Over 36 months with a kids' IT school, we spent a $820,250 budget and brought in 113,925 leads at an average of $7.2. Most importantly, we took the school from one country to 11 EU markets while keeping lead quality. A full breakdown of the strategy follows.
Context: who the client is and what the goal was
In 2023, an international chain of kids' IT schools teaching programming online to children aged 7–17 came to us. At the start, the school operated in one market and wanted to expand into Europe.
The brief was: “Scale the school to 5+ European markets in 12 months while keeping the cost per lead at $10 or less.” The budget was growing; we started at $8,000/month.
The strategy in three words: test, localize, scale
The biggest mistake most agencies make when entering new markets is copying a winning combination from one market to another. We knew it doesn't work: what resonates with Ukrainian parents won't resonate with Swiss ones. So we launched every market as if from scratch.
Stage 1. In-depth market research (2 weeks)
Before launching in any new market, our team carried out:
- Competitor analysis: which IT schools already operate in the country, their prices and positioning
- Customer development: 5-7 interviews with local parents, through personal contacts or Respondent.io
- Language nuance analysis: which terms to use (for example, “programming” in Czech is “programování”, but tech-savvy parents also understand “coding”)
- Lead magnet adaptation: in Switzerland a free trial lesson with a certificate worked better, in Turkey a 50% discount on the first month
Stage 2. Audience testing (1 month)
For each new market, we launched 8 different audiences in parallel, with a $300-500 budget each:
- Broad interest-based: parents aged 30-45 + the “Education” interest
- Parents' professions: IT specialists, engineers, doctors (high income)
- Lookalike of existing clients, 1% and 3%
- Lookalike of people who submitted a lead, 1%
- Kids' content interests: Minecraft, Roblox, LEGO
- Location + language: only Ukrainian speakers in the country (for Germany, Poland)
- Retargeting website visitors: for markets that already have traffic
- Education + premium income: parents ready to pay $200+/month
After 2-3 weeks, 2-3 clear winners emerged and the rest were switched off.
“The main insight from 3 years: a lookalike of clients always beat a lookalike of leads. Paying parents share behavior patterns that Facebook's algorithm picks up very well.” — Natalia Maslakova, founder of MAS Project
Stage 3. Creatives: the key success factor
Over 3 years we tested more than 400 creatives. In our data, the most effective formats were:
1. Video testimonials from student kids
Kids aged 10-14 talking about the first project they built: a game, a website, a bot. These creatives converted 2.4 times better than static banners. The cost per 1,000 impressions fell from $4-5 to $1.8-2.2.
2. “Before → After” carousels
Slide one: a child playing Minecraft on a phone. Slide two: the same child at a laptop coding their own version of Minecraft. The emotional shift from “consumer” to “creator” is the main trigger for parents.
3. UGC from parents
Videos of moms and dads describing how their child changed after 3 months of lessons. The highest conversions came from parents who work in IT, because viewers see expert validation.
Results by market
What didn't work (worth knowing)
We're not perfect, and we made mistakes that cost us budget:
- Launching in Switzerland without a separate test budget: we lost €1,500 testing combinations that didn't work. Lesson: premium markets need 2-3x the test budget.
- Trying to use one creative for all of French-speaking Europe: Swiss parents don't respond to creatives built around Parisian life. Lesson: even within one language, you need localization.
- Scaling the winner too early: we raised the budget from $1,000 to $5,000/day in one step and the cost per lead jumped from €5 to €18. Lesson: scale by no more than 30% a day.
Takeaways for your project
If you run online education for children and want to scale:
- Plan at least $3,000-5,000/month of budget for each new market in the first 3 months
- Don't copy a winning combination: localize every market from scratch
- Invest in video creatives with real students (UGC from parents adds 40% to conversion)
- Use lookalikes of paying clients, not leads
- Scale the budget gradually: +30% a day at most
Want the same result for your online school?
In 30 minutes we'll review your niche and show you how to scale without wasting budget.
Get a free consultationFrequently asked questions
What budget do you need to launch paid social for a kids' IT school?
The minimum test budget is $1,500-3,000/month: $500-1,000 to test audiences and creatives, the rest to scale the winner. In our data, stable performance requires $5,000+/month. In the Algorithmics IT school case, the budget grew from $8,000/month in Turkey to $25,000/month in Europe.
How much does a lead cost for a kids' IT school in 2026?
The cost per lead depends heavily on the market. In our case: Turkey $2.2, Slovakia €5.8, the Czech Republic €6.1, Ukrainian-speaking Europe €7.1, Switzerland €10.2. The average across all 11 markets is $7.2 per lead. In CIS markets, a reasonable price is $5-15.
Which funnel works best for online schools?
In education, a 3-step funnel works: a creative focused on the child's results → a lead form with a free trial lesson → a call from a course advisor offering the course. Lead-to-purchase conversion is 30-46% depending on the niche. In our IT school case, it was 35-40%.
How do you scale from one country to several?
Don't copy the campaign one-to-one. Every market needs language and cultural localization in the creatives, testing of local opinion leaders and a lead magnet adapted to local education standards. In the IT school case, we tested each new market separately for 4 months before scaling.
How long does it take to reach a stable cost per lead?
In our experience, 3-6 months. The first 2 months go to testing audiences and combinations, 1-2 months to scaling the winner and another 2 to optimizing for goals. In the IT school case, each new market stabilized by month 5.