AI Event Drafting Cut Our LiveOps Prep Time From Days to Hours
Our first Grid Pulse event took four and a half days. By early 2026 that was under four hours. Here is exactly what changed and what we kept the same.
Read articleLiveOps, Retention, and Game Design
The Bold Games team shares practical lessons from operating live mobile games, covering event design, player retention, monetization, and AI's role in each.
Featured
All Articles
The Day-7 Drop Is Where Your Revenue Actually Leaks
Day-7 retention sits between day-1 and day-30 and often gets ignored. It should not. It is the window where daily habits form and where IAP conversion is decided.
What Makes a Limited-Time Offer Work in Mobile Midcore Games
LTO conversion in midcore titles comes down to two variables: when in the session the offer appears and how large the perceived value gap is. We tested 13 variants on Scout Horizon.
Running LiveOps Event Tests Without a QA Team
A minimal A/B loop for live events does not require a QA department. Five elements cover what actually goes wrong before it reaches your full player base.
Battle Pass Pacing: Getting Players to the Final Reward
Scout Horizon first battle pass: 58% of buyers never reached the final reward. The dropout was not about reward quality. It was about where the milestones were placed.
Three-Tier Player Segmentation for Small Studio Budgets
Full behavioral cohort analysis takes weeks and costs money most studios do not have. Three tiers built on session frequency and IAP history cover 80% of the LiveOps decisions that matter.
Session Length vs DAU: Which Number to Optimize First
Both metrics matter, but building toward the wrong one first pulls your game in the wrong direction. The answer depends on how your game makes money.
What Soft Launching in Turkey and MENA Actually Showed Us
We ran Grid Pulse through Turkey first in November 2023. Day-1 was 38%, day-7 was 18%. Here is what those numbers did and did not predict for subsequent MENA markets.
In Year Two, Balance Work Outlasts New Content Drops
Content packs feel productive. Balance patches feel like maintenance. After Grid Pulse crossed its second year, the data showed balance work was doing more for retention than new levels.
Building a LiveOps Event Calendar Without a Producer
A working calendar does not require a headcount. It requires three event types you know work, a repeatable weekly rhythm, and discipline to break schedule only when the data says so.
How to Think About Game Economy Design as a Small Studio
Grid Pulse went through three complete economy redesigns in its first six months. Each one taught us something the previous version could not. Here is what we built toward.
Six Months of AI-Generated Events Showed Unexpected Player Patterns
After running AI-drafted events on both titles for half a year, the patterns that surprised us were the ones that changed how we think about player motivation.