Download the PDF version of the case study

We partnered with Belka Games to retarget lapsed users of their flagship match-3 puzzle game, Clockmaker: Jewel Match 3 Game, on iOS and Android globally. With over 10 million downloads worldwide, Clockmaker features core match-3 gameplay enhanced by bingo, puzzles, and tournaments that drive long-term player engagement.
25%
ROAS D7 above the KPI in week 1
3.8x
Campaign scaling in 4 months
>100%
Campaign recoup from week 1 (ROAS D30 >100%)
>200%
Predicted lifetime ROAS
The primary goal was to reactivate churned users who had previously made in-app purchases, with the aim of increasing their LTV and generating incremental revenue.
In addition to the ROAS-based KPI, Belka Games assessed campaign success using their advanced LTV predictions, evaluating how cohorts matured over time to determine performance from the early stages.

We began our collaboration on iOS, where our BI quickly spotted strong cohort growth. Leveraging Persona.ly’s advanced segmentation engine and ML algorithm, they precisely identified lapsed high-LTV audiences and delivered strong results – giving us the confidence to scale.
Aleksei Krizhanovskii – UA Producer at Belka Games
From the initial setup of the campaign with Belka Games, we leveraged our proprietary Live Audiences segmentation engine.
Available to clients at no additional cost, the Live Audiences engine classifies audience segments across multiple signals, including activity level, purchase history, inactivity window, and intent.
Integrating directly and securely with clients’ first-party data, it eliminates the need for third-party tools and adds and suppresses audience segments in real time.

Persona.ly’s Live Audiences segmentation engine in action
Another key advantage of the Live Audiences engine, directly integrated with Persona.ly’s DSP, is its ability to generate data-driven insights that identify and predict LTV from the early stages of the campaign. Our proprietary bidder processes over 4 million ad auctions per second, evaluating each bid in real time against key audience attributes and setting bids based on predicted LTV.
We quickly collected enough data for the bidder to accurately classify which audience segments had a higher probability of converting by returning to play Clockmaker and making new in-app purchases, and which segments had a lower likelihood of conversion. From the first week, we outperformed the D7 ROAS KPI and successfully re-engaged highly converting audiences.
The ability to classify audiences by predicted value at the bid stage and place optimal bids enabled us to safely scale the campaign 3.8x over 4 months. As cohorts matured and campaign profitability became apparent, we deepened our collaboration with Belka Games.
Leveraging our deep gaming expertise, we surpassed Belka’s D7 ROAS target by 25% in the very first week. By prioritizing high-value segments with the strongest likelihood of further in-app purchases, we achieved robust ROAS growth as cohorts matured.
Precise targeting paired with Clockmaker’s exceptional gameplay quality drove strong campaign progression, full recoup, and significant LTV uplift from re-engaged players.

As our programmatic ML-driven bidder accumulated more data, it was able to more confidently predict which audience segments had a higher probability of converting – re-installing or re-opening Clockmaker and making new in-app purchases. This enabled us to continue scaling the campaign effectively.
Our ML-driven DSP determines the optimal bid price: high enough to win the auction while staying safely within the predicted LTV. By combining bid shading and bid optimization strategies, we minimized the risk of overbidding, maximized impression quality, and ensured each audience segment received a bid precisely tailored to its predicted LTV.

To ensure campaign incrementality and avoid cannibalizing organic traffic, we run exclusively full-screen video creatives across both Android and iOS campaigns. Getting the bid price right for video is especially critical, as it is the most lucrative format.
This approach allowed us to scale the campaign by almost 4 times while consistently outperforming the D7 ROAS KPI and maintaining campaign recoup.
Belka Games is a data-driven casual mobile gaming studio founded in Cyprus in 2010, specializing in casual and adventure titles. With over 110 million monthly players worldwide, it ranks among the highest-grossing gaming studios in Europe.

Working with Belka Games has been an absolute pleasure. From day one, we built a strong, transparent partnership based on open data sharing and insights exchange. This level of collaboration was instrumental in achieving the outstanding results we delivered together.
At Persona.ly, we aim to act as a strategic partner, providing data-driven insights and strategies that extend beyond our platform. We’re proud to leverage our advanced machine learning capabilities to support global top-grossing publishers like Belka Games in hitting – and often exceeding – their ambitious growth goals.
Dana Purinson – Head of BD, EMEA & North America at Persona.ly
Persona.ly is a data-driven product company specializing in mobile user acquisition and retargeting, powered by proprietary machine learning algorithms and a robust first-party data management platform.
We help leading mobile companies including King, Tencent, Papaya, NextNinja, Tilting Point, Nexon and many others reach and exceed their growth goals by accurately predicting which users are most likely to engage and convert. Our machine learning models are optimized to target high-value users based on predicted LTV, driving strong ROAS and long-term user quality.
Our in-house programmatic DSP is directly integrated with top-tier ad exchanges and OEMs, processing over 4 million ad requests per second globally.
By combining real-time predictive analytics with programmatic scale, we empower our partners to unlock smarter growth, higher LTV, and measurable incremental impact.