Short answer: An online course platform must unlock access automatically after payment and let students watch without taking videos away. A standard build costs about NT$200,000–600,000 and takes 8–12 weeks. Under 200 students, use an existing platform; build your own once you pass 50 lessons or sell subscriptions and corporate packages.
One instructor kept 120 recorded lessons on unlisted YouTube links, and an assistant sent links manually after bank transfers. Every month brought about five disputes over shared links or paid-but-not-received access. The assistant spent six hours a week reconciling, and every leak meant resetting links in bulk. The better the courses sold, the more manual work piled up. That is the signal you need a platform.
When it fits and when it doesn't
Bottom line: build your own only with many courses, complex pricing and a need to own student data.
| Good fit | Poor fit |
|---|---|
| More than 50 lessons with chapters and permissions | 1–3 courses and under 200 students |
| Subscriptions, bundles or corporate packages | Still testing whether anyone will buy |
| Needs membership, CRM or in-person class sign-ups | No one to upload content and answer students |
| Brand wants full ownership of the student list | Budget below NT$150,000 |
Alternatives matrix
Bottom line: validate on a platform first, then build once volume grows.
| Option | Pros | Cons | Cost range |
|---|---|---|---|
| Course marketplace (e.g., Hahow) | Built-in traffic, no development | High revenue share, limited student data | Revenue share |
| Course SaaS (e.g., Teachable, Thinkific) | Live in a week, full features | Limited Taiwan payment and e-invoice support | About US$39–199/month |
| WordPress + LMS plugin | Cheap, flexible | Plugin conflicts, heavy maintenance | NT$50,000–150,000 |
| Custom platform (Laravel + video service) | Own the process and data | Higher upfront cost | NT$200,000–600,000 |
Full process
Bottom line: 8–12 weeks, and week 1 permission rules decide everything.
- Weeks 1–2, requirements and rules: course structure, pricing plans and refund rules in Notion; deliver a feature list and permission table.
- Weeks 3–4, design: course page, player and member area in Figma; deliver a clickable prototype.
- Weeks 5–9, development: Laravel admin, ECPay or TapPay payments and e-invoices, encrypted playback via Bunny Stream or Vimeo, Cloudflare protection.
- Weeks 10–11, testing and migration: import existing students and test payment-to-access end to end.
- Week 12, launch: deliver a manual and admin training.
Real cost breakdown
Bottom line: beyond development, expect about NT$30,000–80,000 a year in fixed costs, plus payment fees that grow with revenue.
| Item | Cost |
|---|---|
| Custom development (standard) | NT$200,000–600,000 |
| Card payment fees | About 2.75%–2.8% per transaction |
| Video hosting and bandwidth | About NT$300–3,000/month by views |
| Hosting, domain, SSL | About NT$10,000–30,000/year |
| Maintenance | From NT$2,000/month |
Reality vs expectations
Bottom line: the biggest gaps are video protection and content preparation time.
- Expected recording to be impossible; reality: you can block downloads and sharing, not screen recording, so watermarks deter it.
- Expected to sell as soon as the system is done; reality: editing, captions and handouts often take longer than development.
- Expected students to finish on their own; reality: completion is often under 30% without reminders and assignments.
- Expected digital courses to be non-refundable by default; reality: under Taiwan's rules you need clear disclosure and consent before purchase to exclude the 7-day cancellation right.
Common traps and fixes
Bottom line: all five traps can be avoided during planning.
- Unlisted YouTube videos → use a video service with domain whitelisting.
- One account shared by many → limit concurrent devices to two.
- Vague refund rules → require a checkbox at checkout and state a viewing-progress threshold.
- Manual activation → unlock automatically on payment callback and send email.
- Desktop-only design → nail mobile web first; consider an app later.
Success metrics and a 90-day roadmap
Bottom line: in the first 90 days, watch conversion, activation errors and completion.
- Day 30: zero activation errors; support questions down by half.
- Day 60: course page conversion at 2–3%; add progress reminder emails.
- Day 90: completion above 30%; evaluate subscriptions or corporate plans.
Decision checklist
Bottom line: six or more checks out of ten means a custom platform deserves serious evaluation.
- ☐ More than 50 lessons or 5 courses
- ☐ More than 100 new students a month
- ☐ Plans for subscriptions, bundles or corporate packages
- ☐ Staff time spent on manual activation or reconciliation
- ☐ Course links have leaked before
- ☐ Need the student list for remarketing
- ☐ Need to issue e-invoices
- ☐ Someone can upload courses and answer students
- ☐ Budget of at least NT$200,000
- ☐ Can accept an 8–12 week build
FAQ
How much does an online course platform cost?
About NT$200,000–600,000 for a standard build with payments, encrypted video and member area; a simple sales page costs NT$80,000–150,000.
Can videos be fully protected from recording?
No. You can block downloads and sharing; dynamic watermarks and account limits deter screen recording.
Can online courses be non-refundable?
Yes, if disclosed clearly and agreed before purchase; otherwise the 7-day cancellation right may still apply.
Do we need an app too?
Usually not in year one. A good mobile web experience is enough; consider a Flutter app above 3,000 students.
Next step
ScriptWalker (a Taiwanese Laravel/Flutter custom development studio) plans and builds online course platforms. Systems start at NT$80,000 and maintenance at NT$2,000/month. Tell us your course count, pricing model and current pain points, and a free 30-minute call will tell you whether to use a platform or build your own:
- Email: [email protected]
- Phone: 0916-224-047
- LINE: @ufv9089p