Most prompt lists hand you a hundred prompts in no particular order, which is about as useful as a toolbox with no instructions. These eighteen are in launch order: validate the idea, name the thing, work out the money, write the plan, then find the first ten customers.
Every prompt here has blanks to fill in, and that’s deliberate. A prompt without your specifics produces advice that would fit any business in any town, and generic advice is exactly what you don’t need in week one.
One ground rule before you start. ChatGPT is a thinking partner with no access to your market, your bank balance or the truth. It will invent a market size figure as readily as it writes a tagline. Use it to generate options and stress-test your reasoning, then verify anything with a number in it. Scale alone makes that worth saying out loud: compiled ChatGPT statistics put it at roughly 900 million weekly active users as of February 2026, which means a lot of people are acting on unverified output.
These prompts stop at the moment you open the doors. After that the useful prompts are different ones — easy ways to use ChatGPT for small business marketing picks up the day-to-day copy, and ChatGPT for small business owners covers the admin load that arrives with your first real customers.
Lower your delivery costs by 23%
How we reduce costs:
- No delivery vehicle expenses
- Optimized local routes
- Pay-per-delivery model
- Average 23% delivery cost reduction
The Bottom Line
- Work in launch order. Validation prompts before naming prompts, naming before pricing, pricing before the plan.
- Fill in every blank. The specifics are what separate useful output from a horoscope.
- Use it for options and critique, never for facts. Market sizes, regulations and competitor details all need checking elsewhere.
- The highest-value prompt on this list is the devil’s advocate one. It’s the only one that can save you from spending a year on the wrong idea.
- Prompts get you to a plan. They don’t get you to a customer — that part still needs you making calls.
Stage 1: Prompts that pressure-test the idea
1. The honest critique
I’m planning to start [business type] in [city or neighborhood], selling [product or service] to [customer type]. Act as a skeptical small business advisor who has seen this fail before. List the eight most likely reasons this doesn’t work, hardest first. Don’t soften anything and don’t offer solutions yet.
This is the single most useful prompt here. Ask for encouragement and you’ll get encouragement, which costs you nothing and teaches you nothing. Ask for the failure modes and you get a checklist worth working through.
2. The assumption audit
Here’s my business idea: [two or three sentences]. List every assumption I’m making that I haven’t verified, sorted into things I can check in an afternoon and things that need real research. For each one, say exactly how I’d check it.
3. The customer interview script
I want to validate demand for [product or service] with [customer type] before I spend money. Write ten interview questions that reveal whether they have this problem and currently pay to solve it. No leading questions, and nothing that asks them to predict their future behavior.
That last constraint matters. “Would you buy this?” gets you a polite yes every time. “What did you do last time this happened?” gets you the truth.
4. The competitor map
List the types of businesses that already serve [customer type] in [city] with [product or service], including the ones that solve the problem differently. For each type, describe who they’re best for and who they leave underserved.
Use this for the shape of the market, not for named companies. It will get specific local businesses wrong. Checking the actual names takes a map search and twenty minutes.
Stage 2: Prompts for naming and positioning
5. Business name options
Generate 25 business name options for [business type] in [city] serving [customer type]. Mix of: descriptive, founder-name style, place-based, and one-word invented. Avoid anything with “solutions,” “hub,” or “co.” For each, note whether it’s easy to spell out loud over the phone.
That phone test weeds out half of the clever options. If you have to spell it twice, it will cost you orders for years.
6. The one-line pitch
Write 12 versions of a one-sentence description of my business for [customer type]. Facts: [what you sell], [what makes it different], [where you operate]. Each under 20 words. Four should lead with the problem, four with the outcome, four with what makes us different.
7. The positioning gap
Here are the three businesses I’d compete with and what each is known for: [list]. Suggest four positions I could take that none of them occupy, and for each one say what I’d have to be genuinely good at to hold it.
Stage 3: Prompts for pricing and the money
8. The cost checklist
I’m starting [business type] in [state]. List every startup and recurring cost I should budget for, grouped into one-time, monthly fixed, and per-order variable. Include the unglamorous ones people forget. Don’t estimate the amounts — I’ll fill those in.
Let it build the checklist, not the numbers. Cost estimates for your city from a model are guesses dressed as data.
9. Pricing structure options
My cost per [unit or order] is [amount] and similar businesses locally charge roughly [range]. Give me four pricing structures — flat, tiered, subscription, and minimum-order-plus-fee. For each, show the arithmetic on a typical order and say what customer behavior it encourages.
10. The break-even conversation
At [price] per order with [variable cost] per order and [monthly fixed costs], how many orders a month do I need to break even? Show the working. Then show what happens if I’m 20% below that and what levers I’d pull.
Always ask for the working. It makes arithmetic errors visible, and there will be some.
11. The funding and banking question
I’m starting [business type] with [amount] of my own money. Explain the realistic funding options at this stage, what each typically requires, and what order to pursue them in. Flag anything I should avoid as a first-time owner.
