: [WHAT YOU DO, FOR WHOM, AT WHAT PRICE]
CLAIMED ADVANTAGE: [WHAT YOU BELIEVE MAKES YOU HARD TO COMPETE WITH]
COMPETITORS: [NAME THEM AND WHAT THEY DO DIFFERENTLY]
CUSTOMER SWITCHING: [WHAT IT COSTS A CUSTOMER TO LEAVE — time, money, data, habit]

Assess:
1. Whether my claimed advantage is a real moat, a temporary lead, or a preference — be blunt about which
2. How long it would take a well-funded competitor to neutralise it, and what they would have to do
3. Where I am actually differentiated versus where I only believe I am
4. Which competitor is most dangerous to me and why — it may not be the largest
5. What would have to happen in this market for my position to become worthless

Then give me the two moves that would most strengthen my position, with what each costs.

Base this only on what I have told you plus general reasoning about market structure. Do not assert facts about named competitors that you cannot verify — mark anything uncertain as an assumption I need to check.

Cash Flow and Financial Management (Prompts 5–8)

Profit is an opinion; cash is a fact. These four are the ones that keep a business alive.

5. The 13-Week Cash Flow Forecast

The standard operating tool when cash is tight — and the one most founders build too late.

Build me a 13-week rolling cash flow forecast.

OPENING POSITION:
- Cash on hand today: [FIGURE]
- Undrawn credit available: [FIGURE OR "none"]

INFLOWS:
- Confirmed receivables with expected dates: [LIST: amount, expected date, and how reliable each payer is]
- Recurring revenue: [AMOUNT AND COLLECTION TIMING]
- Expected new sales: [AMOUNT, TIMING, AND YOUR CONFIDENCE]

OUTFLOWS:
- Payroll: [AMOUNT AND DATES]
- Fixed costs: [RENT, SOFTWARE, INSURANCE — with dates]
- Variable costs: [WHAT DRIVES THEM]
- Tax and statutory payments due: [AMOUNTS AND DATES]
- Debt service: [AMOUNTS AND DATES]
- Known one-offs: [LIST]

Produce:
1. A week-by-week table: opening cash, inflows, outflows, closing cash
2. The lowest cash point across the 13 weeks, and which week it falls in
3. Any week where cash goes negative, flagged clearly
4. A stressed version where receivables arrive two weeks late and new sales come in 30% under plan
5. The three specific actions that would most improve the low point, with how much each buys and how quickly

Treat receivables as arriving when they realistically arrive, not when they are due. State every timing assumption you made. If my inputs are insufficient to build a reliable week, say which data I need to supply.

6. The Runway and Burn Analysis

How long you have, and what actually changes it.

Calculate my runway and tell me what genuinely extends it.

CURRENT POSITION:
- Cash: [FIGURE]
- Monthly revenue: [FIGURE, and growing/flat/declining at what rate]
- Monthly costs broken down: [PAYROLL / MARKETING / INFRASTRUCTURE / OTHER]
- Committed future costs not yet in the run rate: [LIST]
- Expected inflows outside revenue: [GRANTS, TAX CREDITS, FUNDING — with confidence]

Give me:
1. Gross burn, net burn, and current runway in months — show the calculation
2. Runway under three cases: revenue flat, revenue grows at current rate, revenue declines 20%
3. The date I would need to start fundraising to close before cash runs out, working backwards from a realistic raise timeline you state
4. A ranked list of cost reductions by months of runway bought per unit of damage done to the business
5. Which costs look reducible but are not, and why

Then answer directly: at what point do I stop having a choice about what to cut?

Do not recommend cutting anything without naming its consequence. State whether you are treating my revenue growth as fact or assumption.

7. The Expense Teardown

Systematic cost review that goes beyond cancelling subscriptions.

Review my cost base and find what should not be there.

COSTS (last [PERIOD], itemised):
[PASTE YOUR EXPENSE LINES WITH AMOUNTS AND WHAT EACH IS FOR]

CONTEXT:
- Revenue: [FIGURE]
- Headcount and roles: [LIST]
- What we are trying to achieve in the next 6 months: [GOAL]

Categorise every line as:
- ESSENTIAL — the business stops or breaks without it
- GROWTH — directly generates revenue, with the evidence for that
- INSURANCE — protects against a specific risk, name the risk
- LEGACY — made sense once, unexamined since
- WASTE — no longer serves a purpose

For each GROWTH item, ask whether the return is measured or assumed, and flag the unmeasured ones.

