Online Course Creator

The shortcut: Sell the course before you build it. Pre-sell to 20-40 people with a 3-week recorded delivery window, and you've validated demand and funded production in the same week.

Industry: Media & Content Investment level: Micro — $300-$1,000 Time to launch: 6-10 weeks from idea to first paid student Best for: Someone with a teachable skill peers already ask about — a freelancer, a senior IC at work, a coach, a tradesperson with a system. You don't need credentials. You need a result you can prove and 200-500 people who know you exist. What you'll likely make: $500-$2K/month by month 6, $3K-$8K/month by month 12. Math is in Section 4.

Market Opportunity

Most new course creators spend three months perfecting curriculum and end up with a polished product nobody asked for. The fix isn't better marketing later. It's selling the course before you film a single lesson — which is the part that scares people, and the part that separates a real business from a hobby.

The demand side is real. US digital ad spend hit ~$298B in 2024, with creator-economy share growing fastest (IAB), and a chunk of that traffic ends on landing pages selling courses. But you don't need a national audience. You need a niche where 200 people will pay $200-$500 to skip a learning curve you've already climbed.

The platforms have matured to the point that the technical side is a non-issue. Teachable, Kajabi, Thinkific, and a half-dozen others handle hosting, payments, and student logins. What hasn't gotten easier is the part nobody can do for you: picking a niche specific enough that your offer reads as "for me," not "for everyone." A "productivity course" loses to "Notion templates for solo law-firm intake." A "marketing course" loses to "Google Ads for HVAC companies under $2M revenue."

If you can name three specific people who'd buy your course tomorrow at $300, you have a niche. If you can't, you don't yet — keep narrowing.

Launch With AI

Pro section. AI doesn't replace teaching — it cuts the parts that drained you (writing the pre-sale page, drafting lesson scripts from your raw outline, building the student email nurture, generating FAQ from cohort 1 transcripts). Spend the saved time on what AI can't do: getting on a Zoom with the panicked student in week 3, refining the cohort's curriculum based on the actual questions they asked, and answering DMs in the community channel like a human, not a bot.

The trap most first-year course creators fall into: they let AI write the lessons. Backwards. AI that writes lessons for you produces flat content that students watch at 1.5x and finish at 22%. The craft is YOUR voice + AI's speed on the writing tail (sales page, emails, FAQ, transcripts). Students paid you for your specific take — keep that the part you record.

Important up-front: AI cannot teach the cohort, run the live Q&A, or refund a student who churned because you didn't deliver. It will also confidently write earnings claims that violate FTC 16 CFR Part 255.2 ("our students average $50K in 90 days" — that's an outlier claim without typical-results disclosure). You own every claim, every refund, every methodology decision; AI scales the writing around them.

AI Tools You'll Use

Tool	Price	What it does
ChatGPT Plus	$20/mo	Pre-sale page copy, lesson script outlines, student emails, FAQ, retainer report
Claude Pro	$20/mo	Long-context curriculum review (paste cohort 1 transcripts → cohort 2 outline)
Descript	$24/mo	Edit lessons by editing the transcript, auto-cut filler words, AI voice clone for retakes
Kit (formerly ConvertKit) AI / Beehiiv AI	free-$25/mo	Email automation, paid-tier nudges, abandoned-cart for cohort enrollment
Loom AI	free	Pre-sale video pitches, weekly Q&A recordings, student walkthrough recaps

The Workflow

Pre-sale page + waitlist email (ChatGPT, ~90 min one-time). The pre-sale page validates demand BEFORE you film. Paste:

"I'm pre-selling a cohort-based course on [outcome — e.g., 'land your first 5 freelance Webflow clients in 60 days']. Target: 15-25 paid students at $300-$500. Cohort runs [dates], 4-6 weeks recorded delivery + 3 live Q&As. Build: (a) the sales page (hero promise, 8-10 lesson titles, the live Q&A schedule, the price + payment plan, FAQ, testimonial placeholder), (b) a 5-email pre-sale waitlist sequence — email 1 (announcement), 2 (the trap most people fall into), 3 (curriculum reveal), 4 (24 hours left), 5 (last 4 hours). Stay inside FTC 16 CFR Part 255 — NEVER claim outlier earnings without typical-results disclosure. Tone: senior practitioner, NOT 'unlock your potential.'"