Banking setup is a real early decision and one that costs you to get wrong — it’s worth reading up on the best banks for small business before you open an account rather than defaulting to whoever holds your personal checking.
Stage 4: Prompts that produce the plan
12. The one-page plan
Write a one-page business plan for [business type] in [city] selling [product or service] to [customer type]. Sections: what we do, who buys and why, how we make money, what we need to launch, first 90 days, biggest risk. Plain language, no corporate filler, under 600 words.
One page beats forty. Nobody reads forty, including you.
13. The operations walkthrough
Walk through a single customer order for [business type] from the moment they order to the moment it’s delivered and paid. List every step, who does it, how long it takes, and where it’s most likely to go wrong.
This is the prompt that surfaces the thing new owners consistently underestimate: fulfillment. If your business involves getting physical goods to people, the order comes in fast and the delivery is where the day disappears. Deciding early whether you drive it yourself, hire, or use a courier service in your city changes your pricing, your hours and your radius.
14. The first-90-days sequence
Turn this plan into a week-by-week checklist for the first 90 days of [business type]: [paste plan]. One to four tasks a week, dependencies in the right order, and mark the three weeks where I’m most likely to fall behind.
15. The tooling shortlist
For [business type] with [number] people, list the minimum software I need to operate, grouped by job: taking orders, invoicing, scheduling, bookkeeping, customer messaging. Name two options per job, one cheap and one standard, and tell me which jobs can wait six months.
Resist buying all of it in week one. Most of it can wait, and the ones that can’t are usually payments and bookkeeping. Project and task tooling is a common early over-purchase, so it’s worth skimming a comparison of project management software before committing to a paid tier you’ll outgrow or abandon.
ChatGPT itself belongs on that shortlist, and it can read the documents you’re about to create rather than guessing at them — the best ChatGPT plugins for small businesses covers which connectors are worth wiring up and which to leave alone while you’re still small.
Stage 5: Prompts for the first ten customers
16. The outreach list
I sell [product or service] to [customer type] in [city]. List the ten types of places or people most likely to be my first paying customers, ranked by how fast they could say yes. For each, say what they’d need to hear first.
17. The cold approach
Write three short outreach messages to [customer type] introducing [business]. Facts: [what you do], [what makes you different], [a specific reason to say yes now]. Under 90 words each, no flattery, one clear ask. One for email, one for a text, one for walking in the door.
18. The objection drills
Act as [customer type] who is interested but hesitant about [product or service] from a brand-new business. Raise the five objections you’d actually have, one at a time, and push back on my answers.
Run this out loud before your first real sales conversation. It’s uncomfortable and it works.
What these prompts won’t do
Three limits.
- They won’t verify anything. Market sizes, licensing rules, local regulations and competitor pricing all need a real source. The model will supply numbers with no hesitation and no basis.
- They won’t make the decision. Twenty-five name options is a decision you still have to make. Four pricing structures is four options, not an answer.
- They won’t do the asking. The first ten customers come from conversations. A prompt can write the script and cannot have the conversation.
There’s also a version of this that goes wrong: outsourcing the thinking rather than the typing. Owners who use it well treat it like a sharp intern with a good vocabulary and no judgment. Owners who use it badly end up with a confident plan they don’t understand.
Once you’re actually open
The prompt library changes shape after launch. Validation and naming stop mattering and daily output starts mattering.
Captions, campaign emails, product copy and local ads become the weekly work. Quoting, scheduling and inbox triage become the daily work. Both have their own prompt libraries, linked above, and both are worth setting up in your first month open rather than your sixth.
One last thing, and it’s not a prompt. Starting a business is a years-long physical and mental load, and the owners who last are the ones who treat that as a real operating constraint rather than a weakness. Even solo, it’s worth borrowing from how structured small business wellness programs handle workload and burnout, because month nine of a new business is where good ideas die of exhaustion.
Frequently asked questions
Can ChatGPT write a business plan I could take to a bank?
It can produce a well-structured draft. A lender will press on the numbers, and the numbers have to be yours and defensible. Use it for structure and language, supply every figure yourself.
Which prompt should I run first?
The honest critique in Stage 1. If the idea survives eight hard objections, keep going. If it doesn’t, you’ve saved a year.
Do I need a paid ChatGPT plan to start a business?
No. Everything on this page works on the free tier. Paid plans add speed, a longer memory and connections to your own documents, which matter more once you’re operating than while you’re planning.
How do I keep it from giving me generic advice?
Fill in every blank, name your city, name your customer, and tell it to be specific or say it doesn’t know. Then push back on the first answer. The second answer is usually the useful one.
Is it risky to rely on AI while starting a business?
The risk isn’t the tool, it’s acting on unverified claims. Treat every fact as unconfirmed until you’ve seen it somewhere accountable, and the downside mostly disappears.
