Create a Course to Sell Online: 90-Day Roadmap
9 min read ยท 2026-10-08
To create a course to sell online, first prove that people will pay for one specific result. Then pre-sell it, build a lean version with your first students, pick where to host it and launch to a list. The order matters most. If you build the whole course before anyone pays, you risk weeks of work on something nobody buys.
This roadmap fits into 90 days and is split into phases you can put on a calendar. Each phase has a deliverable, a milestone and a go/no-go check. The checks tell you when to keep building and when to change the offer. The plan works if you already teach your topic at work, consult on it, or are a few steps ahead of the people you want to help.
The roadmap at a glance
Goal: Launch a focused online course that delivers one clear result, and sell it to a first group of paying students within 90 days. Duration: 13 weeks
Validate the idea (Weeks 1 to 2)
Find a course promise that people already want and would pay to reach faster.
- List the questions people ask you again and again at work, in forums or in your messages.
- Read course reviews, Reddit threads and YouTube comments on your topic, and note the complaints.
- Write the promise as one sentence: who it is for, where they start, what they can do at the end.
- Talk to 8 to 10 people who match that student. Ask what they tried and where they got stuck.
- Choose a format: self-paced, a live cohort, or self-paced lessons with live calls.
Milestone: A one-sentence promise that several target students describe as their own problem.
Pre-sell (Weeks 3 to 4)
Get paying commitments before you record a single lesson.
- Write a simple sales page with the result, the module outline, the format and a start date.
- Offer a founding-student price in exchange for feedback and an honest review.
- Message warm contacts directly, post in communities you already take part in, and email your list if you have one.
- Run a free workshop that teaches one useful piece and invites people to join.
- Set the minimum number of buyers you need, and promise a full refund if you miss it.
Milestone: Your minimum number of founding students has paid, or you have refunded them and changed the promise.
Design the curriculum (Weeks 5 to 6)
Turn the promise into a short sequence of modules that each produce something.
- Work backward from the final result and list the skills students need, in order.
- Group them into 4 to 8 modules, each ending with a concrete deliverable.
- Write one learning objective per lesson and cut anything that does not serve it.
- Build the worksheets, templates and checklists students will use to apply each lesson.
- Decide how students get help: a community, office hours or email.
Milestone: A full outline with objectives, exercises and a support plan, shared with founding students.
Produce and deliver (Weeks 7 to 10)
Record clear, short lessons and release them while founding students work through them.
- Set up a quiet room, a decent USB microphone and soft light.
- Write bullet-point notes rather than full scripts so your delivery stays natural.
- Record screen lessons with Loom, OBS or ScreenFlow, and add captions afterward.
- Release one module per week and collect feedback before recording the next.
- Note every question students ask, because each one is a lesson to fix or add.
Milestone: All modules delivered, with at least one student result you can describe.
Set up and launch (Weeks 11 to 12)
Package the course and open it to a wider audience.
- Move the course onto your chosen platform and run a full test purchase.
- Update the sales page with founding-student feedback and real outcomes.
- Plan a launch sequence: a free lesson, a student story, enrollment opening, a firm deadline.
- Host a live Q and A and answer every pre-sale question personally.
- Set up your refund policy and tax settings in checkout before enrollment opens.
Milestone: Public enrollment closed, with buyers beyond the founding group.
Review and plan the next sale (Week 13)
Use what students did to improve the course and decide how you will keep selling it.
- Check where students stop watching or stop submitting exercises.
- Survey students on what helped most and what was missing.
- Re-record or split the lessons that caused confusion.
- Choose between repeat launches and an always-open course with an email sequence.
Milestone: A revised course version and a written plan for the next sales cycle.
Is your course idea worth building? A go/no-go grid
Most guides tell you to validate your idea but never say what counts as a pass. Use this grid at the end of Week 2 and again at the end of Week 4. If an answer is no, fix that point before you move on.
The money question decides everything. The simplest test is to describe the result and who it is for, then offer early access at a reduced price. Set your minimum number of buyers before you start, based on what makes the work worth it for you. If you reach it, build. If you fall short, the offer is wrong, not your effort. Change the promise, the student or the format, and run the pre-sale again.
- Specific student: can you describe their job or situation and their starting level in one line?
- Visible result: can a student show the result, such as a finished portfolio, a passed exam or a working setup?
- Proof: have you, or people you helped, already reached that result?
- Urgency: is there a reason to solve it this quarter, such as a new role, a deadline or a lost client?
- Money on the table: did people pay, or only say they would?
