17 ChatGPT Prompts That Actually Work for Solopreneurs (Structured, Not Vague)

Here's the problem with most ChatGPT prompts you find online: they're starting points, not solutions. "Write me a cold email" returns a cold email that sounds like it was written by someone who has never sent a cold email. "Help me respond to a difficult client" gets you something diplomatic to the point of meaningless.

Vague input, vague output. That's not a ChatGPT problem — it's a prompt structure problem.

As a solopreneur, you're doing sales, delivery, marketing, ops, and client management simultaneously. You don't have 45 minutes to iterate through seven drafts before you get something usable. You need prompts that treat ChatGPT like a smart contractor who already knows the context — not a blank-slate assistant waiting for you to figure out what you want.

The fix is structure: context slots, explicit constraints, and instructions about what *not* to do. Every prompt below follows that pattern. Each has placeholders in [BRACKETS] for your specific context. Copy it, fill in the brackets, and get output worth using on the first try.

Category 1: Client Communication

The highest-stakes writing most solopreneurs do. These three prompts are for the moments where saying the wrong thing costs real money or real relationships.

Prompt 1: Proposal follow-up that doesn't sound like begging

You're a business consultant helping me follow up on a proposal I sent to a prospective client. Here's the context:

- My name: [YOUR NAME]
- What I do: [YOUR SERVICE IN ONE SENTENCE]
- Proposal sent: [X] days ago
- What the proposal covered: [BRIEF DESCRIPTION]
- Last interaction: [WHAT YOU DISCUSSED OR HOW THEY RESPONDED INITIALLY]
- Their likely concern or hesitation: [WHAT YOU THINK IS MAKING THEM PAUSE]

Write a follow-up email that:
- Opens with something specific to our last conversation (not "I just wanted to follow up")
- Acknowledges that decisions like this take time without being apologetic
- Adds one sentence of new value or framing that I didn't include in the original proposal
- Ends with a clear, low-pressure next step (a question, not a CTA)
- Tone: confident, direct, human — not salesy

Do NOT use the phrases "circling back," "touching base," "following up," or "just checking in."

Why this works: The constraint list ("do NOT use") is what separates this from a generic follow-up. Combined with the specific context slots, ChatGPT writes a message that actually sounds like you — not a template.

Prompt 2: Scope creep response that preserves the relationship

I'm a [YOUR PROFESSION] and a client is asking me to do work outside our original agreement. Here's what's happening:

- Original project scope: [WHAT WAS AGREED]
- What they're now asking for: [THE NEW REQUEST]
- Relationship with this client: [NEW / ONGOING / HIGH-VALUE / SENSITIVE]
- My goal: Address the scope creep professionally, offer a path forward to get the work done, and preserve the relationship

Write a short email (under 150 words) that:
- Acknowledges what they're asking without making them feel bad for asking
- Clearly positions this as outside the original scope (without being defensive)
- Offers two options: add it to the current project at a specific rate, or incorporate it into a future engagement
- Does not use legal language or make it feel like a contract dispute

Tone: warm, professional, direct.

Why this works: The two-option structure forces ChatGPT to draft a response with a real decision path, not just a vague "let's discuss." The word limit keeps it from becoming a wall of text the client won't read.

Prompt 3: Price increase email that doesn't lose clients

I need to write an email to [EXISTING CLIENTS / A SPECIFIC CLIENT SEGMENT] announcing a price increase. Here's the context:

- What I do: [SERVICE]
- Current price: [AMOUNT]
- New price: [AMOUNT]
- Effective date: [DATE]
- Reason for increase (honest version): [RISING COSTS / EXPANDED SCOPE / VALUE DELIVERED / MARKET RATE]
- How much notice I'm giving: [X WEEKS/MONTHS]
- Any grandfathering or exceptions for long-term clients: [YES/NO — details if yes]

Write an email that:
- Leads with appreciation, not apology
- States the change clearly in the first paragraph — no burying it
- Briefly explains the reason without over-justifying
- Reinforces the value they've received (one concrete example if possible)
- Makes the next step clear (reply to lock in current rate, or no action needed)

Do NOT: be overly apologetic, use passive voice to soften the announcement, or leave the price change ambiguous.

