Most people use AI for business the same way: describe a problem, ask "what should I do?", and act on the confident answer that comes back. It feels like having an advisor. But generic advice built on a one-line description is worth about what you paid for it — and the part that actually matters, the thinking specific to your situation, got skipped.

The problem isn't the model. It's that running a business follows an order, and most people jump to whatever feels urgent — usually marketing or money — before they've tested whether the thing underneath is sound. This library is built around that order: you validate before you build, decide on a model before you plan, and fix the real constraint before you optimize around it.

Here's how to actually use it.

Give it your real numbers and your real situation

The fastest upgrade to any business prompt is to stop asking in the abstract and feed it the specifics. Compare these:

How should I price my product?

I sell a meal-prep service. It costs me £14 per box to make and deliver, my customers are busy professionals, and competitors charge £25–35. Help me think through pricing.

The first gets you a textbook list of pricing strategies. The second gets you a decision about your box. Every finance, strategy, and growth prompt here has slots for your actual figures and context — and the system prompts are built to reason only from what you give them and never invent market data. Vague in, vague out; specific in, useful out.

Validate before you build

This is the step that saves the most money, and the reason the library leads with Idea & Validation.

When you're excited about an idea, the urge is to start building — the product, the website, the brand. But the most expensive mistake in business is building something nobody wants. Test the demand first:

Pressure-test my business idea.

Who actually has the problem my idea solves, and how badly?

How can I validate this idea cheaply before building it?

If you can't find someone with a painful problem who'd pay, no amount of execution will save it — you'll just build the wrong thing beautifully. Validation is the cheapest insurance in business.

Fix big problems before small ones

When you're working on the business, move in this order:

Validate — is there a real problem someone will pay to solve?
Model & position — how does it make money, and why you?
Plan — the few objectives and milestones that matter.
Build the engine — sales, operations, and the team to run it.
Grow — find the one lever, then scale what works.

The order matters because there's no point optimizing a sales process for an offer nobody wants, or hiring to scale a model that doesn't yet make money. Most business owners spend their energy on step 4 or 5 while a cracked foundation at step 1 or 2 quietly undermines everything above it.

Find the real constraint, not the loud one

At any moment, one thing limits your business more than everything else — and it's usually not the thing screaming for attention. The Operations and Strategy prompts help you find it:

Where's the bottleneck in my business?

What's the biggest lever for growing my business right now?

Help me find my focus.

Pouring effort anywhere except the real constraint just creates slack you can't use. If sales is the bottleneck, a better website won't help; if delivery is the bottleneck, more sales make it worse. Find the one thing, fix that.

Decide like the stakes are real — because they are

Business is a series of decisions under uncertainty, and AI's job isn't to make them for you — it's to make you decide better. The Decisions & Risk prompts are built for that:

Help me make this hard business decision.

Run a pre-mortem on this plan.

Is this decision reversible, and how should that change my approach?

The single most useful question in the set: is this a one-way or two-way door? Reversible decisions deserve speed; irreversible ones deserve a pre-mortem. Most people give both the same anxious deliberation, and waste it on the ones that don't matter.

A 30-second router

Not sure where to start? Match your situation:

Have an idea, not sure it's real → Idea & Validation.
Idea works, unsure how to make money → Business Model & Strategy.
Lots to do, no clear direction → Planning.
Worried about the numbers → Finance & Money.
Need customers and revenue → Sales & Customers.
Drowning in the day-to-day → Operations & Systems.
Hiring or leading people → Team & Hiring.
A hard message or conversation → Communication.
Stuck or want to scale → Growth.
Facing a big or risky call → Decisions & Risk.

A quick example

Say sales are flat and you ask AI "how do I get more customers?" You get a sensible list — ads, content, referrals, partnerships — and you start doing all of them at once.

Don't. Back up and find the constraint: run the bottleneck prompt with your real numbers, and you discover the problem isn't leads at all — plenty of people enquire, but few buy. So the lever isn't more marketing, it's conversion. Now run the sales-process prompt and the "why do deals stall" prompt, and you find prospects go cold waiting three days for a quote. Fix the quote turnaround and sales climb — without spending a penny on ads.

The win didn't come from a better "get more customers" prompt. It came from finding the real constraint before acting, instead of throwing effort at the loudest symptom.

Don't want to choose? Follow the journey

If the full list is daunting, run the six-step journey instead — it walks you through getting a business off the ground in the order that works:

Pressure-test my idea — is it real?
Validate this cheaply — prove demand before building.
Design my business model — how it makes money.
Write a one-page business plan — the whole thing on a page.
Get my first customers — the most direct route to revenue.
Set the right metrics — track what actually matters.

Run those six in order and you've used the whole library's logic without browsing a single category.

The bottom line

Prompts are starting points; your real situation and honest numbers do the heavy lifting. The best results come from three habits, every time:

Validate that someone will pay before you build anything.
Feed it your actual numbers and context, not a vague description.
Find the one real constraint before you act on the loud symptom.

If you only ever use ten of these, use these:

Pressure-test my idea.
Validate this cheaply.
Design my business model.
Write a one-page business plan.
Build a simple budget.
Build a repeatable sales process.
Systemize a recurring task.
Should I hire for this?
Find my biggest growth lever.
Make a hard decision.

Work the order, give it your real situation, and these prompts will do more than generate advice — they'll help you make the calls a good advisor would, on the business only you can see in full.