The Haystack clone saga in SteamDB numbers: the clone with the first public demo won, not the original..
retained text
If you missed it: A dev called Nas tweeted a prototype of a game about finding one needle in 5 million pieces of hay. It went viral, and within a week there was a Steam page for *Needle In A Haystack Simulator*... and clone pages from other devs. Around the same time, the dev of *Drunken Goddess Reflux* cancelled a public demo because they were afraid AI vibe coders would clone it right away. Chris Zukowski (How To Market A Game) pulled the SteamDB follower charts from August 22 to October 28 and went through what actually happened. I do data for a living, so of course this is the part I cared about the most haha : * The original and two clones put up coming-soon pages on the same day * The original ran a playtest (about 23k sign-ups, first spike of 533 followers) and signed with Polden Publishing on September 1, but released no public demo * The clone *Find The Needle* released the first public demo. It took about 13 days to spread on YouTube, and it passed the original in followers on October 5 * *Find The Needle* was then one of the top demos of the October Next Fest * Another clone shipped a full game about a month after the tweet with only 452 followers (roughly 4k wishlists) and stalled * The original still gets around 60 to 70 followers a day without a demo, but it's estimated to be about 20k wishlists behind the lead clone * Only three Haystack games got past 100 followers So the cloner who rushed out a full game with no momentum got nothing. As Chris puts it: "Even AI-cloners can't shadow drop a game and have it work." The clone that won did it the boring way: a competent demo that YouTubers liked, then Next Fest. His conclusion is that indies are not cooked. Haystack was a perfect storm (a viral idea in a simple, well-known "clean up" genre), and the fear we feel comes from the availability heuristic: cloning stories are vivid and everyone shares them, which doesn't mean it's likely to happen to your game ! I mostly agree. The lesson I take from the charts isn't "clones are unbeatable", it's "a closed playtest isn't a demo". YouTubers play what they can download, and whoever is playable first gets the videos. Hiding your game out of fear seems to be exactly what lets a clone get ahead so It's interesting to have the demo ready in case like this ! What I find less comforting is that the speed is new. A clone page in the same week as a viral tweet used to be rare. If your concept is a one-line pitch that fits in a tweet, you probably have days, not months, to get something playable out. Also, a game about finding a needle in a haystack now has its own haystack of clones to find it in. Someone should make a meta game about that, though I guess I'd get cloned by Friday. Full article with the charts and his advice (don't shadow drop, go straight to demo if you go viral, check SteamDB before panicking, use the press if you get cloned): [https://howtomarketagame.com/2026/10/28/ai-cloners-are-here-are-indie-devs-cooked/](https://howtomarketagame.com/2026/10/28/ai-cloners-are-here-are-indie-devs-cooked/) Curious what you all think: would the *Drunken Goddess Reflux* dev have been better off shipping that demo anyway? And does cloning risk change how early you show your prototypes now?
document fetched 2026-10-30 00:26 · canonical https://reddit.com/r/gamedev/comments/1xisr1y/the_haystack_clone_saga_in_steamdb_numbers_the