Send the waitlist sequence to your warm list (50-100 personal messages). Hit your 15-paid-student threshold or refund and rework the angle — that's the whole gate.
Lesson script outlines (Claude long-context, ~60 min/lesson). Paste your raw notes, slides, and any prior content into Claude:

"I'm filming a 25-min lesson on [specific topic] for my [niche] course. Below are my raw notes, prior blog posts, and 3 student questions from the cohort that this lesson must answer. Generate a lesson script outline: (a) hook (60 seconds — the specific pain this lesson solves), (b) 3 core concepts (each with 1 example), (c) the demo or screen-share moment, (d) homework for the week, (e) the 'common mistake' callout. Output as a teleprompter-ready script with timing notes. NEVER guess at numbers or examples I didn't provide."

Read the script aloud once before filming. Edit for your voice. The 25-min lesson takes 30 min to record vs 3 hours hand-written — that's 8x leverage on the build phase.
Descript transcript-based editing (Descript, ~30 min/lesson). Record raw, then edit the TRANSCRIPT (not the timeline). Filler words ("um," "like," long pauses) auto-detected and one-click removed. The AI voice clone fixes the 2-3 sentences you flubbed without re-recording. A 25-min lesson edits in 30 minutes vs 3 hours in Premiere.
Cohort student email nurture (Kit/Beehiiv AI + ChatGPT, ~90 min one-time). Paste:

"I run a 4-week cohort course on [topic]. Build a Kit automation for paid students: (1) Day 0 (immediately after purchase) — welcome + cohort schedule + community channel invite, (2) Week 1 (Friday) — 'how was lesson 1?' check-in + the 1 thing to focus on this weekend, (3) Week 2 — what to do if you're behind, (4) Week 3 — the 'apply this NOW' nudge with a real example, (5) Week 4 (graduation) — survey + alumni community + cohort 2 early-bird offer (20% off if they sign by [date]), (6) Week 6 (post-cohort) — 'did you ship?' check-in + testimonial request. Tone: warm operator, not corporate course platform."

The graduation→cohort 2 sequence is the cohort upsell — converts at 25-35% of cohort 1 grads at full price.
FAQ + cohort 2 outline from cohort 1 transcripts (Claude long-context, ~2 hours per cohort). After cohort 1, paste every Q&A transcript + community channel exports + survey responses:

"Below are the 3 live Q&A transcripts + community channel highlights + post-cohort survey responses from cohort 1 (15 students) of my [topic] course. Generate: (a) the top 12 questions students actually asked (with the answer I gave each time), (b) the 5 lessons cohort 1 students rated highest + the 2 they wanted MORE on, (c) the 3 lessons that didn't land — name what students struggled with and propose a rewrite, (d) cohort 2 outline incorporating these changes, (e) 5 testimonial-quality quotes I can use in marketing (with attribution placeholder for written permission). Tone: senior practitioner."

The FAQ becomes your sales-page section. The cohort 2 outline is your build plan. The student feedback compounds — cohort 3 rated 90% higher than cohort 1 because cohort 2's curriculum was student-driven, not creator-driven.

Time Saved Per Week

Roughly 8-12 hours/week during cohort + 6-8 hours/week between cohorts:

Pre-sale page + waitlist: 8 hours/launch → 90 min (ChatGPT)
Lesson script outlines: 3 hours/lesson → 60 min (Claude)
Lesson editing: 3 hours/lesson → 30 min (Descript)
Email nurture management: 4 hours/week → 30 min (Kit automation)
FAQ + cohort 2 outline: 8 hours post-cohort → 2 hours (Claude long-context)

Trade that time for: the live Q&A you've been dreading, the 1:1 with the panicked week-3 student, and the 5 LinkedIn posts that seed cohort 2's waitlist.

Total AI Stack Cost

Budget tier ($20/mo): ChatGPT Plus only. Kit free covers up to 10K subs; Loom free covers Q&A recordings; Descript free covers basic editing. Right for first cohort.
Full tier ($89/mo): ChatGPT Plus + Claude Pro + Descript + Kit (paid plan when subs > 10K). Worth it once cohort 2 enrolls — Descript alone saves 12 hours per cohort vs Premiere.
Compare: A part-time editor + email manager for a course business runs $1,500-$3,500/mo. The full AI stack is one-thirtieth that cost.

