AI Digital Products
Make something once — a template, a guide, a planner — and keep selling it.
Product sales
Sell once, keep selling
| Event | Product | Source | Sale |
|---|---|---|---|
| Sale | Notion finance pack | Organic search | +€19 |
| Sale | Invoice template set | Marketplace | +€9 |
| Sale | Client onboarding kit | Organic search | +€39 |
| Refund | Notion finance pack | — | -€19 |
| Sale | Notion finance pack | Newsletter | +€19 |
| Sale | Notion finance pack | Organic search | +€19 |
- Notion finance pack38 sales€722
- Client onboarding kit11 sales€429
- Invoice template set9 sales€81
Example figures. These amounts show how the workflow behaves — they are not earnings, not ours and not anyone else's. Nobody made this money.
Open full screen · recording mode — sized for vertical video.
Follow these prompts to start building your income stream
6 steps. Paste each prompt into Claude, use what comes out, move to the next. The last one ends in something live — not in a plan to make something live.
- 01
Pick the question you keep getting
Free sampleOne question, one buyer, and the smallest product that answers it.
In this step you choose what to make, starting from something you already have: a question people keep asking you. That is a market signal you did not have to buy. You end with an outline small enough to finish this month, because an unfinished course earns nothing.
1 — Find the questionHelp me find the digital product I am already qualified to make. What I do or have done: [YOUR BACKGROUND] Questions people actually ask me, or that I see asked repeatedly in places I hang out: [QUESTIONS YOU GET] Who asks them: [WHO ASKS] For each question: 1. What the person is really trying to achieve behind the question. 2. What they do today instead, and why that is unsatisfying. 3. What format would answer it best — template, checklist, guide, spreadsheet, swipe file, short course. 4. What they could reasonably pay, and what comparable things cost today. 5. Why they might not buy even though the problem is real. Then rank by "smallest thing I can finish that still solves it completely", and explain the top one.
2 — Outline the smallest complete productOutline a digital product that answers: [THE QUESTION] for [BUYER]. Format: [FORMAT]. What they should be able to do after using it: [AFTER STATE]. Give me: 1. The contents, in the order the buyer needs them, not the order I know them. 2. For each part: what the buyer does with it, and what it looks like when done. 3. What to leave out, with a reason per item — the cuts are the product decision. 4. The one piece that has to be excellent for the rest to be worth buying. 5. What makes this finishable in [TIME AVAILABLE] of evenings. 6. The file or platform it should be delivered in, and why. Completeness beats length. A twelve-page thing that solves the problem beats a hundred pages that circle it.
Claude. Ask it to cut the outline by a third afterwards and tell you what it removed.
- [YOUR BACKGROUND]
- Jobs, hobbies, things you have done more than once. Credibility comes from having done it, not from certification.
- [QUESTIONS YOU GET]
- Actual questions in their words. If nobody asks you anything, this strategy starts harder.
- [FORMAT]
- Pick the one that fits the problem, not the one with the highest price tag.
- [TIME AVAILABLE]
- Honest evenings per week. The scope has to fit the life you actually have.
Then: Show the outline to one person who asked the question. If they say 'I'd want X too', check whether X is the product and your outline is the add-on.
- 02
Write the product
In the packageA finished draft of the thing you are selling.
In this step the product gets made. The prompts here write with you rather than for you: the model produces structure and first drafts fast, and your experience is what makes it worth money. You end with a draft you can read end to end.
- 03
Build the sales page
In the packageA page that answers the doubts before they are asked.
In this step you write the page that does the selling. The material comes from the people who asked the original question: their words, their doubts, their alternatives. A sales page for a product like this is mostly objection handling with a price at the end.
- 04
Set up payment and delivery
In the packageA working buy button, instant delivery, and the emails around it.
In this step the mechanics get built so a stranger can buy at midnight without you being involved. The after-sale email matters more than people expect: it is where refunds are prevented and where reviews come from.
- 05
Launch to people who know you
In the packageThe product in front of a warm audience, and the first sales.
In this step you sell to the smallest audience that already trusts you — the person who asked the question, the group where it came up, the people who replied to your posts. Cold traffic is a later problem; the first buyers should be people you can talk to.
- 06
Ask your buyers for a review
In the packageReviews you can publish, and a shortlist of what to make next.
In this step the first sales become the asset that makes the second launch easier. You ask buyers what they did with it, publish what they say with permission, and let their questions tell you what to build or improve. This is also where a second product usually appears without you having to invent one.
