Pressure-test any business idea before you spend a rupee on it.
Act as a skeptical early-stage investor who has seen 1,000 pitches and funded 20. My business idea: [describe your idea in 2-3 sentences]. My budget: [amount]. My unfair advantages: [skills, audience, access you already have].
Tear this apart honestly. Answer in order:
1. Who exactly feels this pain weekly, and what do they do about it today?
2. The three strongest reasons this fails in year one.
3. The smallest version I could test in 14 days for under [budget/10] — describe the exact test and the pass/fail number.
4. One adjacent idea that is weaker sounding but stronger commercially.
Score the idea out of 10 on demand, distribution, and defensibility. Do not flatter me. If the verdict is kill, say kill.
A full plan compressed into one sharp page you can actually act on.
Act as a business strategist. Turn my rough notes into a one-page business plan. Notes: [paste everything you have — idea, customers, pricing, costs, channels, timeline].
Structure it exactly like this:
- Problem: one sentence, the pain only.
- Customer: the narrowest slice I can win first.
- Offer: what they get and what it costs.
- Money math: price x realistic monthly customers minus known costs = monthly profit range.
- First 10 customers: the exact channel and motion that gets them, not generic marketing advice.
- Risks: top 3, each with a one-line mitigation.
- Next 7 days: a checklist I can start tomorrow morning.
Keep the whole thing under 400 words. Cut any line that sounds like a pitch deck.
Five-step cold outreach written for a stranger who owes you nothing.
Act as a B2B copywriter with a 40% reply rate. Write a cold email for me.
Who I am: [one line]. Who I am writing to: [role, company type, industry]. What I want: [meeting / reply / intro / sale]. Proof I can offer: [results, clients, numbers, or none yet].
Rules:
- Under 90 words in the body.
- First line about them, not me — give me 3 alternate openers based on things a person in their role actually cares about.
- One clear ask that takes under 2 minutes to say yes to.
- No buzzwords, no 'I hope this finds you well', no fake familiarity.
Then write two follow-ups: one at day 3 (add new value, not just bumping) and one at day 7 (the polite breakup email that often gets the reply).
Find out if your price is quietly killing your sales.
Act as a pricing consultant who has repriced 200+ products. My offer: [what you sell]. Current price: [price]. Who buys: [customer]. Current monthly sales: [number]. Competitors charge: [range].
Audit my pricing:
1. Is my price anchored against anything, or floating in space? Design the anchor I should show next to it.
2. Should I test higher, lower, or restructure into tiers? Give me the exact 3-tier structure with names, prices, and what each tier includes — where the middle tier is the obvious buy.
3. What one feature or bonus, moved between tiers, would most change conversions?
4. Write the one line of price-framing copy for my sales page.
5. Tell me the cheapest A/B price test I can run this week and what result would mean stop or continue.
X-ray any competitor and steal their playbook's best pages.
Act as a competitive intelligence analyst. My competitor: [name, website, or description]. My product: [yours].
Break them down:
1. Positioning: who they sell to and the promise they lead with, in their own apparent words.
2. Moat: what actually protects them — brand, distribution, price, switching cost — and how real it is.
3. Weak seams: 3 places customers are likely frustrated (check their reviews and complaints if you know them, or reason from their model).
4. Their likely unit economics and where margin pressure lives.
5. My wedge: given all this, the positioning I should take that makes comparison shopping favor me.
End with a table: them vs me across price, speed, quality, service, and trust. Mark where I genuinely win and where I am lying to myself.
Assume your project already failed — then work backwards to save it.
Act as a facilitator running a pre-mortem. My project: [describe project, deadline, and stakes].
It is now [deadline date] and the project has failed badly. Write the story of how it failed:
1. The 5 most likely causes of death, ranked by probability x damage. For each: the early warning sign I would see 2-4 weeks before it kills the project.
2. The 2 causes nobody on my team will say out loud.
3. For each of the top 3 risks: one preventive action (do now) and one contingency (ready if it fires).
4. Re-draw my timeline showing where the plan is overconfident.
Finish with a one-line 'kill switch' criterion: the measurable signal at which I should stop rather than pour in more time and money.
Thirty seconds that make the right people lean in.
Act as a pitch coach. Raw material about my [product / self / idea]: [paste everything].
Forge three versions of my pitch:
1. The 10-second version: one sentence, no jargon, built around the outcome the listener gets.
2. The 30-second version: problem, my twist, proof, ask — in spoken language, under 75 words.
3. The weird version: an unexpected angle or analogy that makes it memorable at a party.
For each version, explain in one line why the structure works. Then list the 3 words or phrases in my raw material that sound impressive to me but mean nothing to a stranger, and give plainer replacements. Finish with the single follow-up question I want the listener to ask, and my two-sentence answer to it.
Paste messy notes, get owners, deadlines, and next steps.
Act as a chief of staff. Here are my raw meeting notes: [paste notes, transcript, or bullet dump].
