My part of this is short, two paragraphs, then it's about you. Last month I ran this three-prompt chain as a 10-day sprint and it added $15,000: fourteen ghostwriting retainers at $750 each, plus nine LinkedIn profile packages at $500, on under $1,000 spent up front. That's it. That's the income. I'm putting it here so you know the chain has been run for real, not so you expect the same numbers on your first pass. You won't get them on the first pass. Nobody does.
What you should expect is the part that actually matters: your first $1,000. That's where almost everyone is stuck. Not at $50k wondering how to scale. At zero, drowning in free business advice, taking no action. My own start was one LinkedIn post a day into silence for weeks. The unlock was never more advice. It was picking one path and selling something within the week, because the fastest path to $1 million is your first $1,000. Once one stranger pays you, you have momentum and a data point instead of a theory.
From here on, this is a set of instructions. Run them.
One thing before the prompts. If your plan is to self-teach the client-getting game from YouTube, budget for losing a few weeks to tutorials that contradict each other. One video tells you to niche down, the next says stay broad, a third wants you running ads on day one. A structured course cuts through that so cleanly it almost feels unfair. YouTube is fine once you already know what you are looking for. It is a maze when you do not.
Also, there's a fourth bonus prompt at the end of this post, and it probably matters more than the three main ones if procrastination is your real problem. It's at the end on purpose. Don't skip it.
Prompt 1: figure out what to sell
Log into Claude. Website, desktop app, Cowork, doesn't matter. What does matter is using the version of Claude that has the most context about you, because the answer gets sharper with everything it already knows about your skills and situation. Then run this:
"Give me two ways I could make my first $1,000 in 30 days. Both ways must require skills I already have, not require an audience, and be sellable within 48 hours. Ask me clarifying questions until you're 95% confident in your recommendations."
Every constraint in there is doing work for you. "Skills I already have" kills the two-month detour where you learn video editing before earning a cent. "Not require an audience" covers the realistic case: if you have an audience, congratulations, you have a cheat code, but this assumes you don't. And "sellable within 48 hours" filters out anything that needs a product built first.
The last line is the one people delete, and you shouldn't. If Claude doesn't know you yet, maybe you just moved over from ChatGPT, the clarifying questions are where the tailoring happens. Expect questions about your background, your realistic hours (be honest if you have a full-time job, "under 10 a week" is a fine answer), who's already in your network (your first clients usually come from warm connections), and which options you could do well today without ramping up.
Answer honestly and you'll get two concrete paths. A writer who ran this got LinkedIn ghostwriting for SaaS founders via cold DMs, where a single yes could be $750, and a productized package cleaning up a founder's profile plus five launch posts, a service most people never think to offer.
If you hate the recommendations, push back. Tell Claude that LinkedIn is full of AI comments and you'd rather work Instagram. The point is to finalize two paths and pick one. Not five. One.
Prompt 2: pick one person to copy
"Who is the single most successful person I should learn from to make this happen? Limit it to one person."
One person, on purpose. You want a model of success you can study, and if they're a creator or educator they've usually documented everything that worked for them. If your path is LinkedIn writing, you'll probably get someone like Justin Welsh. If it's health, finance, or ecommerce, you'll get someone else. Fine either way.
Watch for the warning Claude tends to attach here, because it's the important part: don't spend two weeks consuming this person's content instead of selling. Give yourself a hard cap, 3 to 4 hours studying them this week, then start reaching out. Print that on something. Study time feels like progress. It isn't revenue.
Prompt 3: get a plan where selling starts on day one
"Pretend you are that person. Create a 20-day plan that needs less than $1,000 upfront and gets me to my first paying customer as fast as possible. Customer acquisition must start on day one."
That last line is the whole prompt. Ask any AI for a business plan and you tend to get a 90-day monolith where talking to customers shows up somewhere in week seven. Forcing it to day one changes the shape of everything, and the reason most people quit is they never get their first customer. Every day of "setup" is another day to lose interest.
A plan out of this prompt tends to look like: days 1 to 3, build your assets, which mostly means fixing your own profile first because it's the product you're selling. Days 2 to 3, build a prospect list from LinkedIn search. The sharpest targeting cue you'll see: founders posting once or twice a week. That's someone bought into personal branding who hasn't committed the time, which is exactly your buyer. Days 4 to 10, the free-sample close. Write one post in a founder's voice, researched from their actual company and experience, and send it with "if you like it, post it, if you want more, hit me up."
Founders get pitched in their DMs all day long, and the ones they stop and read are the ones that already gave them something. The free sample works when you have no case studies because the work is the proof, and it signals you're confident enough to show value before asking for anything.
Your full spend list, by the way: LinkedIn Premium for the search filters and InMail credits, and a Stripe payment link so people can actually pay you. It is really not expensive to start.
On the skills themselves: if the writing (or whatever your path needs) is already in your toolkit, skip ahead. If it isn't, that's weeks, not months, and format matters more than brand. Short daily lessons you'll actually finish beat a 40-hour course you'll abandon; Mindwand is built around that format, and Coursera or the Skool communities cover similar ground if those fit you better. Pick one. Finish it.
The bonus prompt (the one saved for last)
"Ask me three yes or no questions to identify what will stop me this week. Then cut the plan in half and keep only what directly leads to money."
The three main prompts give you clarity. This one deals with the fact that clarity has never made anyone send a cold DM. Expect questions like: when you hear a no or get ghosted, will you still send DM number eight within the hour? Can you send a free sample this week? Can you say "$500" and then shut up, without discounting, justifying, or offering a payment plan?
Answer honestly, including "not sure." That's the point of the exercise. And sit with the reframe Claude gives on the pricing question, because it's true: silence after a price feels like a no, and in B2B it almost never is. They're doing math, checking budget, asking someone. Buyers regularly come back and say yes after two weeks of dead air. Silence does not mean you've lost them.
The revised plan you get back will probably be almost insultingly simple. One version: a sticky note on your monitor with two numbers, DMs sent today and the 10-day total. Use it. Sales at this stage is a volume game, and volume also gets you feedback faster. DM 200 people and get 20 replies, and you have 20 data points on your offer and your price. DM 20 and you have two.
The hard part isn't the prompts. The hard part is execution, the unsexy thing 300 times in a row. If you want the structured version of this whole path, the skills plus the client-getting reps laid out day by day, that's the lane Mindwand sits in, with Skool as the option if a community format suits you better. Either way, the sticky note is non-negotiable.
One last expectation-setter: averaged over the slow start, your effective rate in the early months will probably be around $20 an hour, worse than what any tutorial promises. It compounds from there or it doesn't, and the difference is the reps. Hopefully some of this lands.