How to price an online course
Price the result, not the hours of video. A two-hour course that gets someone through a certification is worth more to that buyer than ten hours of general theory. Live support, feedback on exercises and a community all raise what you can charge, because they raise the chance that the student actually gets there.
Look at what courses on the same topic charge before you decide, and note what they include. Then apply two rules from the founding-student phase. First, charge something real, because free sign-ups do not tell you whether anyone will buy. Second, raise the price once you have student results to show. Payment plans help with higher prices. Turn them on in checkout before launch rather than adding them when someone asks.
The three tiers below are a simple way to set a first price.
- Short, narrow course that solves one small problem: keep it at the low end of what similar courses in your niche charge.
- Full course that takes a student from start to a finished result: price it against the value of that result.
- Cohort or course with coaching: price it highest, because your time and feedback are part of what they buy.
Marketplace or your own platform: where to sell your course
You have two models. A marketplace like Udemy brings you buyers you did not find yourself. In return, it controls much of the pricing and takes a share of each sale. Check the current instructor terms before you decide. A course platform such as Teachable, Thinkific, Podia or Kajabi charges a subscription. You set the price, own the student list and keep more control over each sale. Payment processing fees usually come on top, so check them on your payment provider's pricing page.
Pick based on how you will sell, not on feature lists. Before you commit, confirm that you can export your student list. Then work out your break-even: monthly platform cost divided by your net income per student. That tells you how many sales a month keep the course from losing money.
- No audience yet and no plan to build one: a marketplace gets you a first test.
- An email list, clients or a work network: your own platform keeps the margin and the relationship.
- Live cohorts: make sure the tool handles start dates, live sessions and a community, or pair it with one that does.
How to create an online course that sells: structure and production
A course sells again and again when students finish it and talk about their result. That depends on structure far more than on cameras. Give each lesson one idea and one action. Let the idea set the length, and split any lesson where students ask the same question twice.
Spend your production time on what students use.
- A worksheet or template for every module
- A short checklist that sums up each lesson
- One worked example that matches your student's real situation
- Clean audio, because students accept plain visuals much more easily than echo or background noise
Online course launch without a big audience
A launch works because it concentrates attention into a short window with a real deadline. You do not need a large following. Your founding students, direct messages to people who fit the profile and one live workshop can fill a first round. Partnering with someone who already serves your students, through a joint workshop or a guest lesson, can bring more.
Keep deadlines honest: a cohort start date, a founding price that ends, or a bonus tied to a live call. Between launches, publish useful content on one channel and offer a free checklist on your topic so your list keeps growing.
Refunds, taxes and legal basics
Most course guides skip this, and it causes trouble after the first sales. Write your refund policy before launch and show it on the sales page, so buyers know the terms before they pay.
Tax on digital products can depend on where your buyers live, not only where you live. Some platforms collect and pay these taxes for you, and others leave it to you. Ask your platform how it handles this, and confirm your situation with an accountant. Also add terms of sale, a privacy notice and a line on who owns the course materials.
Common mistakes to avoid
- Recording the full course before anyone pays, when pre-selling to a founding group would prove demand first.
- Promising a vague topic like "marketing basics", when a specific result for a specific student is what people buy.
- Putting everything you know into the course, when only what moves students toward the promised result belongs in it.
- Spending weeks on video gear, when clear audio, short lessons and good exercises matter more.
- Launching with no student results, when founding students can give you the proof your sales page needs.
- Treating the course as done at launch, when completion data shows exactly which lessons to rework.
Frequently asked questions
How long does it take to create an online course to sell?
A focused course can be validated, built and launched in about 13 weeks if you pre-sell and record alongside a founding group. Bigger courses with many modules and polished video take longer. A lean first version, improved with student feedback, is usually the faster path to sales.
How much should I charge for an online course?
Base the price on the value of the result and the level of support, not on video hours. Short, narrow courses sit at the low end, full courses in the middle, and cohorts or coaching at the top. Start with a founding price, then raise it as results come in.
Should I sell on Udemy or on my own site?
A marketplace brings buyers but takes a share of each sale and controls much of the pricing, so check the current instructor terms. Your own platform costs a subscription, but you set the price and keep the student list. If you already have clients, a network or a list, your own site usually makes more sense.
Can I create an online course that sells without a big audience?
Yes. A small group that trusts you can fill a first cohort. Direct outreach, relevant communities, a free workshop and partnerships with people who serve the same students are enough to get started.
What is the minimum viable course to launch?
It is the shortest path to the promised result: a few modules, each with a lesson, an exercise and a deliverable. Live sessions can stand in for recorded lessons in the first round. Anything students did not need can wait for version two.