Ad spend up 2.76× and CPA down 17% for a European kids’ EdTech studio

About the client

A European kids’ EdTech studio with 5M+ young players in 20+ languages. Over the past five years it has built:

• Reach: over five million children have played its games, in more than 20 languages.
• Catalogue: five apps live on iOS and Android, with web products in active development.
• Category standing: consistently ranked in the top five of the App Store’s under-fives learning category.

‍

Challenge

The studio’s UA managers were optimising paid spend by hand. Moving budget across campaigns, products and localisations ate the week, and it was easy to get wrong: one missed check meant another day of spend on a campaign that was never going to work. Creative testing was where it hurt most, because a test needs watching hour by hour and nobody can watch an ad account round the clock. They wanted the buying run properly, by people who do it full time, so their own team could go back to making the games and the creative.

‍

Solution

The studio brought us in to run paid acquisition end to end: media buying on Meta, the creative testing programme, daily optimisation and the reporting that sits on top of it. Since October 2023 we have taken ad spend up 2.76×, and it scaled without CPA going up with it.

The difference is where the decisions come from. Platform auto-rules only see platform data. We buy against purchase data from the studio’s own database, so a campaign is judged on what it actually returned rather than on what the platform counted.

‍

What we run on the account

1. Media buying

• Turning campaigns on and off: we start and stop campaigns across products and localisations against the media plan.
• Budget moves: budget goes onto what is working, on the schedule the account needs.
• Bid changes: bids go up or down to keep the buying competitive without paying over the odds.
• Cost-spike cover: we watch for sudden CPC and CPM jumps and pull back before they turn into overspend.

2. Daily control

• Daily checks: we monitor ad performance daily, weekends included.
• No manual QA on your side: the studio’s managers stopped doing hourly checks altogether.
• We act first, then tell you: when delivery breaks or spend runs hot we fix it and report what changed.

3. Performance optimisation

• Stopping what is not working: ads come off on historical and live data, not on a hunch.
• Decisions on your data: we optimise against the studio’s own purchase records rather than platform-reported results.

4. Creative reporting

• Week over week: every metric compared against the week before.
• The metrics that matter: spend, CPM, CPC, purchases, CPP and CR.
• Scale and cut: we find the creative worth more budget and retire the creative that is not earning it.

5. Reporting the team actually reads

• A summary in your Slack: the studio sees what changed and why, in the channel it already works in.
• No chasing: nobody has to ask us how the week went.

‍

How it changed the team’s week

Handing the account over changed how the studio’s team spends its time:

• Reporting where they work: our daily summary lands in their Slack, so they see what we changed without opening an ad account.
• Creative direction from evidence: a weekly creative read says which concepts to scale and which to retire, so the next batch is briefed on results rather than opinion.
• Metrics they can trust: spend, CPM, CPC, purchases, CPP and CR tracked week over week, so drift gets caught in days instead of at the end of the month.

‍

The results

In the first year of running the account:

• Ad spend: up 2.76×, in line with the studio’s growth plan, with an immediate 17% drop in CPA.
• Cheaper acquisition: CPM and CPC both came down while CTR went up.
• Creative testing: spend on creative tests rose 84.35%, which let us run cheaper tests and triple the number of creatives in test. Spend on zero-conversion campaigns fell sharply.
• Focus: the studio’s team came off manual adjustments and went back to creative and product.