Why this works: Most "price increase" email prompts produce wishy-washy messages that bury the news and apologize three times. The "lead with appreciation, not apology" instruction and "do NOT be overly apologetic" constraint flip the tone.

Category 2: Content Creation

Four prompts for the content you should be publishing consistently but probably aren't because starting from scratch is slow.

Prompt 4: LinkedIn post from a core idea

I want to write a LinkedIn post based on this core idea: [YOUR IDEA IN 1-3 SENTENCES]

My audience is: [WHO FOLLOWS YOU — be specific, e.g., "B2B SaaS founders" or "freelance designers"]
What I want them to take away: [ONE SPECIFIC INSIGHT OR ACTION]

Write a LinkedIn post that:
- Opens with a line that creates curiosity or tension (not a motivational statement)
- Develops the idea in short paragraphs — no walls of text
- Uses a specific example or data point to make the point concrete
- Ends with a question or a direct statement, not a CTA to "like and share"
- Length: 150–250 words

Do NOT use: hustle-culture language ("grind," "crushing it," "no days off"), vague inspiration ("success is a mindset"), or em dashes as a stylistic crutch.

Why this works: The explicit "do NOT use hustle-culture language" constraint is the difference between a post that sounds like you and one that sounds like every other LinkedIn account. The opening line instruction kills the "I've been thinking about X lately…" opener that gets ignored.

Prompt 5: Newsletter intro that skips the generic opener

Write an intro paragraph (100–150 words) for my newsletter issue. Here's what this issue is about:

- Main topic: [TOPIC]
- The hook or angle: [WHAT MAKES THIS INTERESTING OR COUNTERINTUITIVE]
- My newsletter tone: [DESCRIBE IT — e.g., "direct, no-fluff, talks to operators not dreamers"]
- What I want readers to feel after the intro: [CURIOUS / CHALLENGED / VALIDATED / ENERGIZED]

The intro should:
- Start in the middle of a thought or scene — not with "Welcome to issue #X" or "This week I want to talk about"
- Make the reader feel like they'd be missing something if they stopped reading
- Connect immediately to what the rest of the issue delivers

Do NOT: start with a question, use filler phrases like "hope you're doing well," or open by summarizing what the email contains.

Why this works: Newsletter openers are the most skipped part of most newsletters. The "start in the middle of a thought or scene" instruction forces an entry point that creates momentum instead of warming up.

Prompt 6: Case study from a client win

Write a concise case study (300–400 words) based on a client win. Here are the details:

- Client type: [INDUSTRY / SIZE / ROLE — anonymized if needed]
- The problem they came to me with: [SPECIFIC PROBLEM]
- What I did (approach): [YOUR METHOD OR PROCESS]
- The result: [SPECIFIC OUTCOME]
- Before metric: [QUANTIFIED BASELINE — e.g., "closing 1 in 10 proposals"]
- After metric: [QUANTIFIED RESULT — e.g., "closing 4 in 10 within 60 days"]
- What made the difference: [THE KEY INSIGHT OR UNLOCK]

Structure it as: Problem → What we tried → What changed → Result
Tone: factual and specific — not a testimonial or a sales piece

Do NOT use superlatives ("incredible," "amazing," "game-changing"), vague claims ("dramatically improved results"), or passive voice.

Why this works: The before/after metric slots force the case study to have actual proof points instead of vague "great results" language. Without those constraints, ChatGPT fills the space with adjectives.