Cancel anything you don't open in a 7-day window. The trap is buying 2 course platforms — pick Teachable OR Kajabi (not both).

Your First Win

30 minutes from now your pre-sale page draft is written. Open ChatGPT (free tier works). Paste:

"I'm pre-selling a 4-week cohort course on [your specific outcome — e.g., 'how to land your first 5 freelance Webflow clients in 60 days']. Target: 15 paid students at $300. Generate the sales page (hero, 8-10 lesson titles, FAQ, refund policy, the 'who this is for / who this isn't for' filter, the price + 2-pay option, the cohort dates). Stay inside FTC 16 CFR Part 255 — never claim outlier earnings without typical-results disclosure. Output as Notion-ready Markdown."

Edit for voice. Send to 50-100 personal messages from your warm list. 15 paid students × $300 = $4,500 of validated demand BEFORE you film a single lesson — that's the page worth your first cohort revenue.

Product / Service Offering

Your first course is one cohort, one outcome, 4-6 weeks of recorded video plus live Q&A. That's it. Resist the urge to build a content library on day one.

The core offer. Pick one specific outcome ("Land your first 5 freelance Webflow clients in 60 days," "Pass the AWS Solutions Architect Associate exam in 8 weeks," "Run your first profitable Meta ad campaign for an e-commerce brand"). Price $300-$700 for cohort one. Scope: 8-12 video lessons, a workbook, a private community channel, and 3-4 group Q&A calls during the cohort window.

The validation pre-sell. Open enrollment for two weeks before any lesson is filmed. Set a minimum-viable threshold — say, 15 paid students at $300 — that proves the niche before you commit production time. If you don't hit it, you refund and rework the angle. This is the single most valuable habit you'll build, and it's what separates course-as-business from course-as-creative-project.

The progression. Cohort two is the same course, refined from cohort-one feedback, often at a higher price. Cohort three or four becomes a self-paced version at a lower price point ($99-$199) that runs evergreen. By month nine you have one premium cohort offer and one always-on self-paced offer feeding off the same content library.

Keep the production lo-fi. Loom or QuickTime screen recordings, a $60 USB mic, and natural light beat polished studio video for conversion in most niches. Students buy outcomes, not cinematography.

Revenue Model

You make money two ways: cohort launches (concentrated, higher price) and evergreen self-paced sales (steady, lower price). Most successful micro-creators run both within 12 months.

Path	Students/month	Avg price	Platform fees	Take-home/mo
First $1K month (one mini-cohort)	5 students	$300	~5-8% (Teachable Free or Stripe)	~$1,400 over 1 month
First $3K-$5K month (cohort + early evergreen)	8-12 cohort + 5-10 evergreen	$300 cohort / $149 evergreen	~5-8%	~$3,500-$5,200
12-month repeat-cohort path	20-30 students/cohort, 4 cohorts/year	$400-$600	~5-8%	$30K-$70K/year

Platform economics matter at scale. Teachable's Free plan takes a 5% transaction fee; Pro at $119/month removes it (Teachable). Kajabi takes 0% transaction fees but starts at $149/month and bundles email and community (Kajabi). Thinkific's free plan is unlimited students but capped at one course (Thinkific). Start on a free or low tier and upgrade when monthly revenue clears 10x the platform cost.

Avoid Udemy as your primary channel. Udemy keeps 63% of revenue when a student arrives via Udemy's promotion, and only takes 3% when you drive the traffic with your own coupon code (Udemy). It's a fine secondary discovery surface, never a primary business.

Startup Costs

Lean target: under $700 to first paid student.

Domain + landing page: $12 domain, free or $15/month on a platform like Carrd or your course host's built-in pages.
Email tool: Kit (formerly ConvertKit) is free up to 10,000 subscribers (Kit); Beehiiv is free up to 2,500 (Beehiiv). Either works for cohort one.
Course platform: $0-$119/month. Start on Teachable Free or Thinkific Free until you have students.
Microphone + lighting: $80-$150 total. A USB condenser mic and a window. Don't buy a camera yet.
LLC + EIN: $35-$500 depending on state (LLC University). EIN is free directly from the IRS (IRS EIN) — never pay a third party.
Payment processing: Stripe is built into every major platform. ~2.9% + $0.30 per transaction.