Then give me:
1. Total in each category, as amount and percentage of costs
2. The five lines with the worst ratio of cost to value, with your reasoning
3. What I could cut this month with no operational impact, and the total
4. What looks expensive but is genuinely load-bearing — the things I should not touch
5. Any cost that appears missing for a business of this type and stage

Do not assume a line is waste because you do not recognise it. Ask me instead.

8. The Pricing Change Model

What a price move actually does once churn and volume respond.

Model the full impact of a price change before I make it.

CURRENT STATE:
- Price: [FIGURE AND MODEL]
- Customers: [NUMBER, split by segment if relevant]
- Monthly churn: [RATE]
- Gross margin: [PERCENTAGE]
- New customers per month: [NUMBER] at [CAC]

PROPOSED CHANGE: [NEW PRICE, WHO IT APPLIES TO, AND WHETHER EXISTING CUSTOMERS ARE GRANDFATHERED]

Model:
1. Immediate revenue impact if nothing else changes
2. Break-even churn — how many customers I can lose before the increase costs me money
3. Break-even volume — how many fewer new customers I can afford to acquire
4. The effect on payback period and unit economics
5. The 12-month picture under three elasticity assumptions you state explicitly

Then tell me:
- What signals in my current data suggest whether customers are price-sensitive
- Which customer segment is most at risk and what they have in common
- How to test this on a subset before committing
- What I should measure in the first 60 days, and what result would tell me to reverse it

Be explicit that elasticity is the assumption doing the heavy lifting here, and that you cannot know it from my data. Give me the range where the decision changes.

Financial Analysis and Modelling (Prompts 9–12)

9. The Three-Statement Model Skeleton

Builds the structure and, more importantly, the assumption sheet behind it.

Help me build a three-statement financial model for my business.

BUSINESS: [WHAT YOU DO AND HOW REVENUE IS EARNED]
HISTORICAL DATA: [PASTE WHAT YOU HAVE — revenue, costs, headcount by month or quarter]
PLANNING HORIZON: [MONTHS OR YEARS]

Produce:
1. The driver list — every input the model needs, grouped into revenue drivers, cost drivers, and working capital assumptions
2. For each driver: where the number should come from, and whether it is observable, estimable, or a guess
3. The P&L structure appropriate to my business, with the line items that actually matter for it
4. The cash flow bridge from profit to cash, naming the specific working capital items that will move for a business like mine
5. Balance sheet items that will change materially, and what drives each
6. The formulas linking the three statements, written out

Then flag:
- Which drivers my historical data can support and which I would be inventing
- The two or three drivers that will determine the entire output, so I know where to spend my accuracy budget
- Common modelling errors specific to this business type

Do not produce forecast numbers. Build the structure and tell me what I need to supply.

10. The Scenario and Sensitivity Analysis

Turns one forecast into a range, which is what a forecast should have been.

Run scenario and sensitivity analysis on my plan.

BASE CASE ASSUMPTIONS:
[LIST EACH ASSUMPTION AND ITS VALUE — growth rate, churn, CAC, margin, hiring, price, anything else material]

OUTPUT I CARE ABOUT: [e.g. cash at month 18, EBITDA at year end, months of runway]

Deliver:
1. A sensitivity table showing how the output moves when each assumption varies by minus 30%, minus 15%, plus 15%, plus 30% — one at a time
2. The assumptions ranked by how much they move the output, so I know which ones matter
3. Three coherent scenarios — downside, base, upside — where the assumptions move together in ways that would actually correlate, not independently
4. For the downside case: when I would know I am in it, and which early indicator fires first
5. The combination of assumptions that produces failure, and an honest assessment of how plausible that combination is

Then answer: which single number should I be watching weekly, given that it drives the most outcome variance?

State clearly which correlations between assumptions you assumed, and why. Do not present a single-point forecast anywhere.

11. The Break-Even Analysis

Where the business stops consuming cash, expressed in units you can act on.

Calculate my break-even position and what it would take to reach it.

