Game Launch on Steam

Sky had 270 million players on mobile and console and zero on PC. The Steam launch inherited a hero asset that showed no gameplay, a platform label that undermined confidence, and no backup plan. I built one — a distributed identity-based creator ecosystem across 45 partnerships and thousands of paid creative variations that reached the audiences a single trailer never could.

Sky launching on Steam wasn't just a port. It was an opportunity to reach an entirely new audience — and a risk, since Steam's community is vocal, skeptical of mobile ports, and quick to define a game's reputation in the first weeks.

The launch inherited two structural problems I didn't control. The hero asset delivered was a cinematic trailer for the Sky prequel film — beautiful, and containing no gameplay whatsoever. Over ninety seconds long, incompatible with vertical ad formats, and answering none of the essential questions a new player would have: what is this game, what do I do in it, is it available now? I flagged this repeatedly and was overruled. The second problem: after a year of "coming to PC" messaging, leadership chose to label the release "Early Access" because Steam's native friending features weren't fully integrated — even though the game was fully playable. The label introduced doubt where there should have been confidence.

I fought both decisions and lost both. Then I built a parallel campaign designed to succeed despite them.

The core strategic decision was mine: reject the one-message, one-hero-asset model entirely. The PC gaming audience wasn't one audience — it was dozens of specific communities, each with their own identity and their own reasons to care or not care about a mobile-native social game on Steam.

I rebuilt the audience architecture from scratch around identities rather than demographics: introverted players, neurodivergent players, people who wanted connection without social pressure, cozy gamers, couples who played together, hardcore players who remembered the studio from Journey. Each identity needed a different message, a different messenger, and a different format.

I activated 45 creators across gaming, LGBTQ+, neurodivergent, cozy, and co-op communities — no two partnerships structured the same way. Each got a custom brief built around what that creator's specific audience would find genuinely compelling. Not a script. A framework that gave them the emotional and narrative context to tell the story in their own voice.

The Ironmouse partnership required the most preparation. She is one of the most prominent VTubers in the world, confined to her home due to an autoimmune disorder, beloved for playing games on stream with her friend CDawgVA. The brand alignment was almost too obvious: a game about being guided through a beautiful world by someone you trust, played by someone who experiences the world through screens. But Sky is notoriously difficult to onboard. I had one shot at a two-hour livestream. Her management didn't want to give us prep time beyond what was contracted. I told them directly: without the onboarding session the content will be bad and I'll walk away from the deal. They agreed. I came in with a document of specific anecdotes, emotional framing, and narrative handles. She used almost everything in it on stream. She cried from the emotional impact of the game. The clips went viral. Other streamers started playing organically.

For Cozy K I took the opposite approach: a nine-month partnership built on credibility over reach, timed around the Sanrio collaboration to give her a genuine personal hook into the campaign. Her audience was the North American cozy gaming community — exactly the group I needed to convert. A long-term relationship signals something a one-off placement can't.

On paid media, I created thousands of variations across TikTok, Instagram, Facebook, YouTube, and Reddit — each matched to platform behavior and audience identity. Trickle-released, read early signals, shifted budget toward creative showing the strongest intent signals. Not clicks. Installs, retention, the quality of the acquisition.