Going global: how a Ukrainian online school sells courses to English-speaking markets

What usually pushes an online school abroad is arithmetic, not ambition. The niche dries up, cost per lead climbs with every launch, and the same course that sells for the equivalent of $150 at home goes for $300–400 in Poland, Canada or the UK. Not because it got better — the buyer simply has different purchasing power and different alternatives.

But "going international" is not "translating the landing page". Here is the order of steps that saves six months and a few thousand dollars spent in the wrong direction.

Step 1. Check whether your product travels at all

• Tool-based courses — Figma, paid ads, Excel, 3D, prompt engineering. These travel best: the software looks the same everywhere.
• Country-bound courses — local tax regimes, national law, university admissions, the local job market. These don't travel at all. You need a new product, not a translation.
• Soft skills — negotiation, burnout, management. They travel, but the examples have to be rebuilt.

Quick test: take five random lessons and count the mentions of local currency, local services, local cases and local law. If more than a third of the lessons are affected, you're rewriting the course, not translating it.

Step 2. Narrow down to one segment

You're really choosing one of four entry points:

1. Your own diaspora — the course language stays the same, only the payment setup and context change. Cheapest start.
2. Markets where English is the working language but prices sit below US levels. Lower revenue per student, lower production expectations.
3. US / UK / Canada directly. The most expensive traffic and the highest bar for video quality, support and guarantees.
4. B2B corporate training. One contract replaces 30 retail sales, but the deal cycle runs 1–3 months.

Launch into one segment. Two audiences at once means two landing pages, two funnels and zero focus.

Step 3. Price from alternatives, not from exchange rates

The classic mistake is taking your local price, dividing by the exchange rate and landing on $150. Price on a new market is set by what else the buyer could buy. If comparable courses run $400–600, a $150 offer reads as suspicious, not as a bargain.

What works: find 5–7 competing courses in your topic on that market, write down their prices and what's included, take the median, and don't start below it. Win on what they don't have — homework review, feedback from the author, community, post-course support — not on price.

Build the extras into the price from day one: card processing fees (2–4%), possible tax in the buyer's country, refunds, and support hours in a foreign time zone.

Step 4. Payments and the legal wrapper

• Card acceptance. Most schools end up running two checkouts: a local one for local currency and an international one (Stripe, Paddle or a merchant of record) for USD and EUR. One checkout won't cover both.
• A legal entity with a foreign-currency account. Selling digital services to non-residents is a separate regime with separate reporting.
• Tax in the buyer's country. In the EU, digital services sold to consumers are often taxed with the buyer's local VAT. Talk to an accountant before the first sale, not after the hundredth.
• Terms and a refund policy in English. On Western markets a clear refund policy lifts conversion more than another testimonial block.

Step 5. Translation is not localization

1. Text first: landing page, emails, lesson descriptions, the student dashboard.
2. Subtitles next, on the video you already have. English-speaking students learn from subtitles routinely — it's a standard, not a compromise.
3. Re-recording video comes only after the first sales.

An accent isn't the problem. The problem is examples that mean nothing to the buyer: local currency amounts, local services, untranslatable jokes. Replace those and 80% of localization is done.

Step 6. Where the first traffic comes from

Paid ads to a cold international audience are the worst way to test the hypothesis: you pay $3–8 per click and still won't know whether the offer, the price or the page failed.

What works early: English-language content on YouTube and LinkedIn, answering in topical communities, partnerships with local experts who bring the audience while you bring the product, joint webinars. Turn ads on once the funnel already converts warm traffic.

Step 7. What breaks operationally

• Time zones. A 7 p.m. live session in Kyiv is noon in New York and 9 a.m. in California. Either move it or switch to recordings with asynchronous Q&A.
• Support speed. "We reply within 24 hours" reads as disrespect on Western markets.
• Homework deadlines should be tied to the student's day-since-enrollment, not to a calendar date.

A minimal 90-day plan

• Days 1–14. Pick the segment, analyse 5–7 competitors, lock the price and the English offer.
• Days 15–45. Landing page, terms, checkout, email sequence, subtitles for 20–30% of the course.
• Days 46–60. Pre-sell to a small warm audience: 10–20 seats at an early-bird price. The goal isn't revenue — it's proof that people can pay by card and reach the dashboard.
• Days 61–90. Fix localization based on the first students' feedback, then switch on paid traffic.

No sales by day 60? The problem is almost always the price or the segment, not the translation.

What this requires from your platform

A second market means a second set of everything: pages, emails, currency, funnel. If each piece lives in a separate service, a two-market school burns twice the team's time.

We're building CREO so all of it sits in one place: a builder for sales pages, a CRM, funnels in Telegram, and payments through both WayForPay and Stripe — local currency and dollars without a second service. Gamification keeps course completion at 70–80%, which matters on a market where student reviews outweigh advertising. More than 100,000 students have already gone through the platform, and support replies in about 15 minutes. We're also building MCP control of the school, where an AI agent creates courses, funnels and reports instead of manual admin work — beta testers get it first.

Beta testing starts in August 2026. Join the early-access list at platform.creo.ua.