DATA:
- Fixed costs per month: [FIGURE, itemised]
- Variable cost per unit or per customer: [FIGURE]
- Price per unit or per customer: [FIGURE]
- Current volume: [NUMBER]
- Any step costs — costs that jump at certain volumes: [DESCRIBE]

Calculate, showing your working:
1. Contribution margin per unit and as a percentage
2. Break-even volume and break-even revenue
3. How far current volume is from break-even, in units and in percentage terms
4. Where step costs change the picture, and the effective break-even after each step
5. Margin of safety at current volume

Then give me:
- The three routes to break-even — more volume, higher price, lower variable cost, lower fixed cost — quantified for my numbers, with what each would require
- Which route is most realistic given what I have told you, and why
- The time to break-even at current growth rates

Flag any cost I have classified as fixed that is probably variable, or vice versa — misclassification is the most common error in this calculation.

12. The Financial Statement Interpreter

For reading your own accounts the way an outsider would.

Read my financial statements and tell me what they say about this business.

STATEMENTS: [PASTE P&L, BALANCE SHEET, AND CASH FLOW — as many periods as you have]
CONTEXT: [INDUSTRY, STAGE, ANYTHING UNUSUAL THAT HAPPENED IN THESE PERIODS]

Give me:
1. The three things a sophisticated outsider would notice first — good or bad
2. Trend analysis on the lines that matter: revenue quality, gross margin direction, cost growth relative to revenue growth, working capital movement
3. The key ratios appropriate to my business, calculated with working shown, and what each says
4. Where profit and cash diverge, and the specific reason
5. Anything that looks unusual and would prompt a question in diligence — with the innocent explanation as well as the concerning one
6. What these statements do not tell you that you would need to form a real view

Rules: do not congratulate me. Do not soften findings. If a trend is ambiguous, say it is ambiguous rather than picking a narrative. Distinguish clearly between what the numbers show and what you are inferring.

KPIs and Growth Tracking (Prompts 13–16)

13. The KPI Tree

Connects the one number that matters to the things people can actually change.

Build me a KPI tree for my business.

BUSINESS: [MODEL AND HOW YOU MAKE MONEY]
STAGE AND SIZE: [REVENUE, HEADCOUNT, GROWTH RATE]
STRATEGIC PRIORITY FOR THE NEXT 12 MONTHS: [THE ONE THING]
CURRENT METRICS TRACKED: [WHAT YOU MEASURE TODAY]

Produce:
1. The single north star metric for this business at this stage, with the argument for it over the obvious alternatives
2. The tree beneath it: three to five primary drivers, each broken into the inputs that move them, down to the level where a specific person can act
3. For every metric: definition, calculation, measurement frequency, and who owns it
4. Which metrics are leading indicators and which are lagging, marked clearly
5. Which of my current metrics I should stop tracking, and why they are noise or vanity

Then flag:
- Any metric in the tree that is easy to game, and how it would be gamed
- Where two metrics could be improved at each other's expense, so I watch them as a pair
- The metric I am most likely to be missing entirely

Keep the tree to a size a small team can actually maintain. More metrics is not better.

14. The Cohort Retention Analysis

Retention by cohort is where you find out whether the product is improving.

Analyse my cohort retention data.

DATA: [PASTE COHORT TABLE — customers acquired by month, and how many remained in each subsequent month]
CONTEXT: [WHAT CHANGED OVER THIS PERIOD — pricing, product, channels, target customer]

Tell me:
1. How retention is trending across cohorts — are newer cohorts better, worse, or flat, and by how much
2. Where the curve flattens, and what percentage survives to that point
3. Whether there is a stable retained base, and what it looks like
4. Which cohort performed unusually well or badly, and what was different about that period
5. What the shape of these curves implies about product-market fit, stated carefully

Then:
- Estimate lifetime value from actual retention rather than an assumed churn rate, and explain how the two differ
- Identify the single point in the customer lifecycle where intervention would most improve the curve
- Tell me what a healthy curve looks like for this type of business, and how far I am from it

If the cohorts are too small or too recent to draw conclusions from, say so and tell me how much more data I need. Do not extrapolate from thin cohorts.

15. Growth Decomposition

Separates real growth from growth you happened to receive.

Decompose my revenue growth into its actual components.