5 of the 6 steps are in AI Money
The sample above is a real step, not a teaser — it is the same depth as the rest. The package adds the other 5 for this strategy, and the same treatment for the other eleven.
Your first 7 days
- 1Write down the five questions the people around you ask you most often.
- 2Pick the one you can answer completely in a single document.
- 3Build the smallest version of that answer and give it away to ten people.
- 4Collect their replies — the objections are the sales page.
Your first 30 days
- 1Publish the paid version with a real sales page and a working payment link.
- 2Set up VAT handling and the withdrawal-right notice before the first sale, not after.
- 3Sell to the list you already have; do not open a new channel this month.
- 4Rewrite the sales page using the exact words buyers used when they asked.
What you need
core means you cannot skip it. The rest only if you need it.
- Notioncorefree planBuild and deliver the product itself
- Claudecorefree planDrafting, structuring and editing
- Canvafree planCovers, workbooks and visual formatting
- Gammafree planTurn a document into a presentable format fast
- Stripecorefree planPayments, VAT handling and invoices
- Kit (voorheen ConvertKit)free planThe list that makes the launch work
- beehiivfree planNewsletter as the distribution engine
Want the research?— what it is, what it costs, and how we scored it
What you can actually watch
A product page, then purchases arriving with nobody touching them.
What it is
You make something once — a guide, a template, a workbook, a Notion system, a dataset — and sell it repeatedly through your own store or a marketplace, with AI as the production lever.
Why this opportunity
The unit economics are genuinely attractive: margin above 90% per sale and scalability scored 9 of 10, the highest of the twelve. Production cost has collapsed, which is the part everyone repeats.
Who this is for
For someone who already has an audience, however small, and knows what that audience keeps asking for. Our research is blunt about the reverse: as a first step this is the classic inversion — building supply before you have demand. It is the right second step after a service business has earned you an audience.
The business model
Direct sale of a file or a hosted product, one price, no delivery cost per unit. Optionally bundled or moved to a subscription later, which is the only way this category gets recurring revenue.
How money is made
Price times volume, minus platform and payment fees. Margin per unit is above 90%, but acquisition cost is the number that decides whether this works — and it is the number the promise always omits.
Who pays
Consumers and freelancers inside one defined niche. The narrower the niche, the shorter the explanation, and the explanation is what you are really paying for with your time.
Finding customers
Your own audience. This is the heart of it: the product is not the business, the audience is the business. Without distribution, a digital product is a file on a hard drive.
Delivery
Topic, production, sales page, payment, delivery, support. All of it automates after launch, which is exactly why people mistake the easy half for the whole job.
Where AI actually helps
Very high in production, zero in distribution — and distribution is about 90% of the work. Your competitors' production costs fell exactly as far as yours did, so supply grows faster than demand.
Skills you need
Medium, and on an axis people do not expect: audience understanding and positioning, not production. Writing the thing is the part AI helps with; knowing what to write is not.
What it costs to start
Very low. A payment provider, a hosting page, and whatever tool you write in. There is no inventory and no minimum order.
How to price it
One-off price per product, with bundles as the main lever. Note the EU rules that apply to you as soon as you sell to consumers: the right of withdrawal for digital content, information duties, and VAT in the buyer's country.
Common mistakes
Building the product before the audience. Letting a model generate the whole thing: per the US Copyright Office (Part 2, 29 January 2025, Tier 1), prompting alone is not sufficient human authorship, so raw AI output is copyable on launch day. And pricing on effort instead of on what the buyer avoids.
Scaling
Scalability 9, competition 9, demand 5. You scale by selling more to the same audience — a second and third product for people who already bought — rather than by chasing new traffic for a single file.
The version that holds up
Sell the product to buyers you already have a relationship with, then use their questions to write the next one. A three-product catalogue for one narrow audience beats one product chasing a broad one.
Time to first revenue
2–26 weeks
Two weeks to six months or more, and the spread is the whole message. The low end assumes you already have an audience of some size, you sell something that audience asked for, and you price it low. Without an audience the top end is open: nothing brings traffic to a new digital product except work you do yourself.
How it scored
Ten axes, scored. Risk scored 4 out of 10, where higher is worse.
The state of this market
Competition scored 9 out of 10 and demand only 5. The hard part is the audience, never the product. And per the US Copyright Office, unedited AI output is not protected — substantial human work is what makes it yours.
Not sure this one fits you?
Answer a few questions about budget, time and skills and see which of the eighteen match your situation — and which to avoid.
Run the Opportunity Finder