| | PromptSharp Finance | PRO EDITION |
Copy-paste AI prompts for the deal desk, the trading floor, and the buy side. Friday, November 20, 2026 · For Investment Banking · Sales & Trading · Investment Management | Your full brief — all five sections’ prompts, ready to paste into your own LLM. Plus live MCP delivery into your AI tools. One ready-to-run prompt a day for the exact work you do — banking, trading, buy-side research, modeling, and decks. Paste into your own LLM. No news, no fluff. | SECTION 1 - INVESTMENT BANKING Banking For: IB coverage / M&A / capital markets analysts & associates Strategic-buyer longlist + diligence question tree Early-stage sell-side or a pitch: you need a defensible universe of logical acquirers and the diligence questions each raises. Copy & paste into your own LLM You are an M&A associate building a buyer universe for a target I describe. Produce two things:
A) A strategic + financial buyer LONGLIST. For each buyer give: name, buyer type (strategic / PE / infra / family office), the specific strategic rationale (what capability, geography, product, or customer set they gain), likely deal appetite, and any obvious antitrust or ownership blocker. Group by 'most logical', 'plausible', and 'wildcard'. Do not pad the list — quality over count.
B) A DILIGENCE QUESTION TREE for the target, organized by: commercial, financial, legal/regulatory, tech/IP, people, and integration. Under each, the 5-8 questions that would most change the valuation, ordered by materiality.
Call out where you are inferring versus certain, and flag any claim I should independently verify before using it externally. Here is the target:
[DESCRIBE TARGET: sector, size, geography, what it does, why it might sell]
Rules: Do not invent, estimate, or extrapolate any figure — if a number is not in what I give you, write "not provided" and flag it. Mark every claim I should verify externally before relying on it. Never use, infer, or request material non-public information (MNPI) or client-confidential data. Why it works: Gets you from a blank buyer list to a grouped, rationale-tagged longlist plus the diligence spine in one pass — and the 'infer vs certain / verify externally' flag keeps you from putting an AI guess in front of a client. | | SECTION 2 - SALES & TRADING Sales & Trading For: S&T desk (color / flow / risk) + client-facing sales Red-team my trade thesis (pre-mortem + risk map) Before you put on or upsize a position — or send desk color to a client — force a structured pre-mortem so you are not just reading your own book back to yourself. Copy & paste into your own LLM Act as the sharpest bear on the other side of my trade. I will give you the thesis, the instrument, the entry, the sizing, and my time horizon. Do this:
1) Restate my thesis in one sentence and name the 3 assumptions it silently depends on.
2) Pre-mortem: it is my horizon-end and the trade lost money. Give the 5 most likely reasons, ranked, and the earliest observable signal that each is starting to happen.
3) Map the risks: what's idiosyncratic vs factor/macro, what's crowded, what's the liquidity/gap risk, and what event dates fall inside my horizon.
4) Tell me what price/data would FALSIFY the thesis (my exit), and what would make me want to add.
5) Give a one-line 'the market is offering me this trade because ___' — i.e. what edge or risk premium am I actually being paid for.
Be concrete, no hedging language. Do NOT tell me to buy or sell — pressure-test only. Here is the setup:
[THESIS, INSTRUMENT, ENTRY, SIZE, HORIZON]
Rules: Do not invent, estimate, or extrapolate any figure — if a number is not in what I give you, write "not provided" and flag it. Mark every claim I should verify externally before relying on it. Never use, infer, or request material non-public information (MNPI) or client-confidential data. Why it works: A structured pre-mortem + falsification level is how disciplined desks avoid marrying a position. The 'why is the market offering me this' line surfaces whether you're paid for edge or just taking hidden risk. It explicitly refuses to give directional advice. | | SECTION 3 - INVESTMENT MANAGEMENT Investment Management For: Buy-side analysts & PMs (research / theses / portfolio) 10-K / earnings-call synthesis into a one-page decision brief You have a filing or a transcript and 20 minutes before an idea review. You need the signal, not a summary. Copy & paste into your own LLM You are a buy-side analyst turning a filing/transcript into a one-page DECISION brief (not a summary). I will paste the document. Output:
1) The 3 things that actually changed versus prior period/expectations (and the exact language that signals each).
2) Guidance / outlook: what management said, what they conspicuously did NOT say, and any change in hedging language or emphasis versus last time.
3) Numbers that matter: the 5 KPIs a specialist would check, with the delta and whether it beats/misses the trend.
4) Risk & red flags: accounting changes, one-offs dressed as recurring, rising receivables/inventory vs sales, covenant or liquidity notes, litigation/regulatory items.
5) The 3 sharpest questions to ask management (or to answer before I act).
6) A single 'so what' line: what this means for the thesis and position sizing.
