How to create and sell an online course: 7 steps
How to create and sell an online course in 7 steps: a specific promise, a pre-sale, structure, production, pricing and the numbers, with free calculators.
An online video course is one of the most direct ways to turn what you know into a product, but it is easy to get the order wrong: you buy gear, record thirty lessons and only at the end find out nobody wants that course. This guide puts the steps in the order that costs the fewest mistakes.
An honest limit: nearly everything you find about this comes from people who sell platforms or courses about courses, not from studies of creator results. So we give you the method and ask you to check every number against your own data.
How do you create an online course, in short?
You create an online course in seven steps: choose a specific promise, validate it with a pre-sale, structure modules and lessons, produce the videos, pick where to host and sell, set the price and launch. The order matters: first check that someone will buy, then produce. The reason for this order is cost. Every step you finish before checking demand is time and money you cannot recover: recording thirty lessons for an audience that never asked for them is the most expensive way to learn that nobody wants a course. A pre-sale, even to fifty people, turns the question into a number you can read before you spend weeks on production, and it gives you the first students, who will tell you what is unclear.
- Choose a specific promise about a result for a defined group.
- Validate with a pre-sale or a survey.
- Structure the course into short modules and lessons.
- Produce with the minimum gear that holds up.
- Choose where to host and sell.
- Set the price by running the numbers.
- Launch to your audience and collect first feedback.
The seven steps, one by one
The seven steps to create an online course are: a specific promise, validation through a pre-sale, a module structure, lean production, choosing a platform, a priced-out offer and launch. Each step has a precise job, and skipping validation costs the most, because you find out too late whether anyone will buy.
Each step has a precise job: skip the second and you risk paying for it at the sixth.
Step 1: choose a specific promise
A good online course promises a concrete result, for a specific person, in a defined time. “Learn photography” says nothing; “take portfolio-ready product photos with your smartphone in 14 days” does. That is the example Hostinger uses in its guide, and it works for any subject (Hostinger).
Try writing the promise in one sentence: “After this course, [who] will be able to [do what] in [how long].” If you can't finish it, the course is still a topic, not a product.
Step 2: validate before you produce
The most solid way to validate is to sell before you build. Hostinger recommends offering a beta to a small group at a discount before building the full course (Hostinger), and Kajabi notes that a pre-sale proves demand with money instead of compliments, warning that underpricing it is one of the most common mistakes (Kajabi).
Other useful signals are questions repeated in comments and forums, and a short survey with questions like “What is your biggest challenge with this topic?” and “What would you pay to solve it?”, which are common suggestions in creator guides. A useful rule is to decide beforehand what counts as a green light: for example, a minimum number of pre-registrants by a set date.
A caution: these are vendors' and creators' advice, not measured results. We found no independent studies on how well pre-selling works, and a good outcome does not guarantee the course will succeed.
A template for the pre-sale message
A pre-sale message works if it is short, specific and honest. A structure to adapt: one line on the problem the course solves, the promise in a sentence, what people who join now receive (early access, a live session, the launch price), the opening and closing dates, and a clear invitation to reply or reserve a seat.
Say what is not ready yet: “The first three lessons are recorded, the rest arrive weekly.” That honesty lowers refunds and gets you first feedback while the course can still change. If nobody answers, that is valuable information: it costs far less to find out before recording thirty lessons. Use the replies to fix the promise, not to defend the idea. And decide in advance how many pre-registrations you need to go ahead.
Step 3: structure modules and lessons
A simple structure beats a rich one: a few modules, each with one goal, and short lessons you can watch on a break. Length comes from the result you promise, not from the number of hours you want to sell.
To see how much work you are taking on, use the course planner. With the default values (20 lessons of 12 minutes, 4 hours of work for every hour of video, 6 hours a week) the result looks like this.
- Hours of finished video
- 4
- Hours of work
- 16
- Weeks at 6 hours
- 3
The multiplier is the delicate part: if you edit heavily, add graphics or reshoot, it can be much higher. We found no independent benchmark: the only figure we came across comes from an AI-avatar video vendor and doesn't generalize. Time yourself on the first two lessons and correct the planner.
Step 4: produce with the minimum gear
To start, good light, a clean microphone and a recent phone are enough. Audio weighs more than picture: a student will put up with a simple video but not with sound that is hard to follow. Record the first two lessons, show them to someone in your audience and fix things before you carry on.
Keep lessons between 5 and 15 minutes. Put the lesson's result at the start and an action to take at the end. If you plan handouts or exercises, prepare them during production, not after.
Step 5: choose where to host and sell
You have two separate jobs: taking payments and delivering lessons. Some platforms do both, others only one. Our platform comparison has prices, fees and limits from the official price lists.
Don't choose on the lowest subscription: look at transaction fees, product limits and what happens to your data if you switch. And decide early how to protect the videos: a course that costs hundreds deserves a per-student watermark and a device limit, as we explain in Per-student watermarks for video courses.
Step 6: set the price by running the numbers
Start from the value of the result for the buyer and compare it with what competitors charge. Then check your margin. In the course profit calculator enter price, expected sales, production cost and fees.
At a $149 price, 50 sales, $1,500 of costs and a combined fee of 5% plus 30 cents, each sale leaves $141.25. You need 11 sales to recover the costs, and at 50 sales profit is $5,562.50. These are examples, not forecasts: change the values to your own and try a pessimistic scenario with half the sales.
Step 7: launch and listen
The first launch is mostly for learning. Offer the course to your audience with an opening date and a closing date, and collect first feedback. The people who pre-bought are your first witnesses.
