Sign In Start Free
Research-grounded

Pre-build. Soft launch.
Every verdict cited.

PixyLiv covers two decision points. Gate 1 validates your concept and UA budget before you write a single line of code. Gate 2 returns a go/iterate/kill verdict during soft launch. Every threshold has a named source. If you disagree with a verdict, you can check the framework it came from.

Why we built this

Academic research on independent mobile game studios has found that most cannot systematically collect or interpret soft launch analytics. Studios were making go/kill decisions based on intuition rather than data, losing months of development time and UA budget as a result.

Separately, research on freemium digital services confirmed that even studios with access to analytics do not use them systematically. The data exists but the interpretation framework does not.

PixyLiv is that interpretation layer. You supply the numbers. We supply the framework: named, cited, and auditable.

The tools and their research basis

Each tool implements a specific published framework, not a proprietary algorithm.

Pre-Build Validator
Validates a studio's concept and UA budget before building. The concept stress test is based on five design conditions that correlate with D1 retention: unresolved session end state, daily return mechanic, short core loop under 5 minutes, multi-session progression, and a clear visual hook. The UA budget check uses the Cochran formula. 385 installs minimum for 95% confidence interval at 5% margin of error. Soft launch region guidance follows practitioner consensus on proxy markets.
Cochran (1977) statistical methodology · practitioner market research 2025
Decision Engine
Evaluates D1 retention, CPI, Hook CTR, and UA budget against genre benchmarks sourced from published practitioner frameworks covering D3/D1 retention thresholds and genre-specific benchmarks. The D7/D1 ratio diagnosis identifies whether the issue is FTUE, core loop, or acquisition cost. the ratio is more diagnostic than either retention number in isolation.
D3/D1 threshold framework (2022) · genre benchmark research (2025)
Friction Finder
Level drop-off thresholds derived from genre completion rate norms aggregated from practitioner research and supplemented by community outcome submissions in PixyBench. The four problem classifications. FTUE, difficulty spike, reward deficit, and pacing. follow published taxonomy of player engagement barriers.
Genre completion rate research · engagement barrier taxonomy (2008)
Update Evaluator
Runs a two-proportion z-test on D7 retention before and after an update, approximating the p-value using the normal CDF. This is the standard statistical significance test for A/B retention experiments in mobile games. Threshold: p < 0.05 for significance.
Standard statistical significance testing · cohort analysis methodology
Revenue Forecast
Implements compound daily growth from a weekly growth rate input, producing three scenarios across a 90-day horizon. Burn rate comparison follows the ROAS payback-window methodology. the question is not whether revenue is growing but whether it is growing fast enough to cover costs within the soft launch window.
ROAS payback-window methodology · compound growth modelling
ROAS Waypoints
Tracks the actual measured ROAS payback curve from real spend and revenue data entered by the studio. Never projects past real data. Compares the measured payback curve against the target window to determine whether scaling UA spend is economically viable right now.
Measured payback methodology · ROAS tracking framework

Research basis

All thresholds and frameworks are traceable to published sources. We do not name third-party commercial tools or products in our citations. only academic papers, publicly available frameworks, and standard statistical methodology.

Genre benchmark research (2025) — "Soft launch practitioner benchmarks by genre"
Primary practitioner reference for D1/D7/CPI thresholds by genre, the D7/D1 ratio diagnostic framework, and the two-consecutive-week go signal rule. Published and updated annually.
Practitioner (primary source)
Su X., Backlund P., Engström H. (2023) — "Data-driven method development for indie mobile game publishing decisions"
Multimedia Tools and Applications. Academic study confirming that independent studios cannot systematically interpret soft launch analytics. This is the core problem PixyLiv addresses.
Academic · peer-reviewed
D3/D1 threshold framework (2022) — "D3/D1 retention threshold as predictor of D7 trajectory"
Established the D3/D1 ratio as the earliest reliable predictor of D7 retention trajectory. Published via practitioner channels and widely adopted in the mobile soft launch community.
Practitioner · public framework
ROAS payback-window methodology (2024) — "Cohort analysis and ROAS payback methodology"
Defines the framework for evaluating whether projected revenue covers cost within the soft launch window. used in the Revenue Forecast and ROAS Waypoints tools. The measured waypoint approach tracks actual payback against target window.
Practitioner · cohort analysis methodology
Mäntymäki M., Islam A.K.M.N., Benbasat I. (2020) — "What drives subscriptions to freemium services?"
European Journal of Information Systems. Confirms that studios with analytics access still do not use data systematically for decision-making.
Academic · peer-reviewed
Cochran W.G. (1977) — "Sampling Techniques (3rd edition)"
Wiley. Standard reference for the 385-install minimum sample size. 95% confidence interval with 5% margin of error for a binary outcome. Used in the Gate 1 UA budget check.
Academic · statistical methodology
Schell J. (2008) — "The Art of Game Design: A Book of Lenses"
CRC Press. Taxonomy of player engagement barriers used to classify friction types in the Friction Finder.
Academic · textbook

Common questions

Are the benchmarks exact or approximate?
The genre thresholds are derived from multiple practitioner sources and reviewed quarterly. They represent the minimum performance required before UA scaling becomes viable in that genre — not aspirational targets. Studios consistently above these thresholds have a defensible go signal. Studios below them do not.
What is Gate 1 and when should I use it?
Gate 1 is the Pre-Build Validator. Use it before writing a single line of code. It tells you the exact D1, D7, CPI, and Hook CTR benchmarks your genre requires, whether your planned UA budget generates enough installs for a statistically valid soft launch decision, which countries to test in, whether your concept design has the five conditions that correlate with hitting D1 benchmarks, and the mistakes most likely to affect your specific genre and budget. If your game is already built and in soft launch, skip Gate 1 and go straight to the Decision Engine.
Does PixyLiv replace my analytics setup?
No. PixyLiv reads your numbers and interprets them. Your analytics setup collects the data. Every PixyLiv input is designed to be obtainable without any specific platform — you can estimate from playtests, track in a spreadsheet, or pull from whatever you already use.
How often are benchmarks updated?
Quarterly. We review published practitioner sources, compare against community submissions in PixyBench, and update genre thresholds where the evidence warrants it. The update date is shown on the benchmarks page.
Who built PixyLiv?
PixyLiv is built by an independent team of mobile industry practitioners. We built it because we needed it — a free, research-grounded tool for studios making hard decisions with limited data and no internal analyst. Reach us at info@pixyliv.com.
Try the tools
Free for indie studios. Not yet building? Start with Gate 1. Already in soft launch? Run the Decision Engine. Both give you cited reasoning you can share with your team.
Start Free — No Credit Card View Benchmarks