Merge Mansion — Merge Mission

Merge Mission combined live action, CGI, gameplay composites, and social cutdowns into one production that had to feel ambitious and still work as UA creative. I led creative direction and owned the initiative end to end: the original brief, concept development with our external production partner, key creative decisions, variant planning, and CPP creative.

I created the initial brief to explore a premium new approach for Merge Mansion UA. When the first partner concept came back visually strong but too broad, I re-briefed the direction and pitched three counter-approaches. The final concept combined parts of these into one tighter, more testable scope.

The main creative risk was readability. Stacking live action, CGI, and gameplay builds cognitive load fast, and if the cuts do not connect clearly in the first few seconds, the piece loses its conversion read. The principle I locked early: lead with story, pay off with CGI. The gameplay read had to be unmistakable within the first three to five seconds.

Production ran on two parallel tracks. The partner delivered the hero films and partner-led re-edits. Alongside that, I built an in-house variant track at Metacore, giving us more testable directions without affecting the partner’s delivery schedule. Each variant answered a specific question — story order, gameplay clarity, social cutdown formats, or CPP alignment — rather than producing variants for volume.

The narrative-first hero delivered a clear performance lift against the alternate hero, confirming story-first as the primary creative lever. A follow-up test on the long film with in-frame game UI delivered another measurable improvement. CGI-forward openings, on the other hand, hurt the opening read and showed drop-off around the CGI introduction — a useful negative learning that has shaped how I brief premium concepts since. Beyond the test wins, the project produced a reusable partner-plus-in-house production model and a clearer creative principle for the next premium piece.