Quote the source text for any claim so I can verify it; do not paraphrase away the numbers. Here is the document:
[PASTE 10-K SECTION / EARNINGS TRANSCRIPT]
Rules: Do not invent, estimate, or extrapolate any figure — if a number is not in what I give you, write "not provided" and flag it. Mark every claim I should verify externally before relying on it. Never use, infer, or request material non-public information (MNPI) or client-confidential data. Why it works: Forces signal-extraction (what changed, what was unsaid, what to verify) instead of a bland recap, and the 'quote the source text' rule makes every claim checkable against the filing — essential before it moves position sizing. | | SECTION 4 - CROSS-CUTTING Financial Analysis & Modeling For: All roles — IB modeling, IM valuation, S&T scenario work Three-statement driver interrogation + model sanity-check You have a working model (or a target's historicals) and want a fast, structured challenge of the assumptions before it goes to a VP or PM. Copy & paste into your own LLM You are a skeptical senior analyst reviewing a three-statement model. I will paste the key financials and my forecast assumptions. Do the following, in this order:
1) Restate, in a TABLE, the operating drivers you infer from the historicals (revenue growth, gross margin, EBIT margin, capex % of sales, NWC as % of sales, cash conversion). Flag any that look internally inconsistent.
2) Interrogate my forward assumptions line by line. For each, tell me whether it is aggressive, conservative, or in-line versus the historical trend and typical sector ranges, and WHY. Ask me the 5 questions a VP would ask before signing off.
3) Build the implied bridge from this year's EBITDA to the terminal-year EBITDA and tell me what share of the growth is volume, price, margin, or acquisition.
4) List the 3 assumptions the valuation is MOST sensitive to and the ones that are cosmetic.
5) Give me a one-paragraph 'what would have to be true' summary I could defend in a model review.
Do not invent numbers. If something is missing, ask for it. Here are the financials and assumptions:
[PASTE HISTORICALS + YOUR ASSUMPTIONS]
Rules: Do not invent, estimate, or extrapolate any figure — if a number is not in what I give you, write "not provided" and flag it. Mark every claim I should verify externally before relying on it. Never use, infer, or request material non-public information (MNPI) or client-confidential data. Why it works: Turns the model review from a formatting pass into an assumptions stress-test, and pre-empts the exact questions a VP/PM will ask. The 'do not invent numbers / ask if missing' guardrail keeps the model honest. | | SECTION 5 - CROSS-CUTTING Pitch & Presentation For: All roles — IB pitchbooks/IC memos, IM idea write-ups, S&T client notes IC / investment memo first draft from a raw thesis You have the thesis and the numbers in your head (or a messy notes doc) and need a structured first-draft memo you can edit, not a blank page. Copy & paste into your own LLM Act as an associate drafting an investment committee memo. Turn my raw notes into a structured first draft using EXACTLY these headings: Recommendation & ask - Situation - Business & why it wins - Investment thesis (3-5 numbered points) - Key financials & valuation - Risks & mitigants - Diligence gaps / open items - Proposed next steps.
Rules:
- Lead every section with the conclusion, then support it (BLUF).
- Keep it tight and specific; no filler adjectives, no hype.
- Where I have not given you a number, write [TBD: <what is needed>] rather than guessing.
- Under Risks, steelman the bear case in 3 points and give an honest mitigant for each (or say 'no clean mitigant').
- End with the 5 questions the sharpest person in the room will ask.
Here are my raw notes:
[PASTE YOUR NOTES / THESIS / NUMBERS]
Rules: Do not invent, estimate, or extrapolate any figure — if a number is not in what I give you, write "not provided" and flag it. Mark every claim I should verify externally before relying on it. Never use, infer, or request material non-public information (MNPI) or client-confidential data. Why it works: BLUF structure + a fixed heading set gets you to a defensible draft fast, and the forced bear-case + [TBD] tags stop the memo from reading like a sell-piece. The 'sharpest questions' close is your pre-read prep. | | | | Prompt of the Week (Pro) This week's bonus: a 'comps table builder' that turns a peer set into a formatted trading-comps grid with the multiples flagged as clean vs adjusted. Pro members get the full prompt in the archive. | Your prompt archive Every prompt we’ve shipped — browse the full back-catalog. Open archive → | Prompts reflect real IB/S&T/IM workflows. We make no efficacy or accuracy guarantees — you own the output and must check every figure. Do not paste MNPI or client-confidential data into third-party models. PromptSharp Finance is an educational product. Not investment advice; not a recommendation to buy or sell any security. You are responsible for verifying every output. You are receiving this because you subscribed to PromptSharp Finance at promptsharp.ai. ECWE VENTURES LLC · 202 N Cedar Ave, STE #1, Owatonna, MN 55060 Subscribe · Prompt archive · Go Pro · Unsubscribe |
|