DATA:
- Revenue by month for [PERIOD]: [PASTE]
- Customer counts by month: [PASTE]
- New, churned, expanded, and contracted revenue by month if available: [PASTE OR "not tracked"]
- Price changes during the period: [LIST]
- Marketing spend by month: [PASTE]

Break the growth down into:
1. New customer revenue
2. Expansion from existing customers
3. Revenue lost to churn and contraction
4. Price effect versus volume effect
5. Net revenue retention, calculated and explained

Then tell me:
- Which component is actually driving growth, and whether that is a durable source
- What growth would have been without price increases
- Whether growth is becoming more or less efficient — spend against incremental revenue over time
- The component most at risk of stalling first, and what would signal it starting

If I have not given you the data to split something out, tell me exactly what to start tracking rather than approximating it.

16. The Weekly Metrics Review

Turns a dashboard into a decision meeting.

Design my weekly metrics review, then run this week's.

BUSINESS CONTEXT: [MODEL, STAGE, TEAM SIZE]
CURRENT QUARTER GOAL: [SPECIFIC TARGET]
THIS WEEK'S NUMBERS: [PASTE YOUR METRICS]
LAST FOUR WEEKS FOR COMPARISON: [PASTE]

First, define the review:
1. The six to eight metrics that belong in a weekly review at my stage, and why each earns its place
2. For each: what a normal week looks like, and what threshold should trigger discussion rather than acknowledgement
3. The 30-minute agenda structure

Then run this week:
1. Which metrics are outside their normal range, ranked by how much it matters
2. For each anomaly: the two or three most likely explanations, and what I could check to distinguish between them
3. What is quietly drifting — the slow trend that will not trigger a threshold but matters over a quarter
4. The one decision this week's data actually calls for
5. What to ignore this week, so the meeting stays short

Distinguish between statistically meaningful movement and normal weekly variance. Most weekly changes are noise; say so when they are.

Investment and Portfolio Analysis (Prompts 17–20)

These deal with allocating capital — your own or the business's. They are structured to expose reasoning rather than produce recommendations, for the reason stated at the top of this article.

17. The Capital Allocation Framework

Where the next available money should go, argued rather than assumed.

Help me decide where to allocate my next [AMOUNT] of available capital.

BUSINESS POSITION: [REVENUE, GROWTH, MARGIN, RUNWAY]
CANDIDATE USES:
A: [e.g. hire into sales — cost, expected effect, time to effect]
B: [e.g. product development — cost, expected effect, time to effect]
C: [e.g. paid acquisition — cost, expected effect, time to effect]
D: [e.g. hold as reserve]

CONSTRAINTS: [ANYTHING FIXED — obligations, commitments, minimum cash]

For each option:
1. Expected return, stated as a range with the reasoning, not a point estimate
2. Time until I would know whether it worked
3. What it costs if it fails — money, time, and second-order damage
4. How reversible the commitment is
5. Whether it compounds or is one-off

Then:
- Rank them on risk-adjusted return, showing how you weighted the factors
- Say what you would do with the full amount, including splitting it if that is right
- Make the case for holding it all as reserve instead, honestly
- Name the condition under which your recommendation flips

Treat the reserve option seriously rather than as a default no. Distinguish clearly between my stated expected effects and your assessment of them.

18. The Investment Due Diligence Checklist

For evaluating an acquisition, a partnership, or putting money into something outside your own business.

Build me a due diligence framework for this opportunity, and apply it to what I know so far.

OPPORTUNITY: [WHAT IT IS — acquisition, investment, major partnership]
WHAT I KNOW: [EVERYTHING YOU HAVE — financials, terms, people, market]
WHAT I AM BEING ASKED TO COMMIT: [MONEY, TIME, EQUITY, OBLIGATIONS]
MY THESIS: [WHY YOU THINK THIS IS GOOD]

Produce:
1. The diligence checklist appropriate to this type of opportunity, organised by financial, commercial, legal, operational, and people
2. For each item: what good looks like, what a red flag looks like, and how to verify it independently
3. Which items are deal-breakers versus price-adjusters
4. Applied to what I have told you: what is answered, what is unanswered, and what is concerning
5. The three questions I should ask next, and who I should ask them of rather than the counterparty

Then:
- Attack my thesis. Give me the strongest version of the case that this is a bad idea
- Identify what I appear to be assuming without evidence
- Tell me what a professional buyer would look at that I have not mentioned

Do not value the opportunity or tell me whether to proceed — that requires verified information and professional advice. Tell me what I need to establish before I can decide.

