Serious warnings 2 confirmed warnings, and users report 1 more thing. Checked: 405 reviews, the Android app’s code, the store listing and the privacy label.
From our automated analysis, 30 Oct 2026. Every label links to its evidence in the full report below.
This analysis describes version 8.6 of the app (the version number recorded inside the package we inspected), analyzed 2026-10-30 .
The store currently lists version 8.6, updated 2026-09-23.
Store screenshots supplied by the developer.
Survive endless hordes and unleash pure Bullet Heaven.
20 Minutes Till Dawn is a roguelike, shoot'em up where you mow down endless hordes of monstrous baddies.
【Unique Builds Every Run】
In this roguelite survival game, select from a variety of upgrades to create a unique and overpowered builds every run. You can be a fire wizard, for example, and ignite monsters with every pump of your shotgun, or an agile ninja who controls magic knives to pierce your enemies.
【Select Your Hero】
Choose from a diverse selection of characters and weapons that offer a variety of gameplay experiences.
【Features】
*Over 80 different upgrades to choose from for a unique experience every run!
*Wide cast of characters, with more to come post-launch.
*Casual quick 10-20 minute play sessions, for busy gamers.
*Rune System helps you become stronger…
From the developer’s store description.
For developers and the curious
The evidence, sources, measurements and versions behind every label above.
Findings that met the published evidence standard.
Bundles 10 distinct ad networks / mediation SDKs: Amazon Publisher Services, AppLovin MAX, Fyber / Digital Turbine Exchange, Google Mobile Ads (AdMob / Ad Manager) and 6 more
The package embeds several distinct advertising or mediation SDKs — a stack built for ad volume, not for a single house ad.
Package artefact
Appworthy
sha256:cb796cb6dddd7e9f2a436e50bb70ceeb5d45a518b521ebdfad0fd0be8e309ca2Progress is gated behind watching an ad: "Watch a video to earn a reward!"
Advertising volume or placement that interrupts normal use: ads between every action, unskippable or forced full-screen ads.
Package artefact
sha256:cb796cb6dddd7e9f2a436e50bb70ceeb5d45a518b521ebdfad0fd0be8e309ca2Specific evidence of a useful property, within the stated scope.
The developer's store data-safety disclosure states that data deletion can be requested; we have not seen a deletion control in the app's own text
The developer's data-safety disclosure on the store states that users can request deletion of their data, or the app's own text offers a way to delete the account. This restates what the developer declares; it is not a test of whether deletion works. Apple publishes no equivalent disclosure, so on an Apple-only listing only the app's own text can support it.
Open observations that have not cleared the evidence bar.
Pay to progress: raised by 6 distinct reviewers (about 1 in 68 reviews we read)
This observation has not cleared the required evidence standard.
Advancement or competition is gated behind spending money rather than playing.
405 reviews read for this analysis. Each observation keeps the sample and store it was based on.
you cannot play offline, many features in-game are only unlockeable by watching ads... and they are even more annoying, we are talking about 1:30min unskippable ad
OMG SO MANY ADS KMSKMSKMS
its good game but not complete. maybe its better if all is accessible even if lots of ads rather than paywall. anyways nice game to pass time and enjoy.
it's too pay to win the only way I was able to reach the second stage was by watching ads
This game is sooo hard but suuuper great i just dont like the p2w characters
A sampled review illustrates one person’s experience. It does not establish a finding on its own.
“Not assessed” means we could not measure it. It is never a zero and never a pass.
Android package 8.6, scanned 26 Oct 2026. These describe that package, not the current store version.
Supporting measurements: not a complete list of possible harms or positives. Measurement vocabulary 1.4 · how these are measured
This analysis describes version 8.6 of the app (the version number recorded inside the package we inspected), analyzed 2026-10-30 .
The store currently lists version 8.6, updated 2026-09-23.
Each store’s release notes, as the developer wrote them. They are the developer’s words, not ours.
The Nightmarish Feast Halloween update is live from Oct 25 to Nov 24! ◆ New Hero: Master Abby: Saint Rosa and her Rosethorn Chamber mechanic in battle. ◆ Crimson Wish: Use Pumpkin Keys to draw exclusive Halloween hero and weapon skins. ◆ Nightfall Pass: Complete Halloween Trial tasks to level up the pass and claim rewards. ◆ Festive Atmosphere: Experience Halloween-themed visuals on the main menu and monster sprites.
No release notes were provided for this version.
20-minutes-till-dawn-apple-1635123906Assessed 2026-10-30 from 405 user reviews, the app's package, the store listing and its privacy label.
What the stores publish. Store ratings are the stores’ numbers, not Appworthy’s, and are never averaged across stores.
From the store listing. Billing terms are read from each purchase’s name — the stores don’t publish them — so treat them as a strong hint, not a disclosure.
These are really on sale; we just can’t tell from the name whether they recur.
This assessment is produced automatically. Six independent signals are read — user reviews, SDKs in the app package, the app's own on-screen text, the app manifest, the store listing and the store privacy disclosure — and each candidate finding is filed under the sources that support it.
No charge stands on one kind of evidence — with one published exception. A finding is either a measured fact that anyone can reproduce (an ad SDK is in the package; every recurring plan the store lists is named as weekly, the term read from the purchase names, with a list cut off at the store’s 10-item maximum not assessed; the privacy label omits a tracker we can see), or it is an interpretation — and an interpretation needs independent sources to agree before it reaches the catch list. The exception is a harm only users can see: a charge may publish from reviews alone, as “Reported by N users”, once enough reviewers raise it to clear that charge’s own published bar. That agreement rule is what replaced the human reviewer, not a model’s confidence.
There is no overall score, on purpose. A 0–99 number was published here until 2026-09-15. It was withdrawn because it discriminated too weakly to be the instrument — 44.7% of every pair of apps tied — and because a single number is a verdict where a count is an observation. What publishes instead is a row of measurements, each of which says what was counted and, where we could not look, says that instead.
Findings under One source only are real reports we do not stand behind as findings. They are shown because withholding them entirely would hide a genuine complaint, and labelled because publishing them as charges would be an accusation off a single source.
Write to [email protected], or use the form. Name the finding, the app version and the evidence that needs correcting. Factual corrections are free and never depend on buying anything.