How We Scaled an Online English School Through Google, YouTube, Meta, and TikTok and Increased the Value of Every Lead

Core tasks of the growth system

• not to stop at a single channel even when it was already delivering results
• to continuously improve the economics of existing campaigns without losing volume
• to find new traffic sources before they became mainstream in the market
• to scale working combinations only when the results were stable
• to build a system where new channels did not replace old ones, but strengthened the overall performance model

Meta Ads: scaling through look-alikes and creatives

At the first stage, the main focus was on Meta Ads. In that period, look-alike audiences performed best, so a lot of attention was paid to source-audience quality, segmentation, and constant video creative testing.

Systematic creative testing helped us find the right video format — one that worked especially well in combination with look-alike audiences and delivered stronger results.

Google Ads: searching for new growth points

Over time, the focus expanded to Google Ads. We tested search campaigns for offline directions, children’s courses, and online courses.

Different search approaches were tested separately:

• optimization around English-level tests
• competitor search campaigns
• informational queries related to learning English

Search remained a stable source of hot demand, but the real breakthrough came from YouTube.

YouTube Ads: the strongest scaling driver

Within the Google ecosystem, YouTube Ads started showing the most promising results. Thanks to a large number of tests, we found working combinations across audiences, interests, placements, conversion types, creatives, and landing pages.

As a result, YouTube advertising started outperforming Facebook and significantly increased both the volume of leads and the overall performance of the system.

TikTok Ads: early testing of a new channel

When Meta, Search, and YouTube were already working steadily, the next stage was testing TikTok Ads. At that moment, the platform was still new and not well explored in the market, so most approaches had to be tested from scratch.

Through systematic testing, we found the right balance of bid, conversion cost, and user journey on landing pages. This added one more growth channel and strengthened the overall lead generation system.

Performance Max: additional scaling inside Google

The final major stage was Performance Max. These campaigns improved overall ROAS by optimizing across several Google formats at once, including the Display Network, Demand Gen, video inventory, and other tools in the ecosystem.

Operating strategy

The core strategy had two layers.

The first task was to continuously improve and scale everything that was already working — as long as the economics remained stable.

The second, equally important task was to constantly search for new traffic sources, platforms, formats, and working combinations that could unlock the next level of growth.

This made it possible to avoid dependence on a single channel and constantly discover new scaling opportunities.

Result

The final assessment included results from the core traffic sources: Meta Ads, Google Search, YouTube Ads, TikTok Ads, and Performance Max.

• 30,000+ leads in total over the full period of work
• average lead cost across different sources — from $3 to $10
• ROAS varied depending on the channel, creatives, landing pages, and the stage of scaling

The main result of this case was not a single “magic” channel, but the creation of a system in which every new working tool reinforced the whole performance model.