Buddy AI: worldwide user acquisition at target ROI

TL;DR

Two things closed the old geographies at once: depreciation of the local currency against the dollar and Apple’s IDFA change degrading attribution. Scaling harder was not available – LTV had to move first.
The hypotheses targeted the decision-maker, not the user: parents renew the subscription, so progress emails to parents and a value video on the paywall – which parents see before any onboarding, since the kids go through it alone – were the levers.
LTV rose 32% against two hypotheses of at least 10% and 5%: retention to recurring payment +29% and conversion to subscription +17%, averaged by geography.
Only then came expansion, prioritized by market size so creative localization pays back sooner – LatAm scaled ×1.5 at target ROI and the product launched in Turkey.

Situation

Buddy AI is a language learning app that utilizes artificial intelligence to provide personalized English language instruction to children through a game. The idea of the app is that kids can start to talk to the virtual tutor right away, from the very first session. The app utilizes a subscription monetization model, and one of the most effective ways to acquire users at that time was to promote the app as a game for kids. Summing up, users' acquisition, retention, and monetization are the critical elements of the app's revenue.

As Senior Marketing Manager at Buddy AI, I was responsible for improving those metrics to ensure sustainable business growth for successful fundraising and break-even point achievement.

Task

Upon my collaboration with Buddy AI, the app had active users from two primary geographies and one major and one minor user acquisition channel. After a closer look at the unit economy, I concluded that there is no way to scale while meeting the KPIs on ROI.

The main reasons for that were the unstable economic situation in those geographies (depreciation of the national currency against the US dollar) and the introduction of the implicit consent for IDFA tracking by Apple (which led to a decline in attribution quality).

To solve the business growth problem according to target ROI, I have focused on users' LTV improvement and the search for new geographies for user acquisition.

Action

First, I've analyzed what prevents the app growth via given acquisition channels. Based on given conditions, I've decided to do the following:

formed hypothesis on LTV improvement from both registered and paying users' perspectives;
scale the user acquisition in new geographies according to the target ROI in case of successful LTV growth.

To solve the problem above, I've built the unit economy model and figured out the following:

major users' flows and product metrics;
how marketing affects those metrics;
which product adjustments may let marketing grow the business further.

Users' retention rate, LTV, and ROI are the crucial metrics for subscription businesses.

While selecting the approach for solving the problem from area 2, I've decided to prioritize the larger geographies since it will shorten the payback period for the investments in creative localization. In particular, we focused on LatAm due to its size and potential interest in the product.

I've come up with the following hypothesis:

LTV will increase by at least 10% (due to retention growth) via regular mailing to parents showing their kids' progress in learning English because the parents are the actual decision-makers regarding subscription renewal. Thus, we'll be sure why they should pay for the next period.
LTV will increase by at least 5% (due to ROI growth) via the increase in conversion to paying users since we'll put the video with product values on the paywall because parents often miss the onboarding (the kids may go through it on their own). The first thing they see is the paywall.
Growth of media buying, which meets the target ROI by at least 20% via creative localization and adaptation according to other geographies' cultural contexts.

Besides the major hypothesis mentioned above, I've applied several typical optimizations like onboarding simplification, the process for creatives' handling, creative rotation, etc.

I led the full-stack team to accomplish these tasks, including a data analyst, front-end developer, back-end developer, UA manager, and creative producer. Also, several teams were outsourced, like designers, video editors, localizers, and country advisers.

Result

LTV increased by 32% due to:
- retention to recurring payment growth by 29% (on average by geography);
- conversion to subscription growth by 17% (on average by geography).
Expanded the number of geographies available for user acquisition that met the target ROI, in particular:
- scaled user acquisition in LatAm by 1.5×;
- successfully launched the product in Turkey.

Results in numbers

Metric	Target	Result	How measured
LTV	≥ 10% via retention (parent mailing); ≥ 5% via conversion (paywall video)	+32%	Blended, averaged by geography
Retention to recurring payment	–	+29%	Average by geography
Conversion to subscription	–	+17%	Average by geography
Media buying at target ROI	≥ 20%	LatAm ×1.5, Turkey launched	Geographies meeting target ROI after creative localization

Table 6. Targets are the hypotheses set before the work; results are as stated in this case.

Written by Maria Atamanova – Senior Growth & Performance Marketing Manager and applied mathematician; twelve years growing businesses – apps, e-commerce, EdTech, AI data platforms – on numbers, not gut feeling. The decision to prioritize large geographies was a payback argument: creative localization costs the same everywhere, so the larger market repays it sooner.

More cases

How to Leverage Marketing to Scale a Product and Secure Series A Funding – Praktika

$30M Series A · 5× revenue in <3 months
How to Scale User Acquisition via Improving Operations – Scentbird

27% CAC reduction · Meta spend ×2 in 5 months

How to Perform Worldwide User Acquisition for a New Startup