AI Finance Brief › Finance Prompts › AI Prompts for Sales & Trading
AI Prompts for Sales & Trading
S&T desk (color / flow / risk) + client-facing sales
On a sales & trading desk the constraint is time, not information. The prints are public seconds after the open; the edge is turning overnight moves into desk color, a trade idea into a risk map, and a view into a client-ready note before the market does. These prompts are built for exactly that — fast, structured, and honest about what the model can and can't know. Paste in your own data; the prompts force the model to reason from it, not fabricate.
3 free prompts you can run right now
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.
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.
Morning desk-color note from overnight moves
You need a tight, client-ready morning color note synthesizing overnight action for your coverage.
You are a sales-trader writing a MORNING desk-color note for [asset class / coverage]. I will paste overnight moves, headlines, and levels. Produce: 1. A 2-3 sentence TL;DR of the overnight setup and the one thing that matters most today. 2. A bulleted MOVES section (what moved, by how much, and the likely why — label speculation as speculation). 3. A WHAT-TO-WATCH list (data, events, levels) with times where relevant. 4. A short positioning/flow read framed as observation, not advice. Keep it desk-voice: concise, specific, no filler. Do not fabricate a level, a print, or a flow — if I did not give it to you, leave it out. 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.
Event / catalyst calendar + scenario tree
You want to map the upcoming catalysts for a name and how price might react to each outcome.
You are a trader mapping catalysts for [TICKER / theme]. I will give the known events (earnings, data, product, regulatory, macro). Produce: 1. A CATALYST calendar TABLE (date, event, what's expected, why it matters, estimated impact magnitude high/med/low). 2. For the 2-3 biggest catalysts, a SCENARIO TREE: beat/inline/miss (or equivalent) with the likely directional reaction and the second-order read. 3. The single event with the most asymmetric setup and why. 4. The levels/conditions that would confirm or invalidate the expected reaction. Use only events I provide; do not invent a date or a consensus number. 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.
Get today's prompt free — every weekday
One fresh, desk-ready prompt for banking, trading, or the buy side lands in your inbox each morning. No news, no fluff. Free.
Instant access + the free brief. Unsubscribe anytime.
Want prompts matched to your exact role and coverage? Set your prompt profile →
7 more Sales & Trading prompts in Pro
The full Sales & Trading set — plus every other desk's prompts, personalized to your role — is part of Pro. Here's what's inside:
Options-structure walk-through for a client
A client asked about a structure and you need a clear, honest explanation of the trade-offs.
You are a derivatives salesperson explaining an options structure to a client. The structure is [describe: e.g. call spread, risk reversal, collar, pu
Cross-asset macro read + positioning implications
You want a structured cross-asset read to frame the desk's positioning and client conversations.
You are a macro strategist writing a cross-asset read. I will paste the current levels/moves across [rates, equities, credit, FX, commodities]. Produc
Trade post-mortem + process fix
A trade closed (win or loss) and you want to extract the process lesson, not just the P&L.
You are a trading coach running a POST-MORTEM on a closed trade. I will give the thesis, entry, sizing, management, exit, and outcome. Produce: 1. A v
Client axe / color email draft
You need to send a client a crisp axe or color email that's compliant and value-additive.
You are a sales-trader drafting a client email. Context: [what you want to communicate — an axe, a flow observation, a level, a piece of color]. Produ
Relative-value pair thesis + risk legs
You have a relative-value idea and want it stress-tested with the risk legs made explicit.
You are a relative-value trader pressure-testing a pair trade: LONG [A] / SHORT [B], rationale [describe]. Produce: 1. The thesis restated in one sent
Earnings reaction playbook + levels
A name you follow reports soon and you want a pre-built reaction plan across outcomes.
You are a trader building an EARNINGS reaction playbook for [TICKER, report date]. I will give the setup: consensus, whisper if known, implied move, k
Client meeting prep brief (talking points + likely questions)
You have a client meeting and need a tight prep brief: what to raise and what they'll ask.
You are a sales-trader prepping for a client meeting. Context: [client type, mandate, recent activity, what you want to achieve]. Produce: 1. THREE ta
Frequently asked
Can AI generate trade ideas?
It can pressure-test and structure your ideas — that is where it earns its keep. The strongest use is a pre-mortem: paste your thesis and have the model build the bear case, list what you're assuming but haven't verified, and name the single data point that should make you exit. It should not be handed discretion over a live book, and any levels or figures it cites must be checked. Educational only, not a recommendation to buy or sell any security.
How do traders use ChatGPT for desk color?
By pasting the overnight tape — index moves, notable single-name gaps, macro prints — and asking for a ranked 'what actually changed' read a PM could act on in 60 seconds, plus the one thing the market may be mispricing. The value is compression and ranking, not new information. Feed it only data you're cleared to use, and never client flow or confidential axes on a consumer tool.
Will AI replace trading desks?
No — front-office AI adoption is rising quickly, but institutions keep humans accountable for risk and client outcomes for regulatory and fiduciary reasons. The realistic gain is speed on the repetitive layer: summarizing releases, drafting color, mapping scenarios. Judgment, execution, and relationships stay with the desk.