GDC 2022

Identify the Best Growth Strategy for Your Game with Azure PlayFab Live Ops Services

NBA Clash

When Nifty Games launched NBA Clash on mobile, they needed to find the best engagement strategy that would set them up to successfully release a new Starter Pack. They turned to Azure PlayFab Live Ops Services to help them easily configure and manage experiments to test the content against targeted player segments.

 

This demo shows how easy it is to configure experiments, monitor in-game events in real time with PlayStream, and use data to make confident decisions about how to grow your game.

Benefits

Custom KPIs and Events

Run A/B tests and experiment with variable gameplay experiences through coded or override-based configurations.

A screenshot of the Chicago Bulls playing in NBA Clash.

Concurrent Player Experiments

Deploy player-targeted experiments concurrently without the risk of interactions to optimize the rollout of new game flows.

A screenshot of the Card Deck in NBA Clash.

Insightful Engagement Analysis

Make confident, data-driven decisions based on insights into player behavior. Actionable metrics include Sample Ratio Mismatch and Statistical Significance Indicator.

A screenshot of NBA Clash revealing a new player card coming out of a duffel bag.
PlayFab has made a big difference for us – it has driven measurable efficiencies around our experiment management and analysis. PlayFab continues to improve, and we look forward to continued collaboration.

 

— Jonathan Bach, Head of Central Product at Nifty Games

Resources

Developer Info

Every shot counts in NBA Clash! Download this real-time 2 player basketball game for amazing head to head NBA action on your Android or iOS device.

Learn More

Solution website

Azure PlayFab LiveOps is the complete back-end platform for building and operating your game as a service. Use PlayFab LiveOps to improve your game.

Learn More

Documentation

PlayFab Experiments feature helps you identify the best strategies for your game. It does so by helping you run multiple concurrent experiments and ensure statistical trustworthiness.

Learn More