“First, it’s worth noting that of the 20,000+ Steam releases approximately three-quarters are student projects, asset flips, AI-generated content, and hobby games. These games don’t get views or ratings, and constitute a “ghost” market that can be overlooked in analytics.
90% or die’ is likely an exaggeration, born from the fact that most games with huge sales can have high ratings. Of course, the higher the player rating percentage, the better. Below 80% is probably fatal, but 80-90% is viable if you have traffic; above 90% is where commercial outperformance concentrates.
More important, however, is the store page conversion rate, and a rating above 90% isn’t always required to become a profitable game. For small studios, which are the majority in the current market, the most realistic goal would be to aim for >80% rating and >50 reviews. With these results, player conversion averages 2%, which is a decent result. The biggest influence on success is likely a strong, high-quality demo for various events like Steam Next and building a community before its release/Early Access.
Yes, there is a correlation. For example, an increased rating and an increased number of reviews indicate an increase in conversion. The correlation is strong, but the reasoning is nuanced: review scores don’t trigger visibility; they convert traffic coming from other signals (wishlists, demos, press, word of mouth). A high-quality game that no one discovers still fails.
The counterintuitive data point: underperformed games had an average of 400+ days of pre-release marketing; overperformers averaged 200+ days. Over-polishing with no audience is a trap.
Small studios may decide to release a game if they have been able to achieve a review score that’s consistently above 80% in playtester/demo feedback (proxy for launch reviews); a demo that’s been run and achieved meaningful concurrent player counts (~100+ during Next Fest); wishlist base is large enough to seed launch day traffic (10,000+ for small studio survival); day-1 crash bugs and progression blockers are zero; the core loop is working, fun, and complete.
Everything else can be patched and polished after release. Further polishing before release may be excessive. It’s crucial to make a good impression on the release, as a game’s rating in the first week is crucial. There are studies that show that low ratings and bad reviews in the first days of release can ruin a game’s overall performance, even if developers subsequently fix all the bugs and address all the feedback, bringing the product to perfection.”
– Alex Sevastyanov, Senior Game Evaluation Analyst • Global Strategic Initiatives & Partnerships (GSIP)