If you run paid acquisition for cohort one, budget $200-$400 for Meta or LinkedIn ads testing. Optional. Most successful first cohorts sell entirely from organic outreach to existing networks.

Legal & Formation

Business entity. A single-member LLC is the right call once you've collected your first paid student. Until then, sole proprietorship is fine — you can operate, invoice, and bank under your own name. The LLC matters when you start taking real revenue and want personal-asset separation. Filing fees range $35-$500 by state (LLC University). Get your EIN free at IRS EIN Online — formation services charge $50-$200 for the same form you can submit yourself in 10 minutes.

Licenses & sales tax. Course revenue is digital services in most states, and you generally have no sales-tax nexus until you cross ~$100K in a state ( specific state thresholds via your state department of revenue). A handful of states do tax digital products and SaaS — your course platform's checkout often handles this collection automatically if you enable it in settings. Check the platform docs before launch.

Industry-specific risk. The single biggest legal trap for course creators is testimonial and earnings claims. The FTC requires testimonials in marketing to reflect typical results, not outlier success stories — and "results not typical" disclaimers are insufficient. The typical result must be stated (16 CFR Part 255.2) (FTC Endorsement Guides). Translation: if you feature a student who made $50K from your course, you must also disclose what the typical student made. If you don't track it, don't feature outliers. Same rule for paid affiliate promotion — disclosure is mandatory in any sponsored content.

Marketing & First Customers

The first 30 students come from people who already trust you. Make a list of every coworker, ex-coworker, freelance client, online community where you're recognized, LinkedIn connection in your niche, and people who've DMed you for advice in the last 12 months. That's your launch list.

Send 50-100 personal messages — not a broadcast email — with one ask: "I'm running a small cohort on [outcome] for $300. Five spots. Want one?" This converts at 5-15% from warm contacts and is how most first cohorts fill.

Once cohort one is delivered, publish weekly content in the format your niche reads — LinkedIn posts for B2B niches, YouTube videos for skill niches, a Substack or Beehiiv newsletter for analytical niches (Substack takes 10% of paid subs plus ~3% Stripe processing — Substack Pricing). The content's job is to seed cohort two, not to sell directly. Each piece should make one specific point that demonstrates you've thought about the problem more carefully than the reader has.

Amy Porterfield's documented playbook — 90 days of email list-building before each launch — is the conventional cohort math. You don't need to copy her scale to copy the rhythm.

First 90 Days

Week 1. Pick the niche. Write a one-paragraph outcome statement. Test it against three people in the niche by DM — do they say "I'd buy that" or "interesting"? Iterate until it's the first.
Week 2. Build a one-page sales page with the outcome, 8-10 lesson titles, the cohort dates, the price, and an enrollment button. No course built yet.
Week 3. Open pre-sale to your warm list. Target: 15-25 paid students at $300-$500.
Week 4. If you hit your minimum, lock the cohort and start filming lesson one. If you missed it, refund everyone, gather feedback, and rework the angle.
Weeks 5-8. Film and release lessons weekly while the cohort runs. Host one live Q&A per week. Over-deliver on the community channel.
Week 9. Survey every student. Ask what worked, what didn't, what they'd pay double for. This builds cohort two.
Week 10. Publish three pieces of public content — a blog post, a LinkedIn case study, a free resource — built from cohort-one wins.
Weeks 11-12. Open cohort two enrollment at a 20-30% higher price. Target: 25-40 students. By the end of week 12 you should have run one cohort, banked ~$5K-$8K, and pre-sold the next one.

Common Pitfalls

Building before selling. You film 12 lessons, launch, and three people buy. Fix: never film a full lesson before you have 10+ paid students for that specific cohort.
Niche too broad. "How to be a better writer" doesn't sell; "How to write technical case studies for B2B SaaS marketing teams" does. Fix: if you can't name three real people who'd buy tomorrow, narrow until you can.
Outlier testimonials without typical-result disclosure. A featured "I made $30K" student without typical-results context is an FTC compliance gap. Fix: track and disclose typical student results, or don't feature earnings claims at all.
Premature platform spend. Paying $149/month for Kajabi before you have students burns $1,800/year on a tool you don't need yet. Fix: stay on a free tier until monthly revenue clears 10x the platform cost.

Get your full launch plan — take the free 60-second quiz.