Return:
1. Decisions made — one line each, only what was actually decided.
2. Action items in a table: task, owner, deadline. Where the notes name no owner, assign the most likely owner and mark it (assumed). Where no deadline exists, propose one and mark it (proposed).
3. Open questions that were raised but parked, with who should answer them.
4. A 3-sentence summary I can paste into a follow-up email to everyone who attended.
5. Anything in the notes that sounds like a commitment someone made but nobody wrote down — flag it so it does not evaporate.
Scripts and counters for the ask you have been putting off.
Act as a negotiation coach trained in interest-based bargaining. My situation: [what you are negotiating — salary, rent, vendor contract, price]. My side: [your position, constraints, and your walk-away point if you have one]. Their likely side: [what you know or guess about the other party].
Prepare me:
1. My three strongest leverage points and their likely three.
2. The opening number or position I should state, with the exact sentence to say.
3. A decision tree: their 4 most likely responses, and my scripted counter to each — worded so I can say them naturally.
4. Three concessions I can offer that cost me little but feel valuable to them.
5. The phrase to use when I need to pause: how to buy 24 hours without weakening my position.
6. What I should never say in this negotiation and why.
Questions that get honest answers instead of polite lies.
Act as a user researcher who follows the Mom Test: ask about past behavior, never about future promises. My product idea: [describe it]. Who I am interviewing: [customer type].
Build me a 30-minute interview script:
1. 3 warm-up questions that get them telling real stories about how they currently handle [the problem area].
2. 8 core questions digging into: the last time they hit this problem, what they did, what it cost them in time or money, and what they have already tried. No question may mention my product.
3. 2 money questions that test willingness to pay without asking 'would you pay?'.
4. The closing ask: how to request a referral to two more people like them.
Then list the 5 most common ways founders ruin these interviews, and the exact phrases I must avoid.
A SWOT that ends in moves, not a poster for the wall.
Act as a strategy consultant. Business or project: [describe it, its market, and its stage].
Run a SWOT, but with rules:
- Each quadrant gets exactly 4 items, each item one sentence with a number, fact, or concrete detail attached — no vague entries like 'strong brand'.
- For each Weakness: the cost of leaving it unfixed for 6 months.
- For each Threat: an early indicator I can watch monthly.
Then the part most SWOTs skip — cross the quadrants:
1. Which Strength cancels which Threat? Name the single best pairing and the move it suggests.
2. Which Opportunity is only reachable if I fix which Weakness first?
3. End with 3 strategic moves for the next quarter, each tied to the analysis, each with a first step completable this week.
Act as a launch operator. What I am launching: [product/service/content]. My resources: [hours per week, budget, help available]. Hard deadline: [date or none].
Build my 90-day plan:
- Phase 1 (weeks 1-2): the smallest testable version and the validation target it must hit.
- Phase 2 (weeks 3-6): build weeks. Break into weekly milestones, each with a done definition.
- Phase 3 (weeks 7-10): pre-launch — audience warming, waitlist or seed users, assets needed.
- Phase 4 (weeks 11-13): launch week day-by-day, then the 2 weeks after that everyone forgets to plan.
For every week: 3-5 tasks max, in priority order. Mark which tasks I can do badly and which must be done well. Include a weekly 30-minute review ritual with the 3 numbers I should track. Assume I will fall behind once — show me which weeks have slack and which do not.
Turn a task you keep redoing into a checklist anyone can follow.
Act as an operations specialist. The repeating task: [describe the task — what triggers it, what done looks like, who does it]. My rambling explanation of how it works today: [paste everything, mess is fine].
Produce a standard operating procedure:
1. Trigger: what starts this task and how the person knows it is time.
2. Inputs: everything they need before starting, with where to find each.
3. Steps: numbered, each one action with a verb, each testable — a stranger should be able to follow without asking me anything. Flag any step where judgment is required and give the rule of thumb for the judgment call.
4. Done check: the 3-point definition of complete.
5. Failure modes: the 3 most common mistakes and how to recover from each.
6. Time budget: how long this should take, and the signal that it is going off the rails.
The monthly update that makes investors lean forward, not skim.
Act as a startup advisor who has read thousands of investor updates. My numbers and news: [revenue/users/growth, wins, losses, runway, anything else — messy is fine].
Write my monthly investor update:
1. Subject line that gets opened.
2. TL;DR: 3 bullets — best news, worst news, what I need.
3. Numbers table: metric, this month, last month, delta. Only metrics that matter.
4. Wins: 2-3, with the why behind each.
5. Losses and learnings: written with the calm honesty that builds trust — no spin.
6. Asks: 2 specific, easy-to-act requests (intros to [role/industry], advice on [topic]).
7. Runway and burn in one line.
Keep it under 300 words. Tone: confident, direct, no adjectives doing the work numbers should do. Then give me a reusable template version with [brackets] I can fill monthly.