Prompt 7: Turn a long article into a tweet thread with a hook first

Turn the following article into a tweet thread. The article: [PASTE ARTICLE TEXT OR SUMMARY]

Thread requirements:
- First tweet: A hook that works standalone — either a surprising claim, a contrarian take, or a specific number that creates curiosity. NOT a "thread incoming 🧵" opener.
- Tweets 2–8: One idea per tweet, max 240 characters each, connected logically
- Last tweet: A practical takeaway or a pointed question — not "follow me for more"
- Thread length: 7–10 tweets total

Do NOT: add emojis unless they add meaning, start consecutive tweets with "And," or pad tweets with filler to hit a length.

Format each tweet on its own line, numbered 1/ through [X]/.

Why this works: The "hook that works standalone" instruction is critical — it's the only tweet that most people will see in their feed. The numbered format makes it easy to copy-paste directly without reformatting.

Category 3: Business Decisions

These prompts turn ChatGPT into a useful thinking partner instead of a yes-machine.

Prompt 8: Devil's advocate against your business idea

I have a business idea and I want you to argue against it. Don't just list generic risks — pressure-test the specific logic.

Here's the idea: [DESCRIBE YOUR IDEA IN 2–4 SENTENCES]

My current reasoning for why it works: [YOUR CORE THESIS]

Play devil's advocate and tell me:
1. The most likely reason this fails in the first 12 months
2. What I'm probably underweighting (customer behavior, competition, market timing, execution complexity)
3. The assumption that, if wrong, would collapse the whole thing
4. A version of this idea that would be obviously worse — and why mine might actually be that version

Be direct. Don't soften the critique or balance it with positives. I'll ask for the positives separately.

Why this works: "Be direct. Don't soften the critique" is load-bearing. Without it, ChatGPT appends "but there's also a lot of potential here!" to every critique. The "what I'm probably underweighting" instruction surfaces blind spots, not just obvious risks.

Prompt 9: Evaluate two options with tradeoffs, not just pros/cons

I'm deciding between two options. Help me think through the tradeoffs clearly.

Option A: [DESCRIBE OPTION A]
Option B: [DESCRIBE OPTION B]

My situation: [RELEVANT CONTEXT — constraints, goals, timeline, what matters most to me]

For each option, give me:
1. What it optimizes for (and what it sacrifices to do so)
2. What has to be true for this to be the right call
3. The hidden cost or downside most people don't consider
4. Which type of person or situation this is right for

At the end: If you had to pick one based only on what I've told you, which would you choose and why? Give me a direct answer.

Why this works: "Pros/cons" lists are symmetric — they make both options look equally valid. "What has to be true for this to be the right call" forces asymmetric evaluation that's actually useful for making a decision.

Prompt 10: Surface the assumptions you're making

I'm working on [PROJECT / DECISION / PLAN — brief description].

Here's my current thinking: [EXPLAIN YOUR APPROACH OR REASONING IN 3–5 SENTENCES]

List the assumptions embedded in this plan — especially:
1. Assumptions about customer behavior or demand
2. Assumptions about my own capacity or execution speed
3. Assumptions about how competitors or the market will respond
4. Assumptions I'm probably making unconsciously (things I'm treating as facts that are actually bets)

For each assumption, rate how testable it is (easy / hard / untestable) and what I'd need to do to validate it before committing.

Why this works: The "unconscious assumptions" category is the one that actually matters. Prompting ChatGPT to surface the assumptions you're making without realizing it produces a different class of insight than just listing "what could go wrong."

Category 4: Sales & Outreach

The four situations where bad messaging costs you the deal.

Prompt 11: Cold DM that doesn't sound like a cold DM

Write a cold DM I can send on [PLATFORM: LinkedIn / Twitter / Instagram] to a potential client or collaborator.

Here's the context:
- Who I'm reaching out to: [THEIR ROLE / WHAT THEY DO]
- Why I'm reaching out to them specifically: [WHAT CAUGHT MY ATTENTION — be specific]
- What I do: [ONE SENTENCE]
- What I want to happen next: [A REPLY / A CALL / TO SHARE SOMETHING RELEVANT]

Write a DM that:
- Opens with something specific to them (not a compliment, a specific observation)
- Gets to the point in 3–4 sentences
- Makes it easy to say yes or reply (low friction ask)
- Sounds like one human writing to another, not a sales sequence

Do NOT: use "I came across your profile," "I'd love to pick your brain," "I think we could really add value," or any opener that makes it obvious this is a template.