19. The Portfolio Review

Structured review of holdings, focused on concentration and reasoning rather than picks.

Review the structure of my portfolio and identify the risks in how it is built.

HOLDINGS: [LIST EACH WITH APPROXIMATE ALLOCATION PERCENTAGE AND ASSET TYPE]
CONTEXT:
- Time horizon: [YEARS]
- Purpose of this capital: [WHAT IT IS FOR]
- Income source outside the portfolio: [YOUR BUSINESS — including which sector and how correlated]
- Liquidity I may need and when: [DESCRIBE]

Analyse:
1. Concentration — where allocation is clustered, whether by asset, sector, geography, or currency
2. Correlation risk, including correlation with my income source and my business — this is the exposure founders most often miss
3. Liquidity profile against my stated liquidity needs
4. Whether the structure matches the stated time horizon and purpose
5. Any holding whose role in the portfolio is unclear from what I have told you

Then:
- Describe what a structurally sound allocation looks like for this horizon and purpose, in general terms
- Identify the largest structural gap between that and what I hold
- Give me the questions I should be asking a qualified adviser

Do not recommend specific securities, predict returns, or tell me to buy or sell anything. Analyse structure and risk only. Flag where my position is concentrated in the same risk as my business income, since that is the exposure my situation makes likely.

20. The Build, Buy, or Partner Analysis

Analyse whether I should build, buy, or partner for this capability.

CAPABILITY NEEDED: [WHAT AND WHY]
STRATEGIC IMPORTANCE: [IS THIS CORE TO DIFFERENTIATION OR SUPPORTING]
BUILD: [ESTIMATED COST, TIME, AND WHO WOULD DO IT]
BUY: [WHAT IS AVAILABLE AND AT WHAT PRICE]
PARTNER: [WHO, AND ON WHAT TERMS]
CONSTRAINTS: [CAPITAL, TIME, TEAM CAPACITY]

For each route:
1. True total cost including the parts usually forgotten — maintenance, integration, management overhead, opportunity cost of the team's attention
2. Time to working capability, and how confident that estimate should be
3. What control I retain and what I give up
4. Switching cost if it goes wrong
5. Effect on the defensibility of my business

Then:
- Recommend a route, with the reasoning exposed
- State the conditions under which the answer would be different
- Flag the classic error for this specific decision type

Be direct that build estimates are systematically optimistic, and apply an honest adjustment — state what adjustment you applied and why.

Reporting and Communication (Prompts 21–23)

21. The Monthly Investor Update

The discipline that keeps investors useful rather than anxious.

Write my monthly investor update.

PERIOD: [MONTH]
METRICS: [REVENUE, GROWTH, CASH, RUNWAY, AND YOUR CORE KPIS — this month and last]
WHAT WENT WELL: [LIST]
WHAT DID NOT: [LIST — be honest here, the model can only work with what you give it]
KEY DECISIONS MADE: [LIST]
WHAT I NEED FROM INVESTORS: [INTRODUCTIONS, ADVICE, HIRING HELP — be specific]

Write an update that:
1. Opens with the two or three numbers that matter, and the direction of each
2. States the honest position in the first paragraph, good or bad — no burying
3. Covers what went wrong with the same specificity as what went right, including what I learned and what I changed
4. Makes the asks concrete enough to act on — a named type of person, a specific company, a defined question
5. Ends with the top priorities for next month

Constraints: under 600 words. No adjectives doing the work of evidence. No spin. If the month was bad, the update should read as a bad month handled well, not as a good month.

Flag any claim in my inputs that I should be able to substantiate if an investor asks.

22. The Board Deck Builder

Build the structure and narrative for my board meeting.

MEETING PURPOSE: [ROUTINE REVIEW / DECISION REQUIRED / DIFFICULT NEWS]
PERIOD PERFORMANCE: [KEY NUMBERS AGAINST PLAN]
DECISIONS I NEED FROM THE BOARD: [LIST]
ISSUES I WOULD RATHER NOT DISCUSS: [BE HONEST — these are usually the ones that matter]
BOARD COMPOSITION: [WHO IS IN THE ROOM AND WHAT EACH CARES ABOUT]

Produce:
1. The deck structure with a stated purpose for each section and a time allocation
2. For each decision required: the framing, the options, my recommendation, and the information the board needs to decide
3. How and where to raise the issues I listed as uncomfortable — early, directly, and with a plan attached
4. The three hardest questions I will be asked, and what a good answer to each contains
5. What belongs in a pre-read rather than in the room

Then tell me:
- Where my framing is defensive rather than informative
- Which numbers will be challenged, and what supporting detail to have ready
- What I should be asking the board for that I have not thought to ask

Optimise for the board being useful to me, not for a comfortable meeting.

