Thick as Thieves | Otherside Entertainment
Emergency 24/7 launch support in just 14 days: How Multiplayer Wizards helped the Otherside Entertainment team de-risk the launch of Thick as Thieves
“Your company was a lifesaver for us.”
— Jay Brown, Director of Engineering, Otherside Entertainment
Executive Summary
Two months before the launch of Thick as Thieves, Otherside Entertainment asked Multiplayer Wizards, Code Wizards Group’s multiplayer-focused division, for a launch-readiness review. The request reflected Otherside’s own prep: the team had a solid base covering a range of launch tasks. They had already invested in runbooks, the OTEL observability stack, and early planning around patching and server locations. Multiplayer Wizards helped sharpen that work for a live multiplayer launch, adding games-specific experience without discounting the enterprise SRE strength already in place.
Over the following month, Multiplayer Wizards worked closely with Otherside to make the launch model practical for launch, helping to refine runbooks, improve monitoring signals and alerts, strengthen launch and patching processes, validate server-location assumptions with latency evidence, and tighten up the process and tooling for the outsourced Network Operations Centre (NOC) expected to provide round-the-clock coverage.
Two weeks before launch, Otherside moved to Multiplayer Wizards’ full-support package after recognising that the planned support model still needed deeper games-specific context and tool familiarity; emergency onboarding was completed in just 14 days.
Thick as Thieves launched with 24/7 monitoring, war-room support, and zero player-facing issues related to infrastructure or monitoring systems through the first week; a testament to a fantastic collaboration with the team at Otherside and the result of many hours of hard work by both teams.
| 14 DAYS from urgent full-support decision to launch-ready support | 24/7 games-specific monitoring and launch coverage | 0 player-facing infrastructure or monitoring issues in launch week | 1st patches supported through the early live-operations window |
Game & Studio Context
Developed by Otherside Entertainment, Thick as Thieves is a stealth-action heist game playable solo or in two-player online co-op. Its connected multiplayer format made server availability, matchmaking, monitoring and operational response part of the launch experience. The game released on Steam on 20th May 2026.
For Otherside, the immediate issue went beyond whether the underlying services worked, into whether the whole operating model could detect, understand, and respond to problems quickly when players arrived without pulling the engineering team away from finishing and launching the game.
The Challenge
With two months remaining, Otherside contracted Multiplayer Wizards for a focused launch-readiness review. The early sessions confirmed meaningful preparation inside Otherside, particularly around runbooks, observability and SRE practice, while identifying where specialist game-launch experience would strengthen the final approach.
The work centered on extending load testing, validating server performance, tightening launch plans and checklists, completing patching plans, refining alerts, augmenting runbooks, and using latency-tolerance testing to inform server placement. Some of these elements were already in motion; the task was to finish, validate, and adapt them ready for launch day.
The deadline made prioritisation essential. Multiplayer Wizards helped convert the remaining readiness work into a practical launch plan so that the NOC staff covering the game could interpret live signals and respond quickly to issues.
Finding the Launch Risks Early
Multiplayer Wizards’ initial work turned Otherside’s readiness concern into a defined programme. The team reviewed the launch model, recognised the preparation already completed by the game’s engineers, and prioritised the remaining technical and operational work needed before launch.
Over the next month, the team improved launch and patching plans, refined monitoring signals, strengthened checklists, augmented runbooks, and prepared the existing outsourced NOC through training. Otherside’s runbook work and OTEL-based observability provided a strong base; Multiplayer Wizards added games-industry launch experience across Pragma, Edgegap, AWS EC2, Grafana and Prometheus.
Two Weeks to Launch: Replacing a Fragile Support Model
As launch approached, the planned NOC partner Otherside had contracted prior to working with the Multiplayer Wizards team still needed the games-specific judgement and tooling context the launch required. This external team could follow documented steps, but unfamiliar tooling limited their ability to interpret unfamiliar situations or act beyond prescribed runbooks.