Why this works: The "specific observation, not a compliment" instruction breaks the formula. Most cold DMs open with a flattering generic line — this one tells ChatGPT to reference something real, which produces a message that feels researched.

Prompt 12: Follow-up message for someone who went quiet

I reached out to someone [X DAYS / WEEKS] ago and they went quiet. Here's what happened:

- Initial contact context: [HOW YOU FIRST CONNECTED / WHAT YOU SENT]
- Their initial response (if any): [WHAT THEY SAID OR DID]
- Why I want to follow up: [SPECIFIC REASON — not just "checking in"]
- What I want to happen next: [ONE CLEAR NEXT STEP]

Write a short follow-up message (under 100 words) that:
- Doesn't make them feel guilty for going quiet
- Leads with something new — a piece of value, a relevant observation, or an update on my end
- Makes it easy to re-engage with a simple yes/no question or action
- Leaves the door open if the timing is wrong

Do NOT: open with "Just following up," "I wanted to circle back," or any variation of "Did you get a chance to…"

Why this works: "Leads with something new" is the structural fix — it gives the recipient a reason to respond beyond obligation. Most follow-up prompts produce polite nudges; this one produces messages that restart conversations.

Prompt 13: Objection handler for your most common pushback

Help me write a response to a common sales objection. Here's the context:

- What I sell: [YOUR PRODUCT OR SERVICE]
- The objection I keep hearing: [EXACT WORDS OR CLOSE PARAPHRASE]
- Why this objection usually comes up: [THE REAL CONCERN BEHIND IT]
- What I believe is actually true: [YOUR HONEST COUNTER-PERSPECTIVE]
- The outcome I want: [CONTINUE THE CONVERSATION / REFRAME THEIR THINKING / CLOSE THE DEAL]

Write a response that:
- Acknowledges the objection without dismissing or over-validating it
- Reframes the underlying concern rather than just arguing against the surface statement
- Uses a specific example or analogy to make the reframe concrete
- Ends with a question that moves the conversation forward

Tone: confident and grounded — not defensive, not pushy.

Why this works: The "real concern behind it" slot is what makes this work — you're telling ChatGPT what's actually going on, not just the surface objection. The reframe instruction means you get a response that actually shifts perspective instead of just pushing back.

Prompt 14: Discovery call prep

I have a discovery call coming up with [PROSPECT DESCRIPTION — role, company type, size]. Help me prepare.

What I know about them so far: [WHAT YOU'VE FOUND — website, LinkedIn, referral context, etc.]
What I think they're trying to solve: [YOUR HYPOTHESIS]
My service and how it's relevant: [ONE PARAGRAPH]

Give me:
1. 3 things I should research before the call (and where to find them)
2. 5 discovery questions that uncover budget, urgency, and decision-making authority without sounding like a script
3. The 2 questions most people forget to ask on discovery calls that I should ask
4. What to listen for that signals a good fit vs. a red flag
5. How to end the call with a clear next step (not just "I'll send a proposal")

Why this works: "Questions most people forget to ask" consistently surfaces useful angles ChatGPT wouldn't generate if you just asked for "good discovery questions." The structured five-part output also means you walk away with a complete prep doc, not a list.

Category 5: Operations & Thinking

The three prompts that turn messy input into structured output — fast.

Prompt 15: Meeting notes to action items only

Here are my raw notes from a meeting: [PASTE YOUR NOTES]

Extract only the action items from these notes. For each one:
- Who is responsible (if mentioned, otherwise write "unassigned")
- What the action is (specific, verb-first)
- Any deadline or timing mentioned

Format as a numbered list.