Your audience lives on social platforms, but the only channel you really own is your contact list: that is the subject of our article on creator newsletters. To see where a course sits in the wider picture of income, read how creators make money.
What happens after launch: completion and leaks?
Expect few people to finish. A 2013 study of 29 MOOCs found an average completion rate of 6.8%, with a low of 0.8% and a high of 19.2%, counted, for most of the courses, as the share of enrollees who earn a certificate (Inside Higher Ed).
See the data as a table
| Value | Completion |
|---|---|
| Lowest | 0.8% |
| Average | 6.8% |
| Highest | 19.2% |
A paid creator course is something else, and the data is old: it is there to remind us not to confuse sales with completions. The best way to improve the second is brevity, real support and a community. If you want to know what happens when a course ends up in other hands, our guide to course piracy explains what can be done, and the first-48-hours checklist gives you a ready-made template.
How much time and work does it really take?
Time depends on how much editing you do. With the production planner you can try scenarios. With 20 lessons of 12 minutes (4 hours of video) and 6 hours a week, a multiplier of 4 gives 16 hours of work, or 3 weeks; at 8 it becomes 32 hours and 6 weeks; at 12, 48 hours and 8 weeks. These are the tool's calculations, not benchmarks.
Three habits shorten the time: writing an outline for each lesson before recording, recording several lessons in a row with the same light and sound, and settling for “good enough” on the first edits. If you sell through a pre-sale, you also have a deadline that forces you to finish.
How do you write the sales page?
The sales page answers six questions, in order: who the course is for, what result it promises, what it contains, how it works (length, format, support), why to trust you (examples, testimonials, your background) and what it costs, with the refund policy. The FAQ closes the doubts that would otherwise arrive in private.
Put the promise in the headline, the programme in a readable list and the price near a clear call to action. Don't promise results you can't guarantee: write what the student will do, not what they will achieve. And say honestly who it is not for: it lowers refunds and raises trust. For details on disclosures and sponsorships, see media kit and rates.
How to set pricing in tiers
A common structure is to offer three tiers: a base edition with the lessons, a complete one with materials and exercises and one with direct support (a call or a review). They do two things: let people with less budget in and give people who want more help a way to pay more. There is no right number of tiers for everyone: start with two and add the third only if you get requests.
Use the calculator for every tier: the supported tier costs your time, so check the margin holds even with few buyers. And decide beforehand how you will handle refunds: a clear policy written on the page lowers disputes.
What should you do in the first thirty days after launch?
In the first thirty days, listen; don't change everything. Read every student reply, note the questions that repeat and fix first the audio or content errors that block understanding. A short message at the end of the first week (“How is it going? Is anything unclear?”) gives you information no survey replaces.
Check the essential numbers too: how many started, how many reached the third lesson, how many support requests you got. If many stop at the same lesson, the problem is that lesson. After a month you have the data to decide what to improve, whether to open a second edition and which part of the course to turn into free content to bring in new students.
What should you check before launch?
- The promise fits in one sentence.
- You have at least a pre-sale or a list of interested people.
- The first two lessons have been seen by someone in your audience.
- You have chosen where to take payments and deliver, and know the fees.
- You have a watermark or a device limit, if the course is your main product.
- Your tax side is in order: if you don't know whether you need a VAT number, read the guide.
Key terms. A pre-sale is selling a course before it is built, to test demand. A production multiplier is the number of work hours per hour of finished video. A completion rate is the share of enrolled students who finish a course.
How we checked this. We ran the production planner with multipliers of 4, 8 and 12 and found 3, 6 and 8 weeks of work for the same four hours of video. We found the completion figures in the Inside Higher Ed report and checked them there. We found no independent benchmark for production time, so we say so. Contact: hi@heytea.app. Editorial note: we fact-check every figure against the sources listed below and correct mistakes openly.
Frequently asked questions
How do you create an online course from scratch?
Start with a specific promise about a result a defined group of people wants, test it with a pre-sale or a survey, design a structure of short modules and lessons, record the videos and choose where to host and sell them. Only then do you set the price and launch. Validating before you produce avoids investing months in an idea nobody buys.
How much does it cost to create an online course?
It depends on how much you do yourself. Out-of-pocket costs can be low if you record with a phone and a microphone, but the real cost is time: the default in Tea's free planner is 4 hours of work for every hour of video, so 16 hours for a 4-hour course. For your own case, use the course profit calculator with your numbers.
How much should I charge for an online course?
There is no single right price: it depends on the result you promise, the audience and what competitors offer. Start from the value of the result for the buyer, not from the number of video hours, and calculate how many sales cover your costs. You can then test two prices with two groups of pre-registrants.
How many students actually finish an online course?
Few. A 2013 study of 29 MOOCs found an average completion rate of 6.8%, with a low of 0.8% and a high of 19.2%. It covers open, free university courses, not paid creator courses, so it is an order of magnitude and not a forecast. A short course with deadlines and support tends to do better.
Do I need to protect a course before selling it?
If the course is your main product and costs hundreds, think about it before launch, because after a leak you can't recall anything. No protection stops someone from filming the screen, but a per-student watermark and a device limit make a copy riskier for whoever makes it and traceable for you.
Sources
- Hostinger, guide to selling courses online in six steps. A vendor source. hostinger.com
- Kajabi, “How to Sell an Online Course Before You Build It”. A platform vendor's source. kajabi.com
- Inside Higher Ed, 10 May 2013, on Katy Jordan's study of MOOC completion rates. insidehighered.com
- Tea's free tools: course profit calculator and production planner.