He spent $220 on Google ads — and 60% of the installs were robots

A developer who runs a small puzzle app has published a detailed account of a Google Ads experiment in which most of the installs he paid for turned out to be bots. Nick Abe, who runs the Android puzzle game Dayzle, described the case in a post dated September 11, 2026, first published on LinkedIn.

A campaign that stopped adding up

Abe set up a Google Ads campaign for Android at CA$40 a day, with the goal set to installs. For the first few days it barely spent anything: with a target cost per install of $1.50, Google could not find installs at that price. Removing the target as a test changed that immediately — the campaign spent double the daily budget, CA$80, and reported 21 installs in a day.

His own admin panel, meanwhile, showed one install. The gap came down to app versions: old versions of Dayzle do not report an install date. In the raw analytics, 21 new Android devices appeared that day, 20 of them running an old version the Play Store had stopped serving days earlier. An old version cannot be downloaded from Play, so those phones had obtained the app somewhere else — even though every one of them reported Google Play as the installer. Each opened the app once, spent zero seconds on any screen and never came back. The devices covered 28 phone models across 19 states.

56 installs billed, 13 real people

Over the full two weeks the breakdown was: 56 installs billed, 33 of them matching that pattern, seven more from countries the campaign was not targeting, and 13 actual people. Those 13 users finished 92 games between them, which Abe calls a good signal that real players enjoyed the app.

The 33 installations did not behave like people, and the analytics export, he writes, supports a bot farm. The scheme he describes works like this: the farm would watch the shortest video in his ad group without clicking it, then install the app from a saved copy of the file rather than from the store — faster, and less likely to be noticed by Play. Because Google counts a view followed by an install as a conversion, the more the farm "installed" Dayzle, the better the campaign looked to the algorithm, which then sent more ads to the farm.

What changed

The campaign goal is now "won a puzzle" instead of "opened the app". A script that opens an app and clicks around is cheap to build; one that solves a Sudoku is not — the aim is to make Dayzle more expensive to farm than the next app. Abe is waiting for Google's response to an invalid-traffic form and says he will report back on a refund.

His advice to other developers: Google's install count is a real number, but it is worth digging into. "If a bot farm can find my tiny ad budget, it can definitely find yours," he writes.