Do NOT: summarize the meeting, include discussion points or context, or add anything that wasn't in the notes. If something is unclear, flag it with a "?" rather than interpreting it.

Why this works: The explicit "do NOT summarize" instruction prevents the default ChatGPT behavior of re-narrating your meeting before getting to the point. The "flag with a ?" instruction handles ambiguity cleanly without inventing details.

Prompt 16: Brain dump to prioritized to-do list

Here is a brain dump of everything on my mind for [TODAY / THIS WEEK]: [PASTE YOUR BRAIN DUMP]

Organize this into a prioritized to-do list using this framework:
- Must do today (blocking or time-sensitive)
- Should do this week (important but not urgent)
- Can wait (low priority or can be delegated)
- Capture only (ideas, someday items — not a to-do right now)

For items in "must do today," put them in the order I should tackle them (hardest/highest-stakes first).

Do NOT: add tasks I didn't mention, combine tasks into vague categories, or create subtasks I didn't already reference.

Why this works: The four-tier framework forces ChatGPT to make prioritization decisions, not just reformat your list. The "hardest first" ordering instruction for today's items adds a further layer of structure that most task lists skip.

Prompt 17: Draft an SOP from a process you just described

I'm going to describe a process I do regularly. Turn it into a clear SOP (Standard Operating Procedure) that someone else could follow.

Here's the process: [DESCRIBE THE PROCESS — can be messy, step-by-step or narrative]

Format the SOP as:
- Purpose: What this process accomplishes (one sentence)
- Who does this: [ROLE]
- When to use this: [TRIGGER — what kicks off the process]
- Steps: Numbered, verb-first (e.g., "Open X," "Send Y to Z"), in the order they happen
- Notes / edge cases: Anything that varies or needs judgment

Keep it scannable — each step should be one clear action. If a step has a sub-step, use a lettered list (a, b, c) under the parent step.

Do NOT: turn this into a training document, add "why" explanations for each step, or include advice I didn't ask for.

Why this works: "Verb-first steps" and the explicit format instruction prevent ChatGPT from producing paragraph-style descriptions that require interpretation. The "do NOT add why explanations" keeps the SOP functional rather than instructional.

Why Structure Changes Everything

The difference between "write me a cold email" and a good prompt isn't just detail — it's architecture.

Here's the same request, two ways:

Vague: *"Write me a cold email to send to potential clients."*

Structured: *"Write a cold email to a [ROLE] at a [COMPANY TYPE]. I'm a [YOUR ROLE] who helps [TARGET CLIENT] with [SPECIFIC PROBLEM]. The tone should be direct and human. The email should be under 100 words, lead with a specific observation about their business, and end with a low-friction question — not a pitch. Do not use the phrases 'I hope this finds you well,' 'reaching out,' or 'synergy.'"*

The structured version gives ChatGPT a specific audience, tone direction, format constraints, a clear call-to-action structure, and explicit exclusions. The vague version gives it nothing to work with — and it will fill in the blanks with the most common patterns it's seen, which are bad.

Every prompt in this post follows that pattern: audience, constraint, what NOT to do. Once you internalize the structure, you stop needing to iterate through five drafts to get something usable.

If you work with Claude instead of (or alongside) ChatGPT, the Claude AI Prompt Pack — 30 Business Prompts covers the same core functions optimized for how Claude processes context. $17.

Try More Prompts — Free

If these were useful, the full ChatGPT Prompt Pack for Business includes 50 structured prompts for solopreneurs and small business operators — covering sales, client management, content, operations, and decision-making.

[Try 5 prompts free — no email required →](https://stackd-store.madethis.app/preview)

If the preview is useful, the full pack is $19:

[Get the ChatGPT Prompt Pack (50 prompts, $19) →](https://madethis.com/checkout/stackd-store/md74sn22t94cq3hmh9ze98sk4h887csk)

Every prompt in the pack follows the same structured format: context slots, constraints, and what NOT to do. Built for people who don't have time to iterate.