Kohort, a London-based mobile gaming analytics firm, closed a €5.9 million Series A to build AI agents that automate user acquisition optimization for mobile game studios. The round signals investor confidence in applying autonomous systems to gaming's most expensive, least efficient channel: UA spend typically runs 30-40% of mobile game budgets but remains heavily manual despite decades of analytics software.

The startup positions itself between data (which already exists) and action (which studios still do by hand or through agencies taking 15-20% cuts). Kohort's agents would supposedly make real-time bidding, audience segmentation, and creative optimization decisions across networks like Apple Search Ads, Google App Campaigns, and Facebook without human intervention. The Series A arrives as European gaming infrastructure consolidates: Superawesome (kid-safe ads) sold to Roblox, Scopely acquired Spain-based Afterverse, and Unity cut 25% of headcount, leaving room for specialized automation plays.

UA in mobile gaming is structurally broken. Studios spend billions annually acquiring players for games with <5% retention at day 30. The inefficiency persists because (1) each game's economics differ wildly, making scaled playbooks impossible; (2) human UA managers cost $80-150K and turn over constantly; (3) agencies capture margin without bearing risk. An AI agent that learns per-game conversion funnels and adjusts spend in hours rather than days theoretically unlocks 15-25% ROAS improvements, which at scale is nine-figure value.

The European angle matters: European gaming studios (Germany's Goodgame, Scopely, Miniclip heritage) have tighter ARPU and lower ad spend per install than US counterparts, making automation more economically urgent. Kohort entering now hits a consolidation moment. If agents work even 60% as well as claimed, they threaten both legacy UA consultancies and in-house UA teams. If they fail, they validate that mobile gaming's acquisition problem is unsolvable by software alone.

Kohort wins if it captures 10+ percent of mid-tier European game studios (companies spending €500K-5M annually on UA). Legacy UA agencies and consulting firms lose if automation proves repeatable. Winners: studios with thin margins that currently can't afford dedicated UA staff. Losers: humans doing manual UA optimization at mid-size publishers. Hyper-scalers (King, Scopely, Take-Two) already have proprietary systems and won't adopt third-party agents; Kohort's real market is 50-500 person studios.

Monitor (1) Kohort's first customer cohorts and their ROAS delta vs. manual UA over 6 months; (2) whether the startup lands a second enterprise customer outside gaming (e.g., fintech app installs); (3) whether larger consolidators like Zynga (now Take-Two) or AppLovin acquire similar capabilities rather than licensing; (4) if Google or Apple integrate agent-like features into their own UA platforms, commodifying Kohort's moat.