This was a support-model mismatch, not a question of effort or engineering capability at Otherside. With two weeks remaining, Otherside moved urgently onto Multiplayer Wizards’ full-support package.
Emergency Onboarding in 14 Days
Multiplayer Wizards completed emergency onboarding in 14 days, establishing 24/7 games-specific monitoring, escalation routes, launch support and working knowledge across Otherside’s tooling and infrastructure.
The compressed schedule worked because neither team was starting cold; Otherside had already built useful runbooks and observability foundations, and Multiplayer Wizards had already shaped the launch risks, alerts and processes during the review stage.
Otherside launched Thick as Thieves with Multiplayer Wizards in the war room and providing round-the-clock operational coverage. Launch week completed with no player-facing issues related to infrastructure or monitoring systems, and the team supported the first patches as the game moved into live operations.
Latency Testing Changed the Hosting Strategy
Latency-tolerance testing examined how the game behaved under increasing latency and packet loss. The resulting report showed that Thick as Thieves remained playable at substantially higher latency than the team had expected.
That evidence changed planning assumptions; Otherside found they could use fewer server-hosting locations, widen matchmaking tolerances, and bring players together across broader geographic regions. This delivered more flexible hosting and matchmaking strategy grounded in observed game behaviour, saving costs on unnecessary hosting provisioning.
Results & Benefits
The engagement moved from a small review to full operational support without losing momentum. Otherside’s early preparation gave the work a solid base to begin with; Multiplayer Wizards helped complete the remaining readiness work and onboard full support inside the final 14-day window.
The technical launch was smoothly handled. Otherside combined its SRE experience and preparation with Multiplayer Wizards’ games-specific monitoring, war-room support, escalation paths and early patch coverage. Most importantly, no player-facing infrastructure or monitoring issues were recorded during launch week.
| Area | Before | After |
|---|---|---|
| Launch readiness | Preparation under way; remaining launch-readiness work required prioritisation and completion | Launch processes, checklists and runbooks prepared |
| Support model | General monitoring with limited player context | 24/7 games-specific coverage |
| Observability | Useful SRE foundations and OTEL observability in place; game-launch metrics and alerting needed refinement | Grafana and Prometheus monitoring aligned to response |
| Hosting strategy | Patching and server-location strategy planned, with final rollout awaiting stronger demographic and latency evidence | Fewer locations and wider matchmaking tolerances supported by testing |
| Launch response | Limited flexibility beyond prescribed steps | War-room, escalation and response support with game context |
Why It Mattered
A multiplayer game NOC must do more than watch dashboards and relay alerts. It needs to know which signals to respond to, how the tooling behaves, what player impact looks like, and which actions are safe when reality doesn’t match what’s written in the runbook.
Each element a NOC handles was vital for a successful launch of Thick as Thieves. Otherside had prepared the foundations; Multiplayer Wizards added the game-launch context, tooling familiarity and judgement needed to act as part of the launch team rather than a detached alerting layer.
For the Otherside team, the result was a safer launch without building an internal round-the-clock function at the very last moment. For the wider industry, it shows why games-specific operations can reduce launch risk when timing is tight and the unexpected cannot be fully scripted.
What We Learned
The Thick as Thieves launch showed that readiness is both technical and operational. Tests, alerts and runbooks matter, but they only protect a launch when the support team on-hand also has the context and authority to interpret events and respond at speed.
| Start with focused discovery: Otherside’s awareness and preparation led to a small engagement that exposed remaining risks early enough to act before launch pressure peaked. |
| Design alerts for decisions: Metrics are useful when they identify meaningful conditions and connect directly to response paths. |
| Test game behaviour, not assumptions: Latency and packet-loss evidence can materially change hosting and matchmaking strategy. |
| Games-specific operations add judgement: Tool familiarity and context allow a NOC to act when rigid runbooks are not enough. |
Subscribe to our
newsletter
"*" indicates required fields