23. The Difficult Update

For missed targets, a bad quarter, or news that has to be delivered before it is discovered.

Help me communicate difficult news to [INVESTORS / BOARD / TEAM].

THE SITUATION: [WHAT HAPPENED, PLAINLY]
NUMBERS: [THE ACTUAL IMPACT]
CAUSE: [WHAT YOU BELIEVE CAUSED IT, INCLUDING YOUR OWN ROLE]
WHAT I HAVE ALREADY DONE: [ACTIONS TAKEN]
THE PLAN: [WHAT HAPPENS NEXT AND BY WHEN]
WHAT I DO NOT YET KNOW: [BE HONEST]

Write a communication that:
1. States the situation in the first two sentences, without preamble
2. Gives the real cause rather than a comfortable one, and takes appropriate ownership without excessive self-flagellation
3. Separates what is known from what is still being established
4. Presents the plan with dates and owners, not intentions
5. Says clearly what I need from the reader

Then:
- Identify anything in my framing that reads as deflection, and rewrite it
- List the questions this will provoke, and whether I am ready to answer each
- Tell me what would be worse: what I am omitting that will surface later

Rules: no softening language that obscures the facts. Do not open with positives before the bad news. Do not overclaim on the recovery plan — credibility now is worth more than reassurance.

Chaining These Into a Rhythm

Individually they answer questions. On a schedule they run a business:

• Weekly: prompt 16 for the metrics review, prompt 5 refreshed if cash is tight.
• Monthly: prompt 12 to read the management accounts, prompt 15 to decompose growth, prompt 21 for the investor update.
• Quarterly: prompt 2 on unit economics, prompt 14 on cohorts, prompt 13 to check the KPI tree still reflects the priority, prompt 22 for the board.
• Annually or at inflection points: prompts 1, 3 and 4 on model, strategy and position; prompt 10 to rebuild the scenario range.
• As decisions arise: prompts 8, 17, 18 and 20.

The compounding value is in the repetition. Running prompt 2 once tells you your unit economics. Running it quarterly with the same format tells you whether they are improving — which is the actual question.

Mistakes That Waste These Prompts

1. Not pasting real data. The most common failure by a wide margin. A prompt with placeholders left unfilled returns a template, not an analysis.
2. Accepting the first answer. These are structured to expose reasoning. Read the assumption list, challenge the two weakest, and re-run.
3. Letting it estimate silently. If you do not tell the model to flag gaps, it will fill them plausibly and you will not notice.
4. Treating output as verified. Check the arithmetic on anything material. Models make calculation errors, and a confident wrong number is more dangerous than an obvious gap.
5. Skipping the case against. The section you are tempted to skim is the one you are paying for.
6. Using it as a substitute for an accountant. Tax treatment, statutory obligations and anything filed with a regulator need a professional. These prompts help you arrive at that conversation prepared, not avoid it.
7. Running them once. The value is in the trend, and the trend needs the same prompt run the same way over time.

Which Model to Run These On

These prompts are long, involve multi-step arithmetic, and reward careful reasoning — which points to a frontier model rather than a fast one. Anything involving a table of numbers benefits from a model that can hold the whole dataset in context and show its working. Our Claude Opus 5 prompt guide and Claude Sonnet 5 prompts cover where each fits, and Claude pricing explained covers what the tiers cost.

For prompts needing live market or competitor data — 4, 18 and 19 in particular — an agentic research tool that can actually browse will beat a model working from training data. Our Genspark review and plan comparison cover that category, and AI skills vs AI agents explains when you need one rather than the other.

Keep Reading

More on getting real work out of AI: 33+ Claude Opus 5 prompts, 40+ Claude Sonnet 5 prompts, 100 prompts for power users, the Genspark review, what ChatGPT Work is, and AI skills vs AI agents. Or browse all guides and prompts on PromptsRush.