Celestial Return

Metaphor Games · 2026 · $14.99 · Adventure · deal grade A

Measured on 2026-09-24 01:14 UTC, 12 days ago. The whole catalog is re-measured daily; a game that just launched can move fast between passes.

• Estimated net residual revenue: $5.5k to $8.2k per month
• Opportunity: $8.9k per month at x1.30 dormancy
• Estimated lifetime owners: 5.6k
• 87% positive across 175 reviews · 54.4 reviews/month

Why it's flagged

• Developer active, last post 1 months ago
• Studio active elsewhere (2 titles tracked), so this one may welcome a partner
• Store page localized in only 1 language(s)

Analyst notes (AI-assisted)

Note written 2026-08-28; the live figures above take precedence over any numbers in it.

A surreal cyberpunk narrative adventure with dice-based resource management and branching endings, released in 2026 by Metaphor Games via Shoreline Games.

Celestial Return has quietly accumulated solid residual revenue on minimal marketing, driven by an 88% positive rating and strong word-of-mouth comparisons to cult favorites like Disco Elysium and Citizen Sleeper. The title's core tension, narrative ambition and visual identity resonate with players seeking character-driven, experimental indie fiction, though storytelling execution and mechanical friction divide its audience. For publishers or developers seeking catalog depth in the narrative-adventure segment, or studios building a prestige indie label, this represents an undervalued backlist asset with room for targeted revival if community feedback on story clarity and dice friction is addressed. Most realistic play: revival.

• Risk: Short playtime (median 9 hours) and narrative ambiguity create high replayability expectation but also high abandonment risk if players feel endings lack closure.
• Risk: No recent developer communication patterns visible; studio status marked operating but engagement cadence unclear.
• Risk: Dice economy frustrates a vocal subset of players, with at least one prominent review critiquing haggle mechanics as illusory choice rather than genuine negotiation.

This case as Markdown · Estimates carry ±30-50% error per title.