a/b测试
In games, at any one time, countless variables have an impact on each player’s experience. That variability has a knock-on effect and makes optimization a much more complex undertaking than it might seem.
在游戏中，任何时候，无数的变量都会影响每个玩家的体验。 这种可变性具有连锁效应，并使优化工作比看起来更复杂。
If you want to improve any aspect of your game, you need technology that isolates the aspect in question and tests your “improvements” effectively. If rigorous A/B testing bears out your hypothesis in specific conditions, on a subset of your total audience, you can roll it out on a larger scale.
如果您想改善游戏的任何方面，则需要能够隔离问题方面并有效测试“改进”的技术。 如果严格的A / B测试在特定条件下证明了您的假设，那么您可以在部分总受众群体中进行大规模推广。
An A/B test essentially compares different versions of the same variable to determine the impact of changes to that variable. The idea is to run concurrent tests in which conditions are as close to identical as possible, save for the variable being tested.
对照组继续经历“默认”行为，治疗组收到新的行为，因此您可以比较结果并确定更改所选变量的影响。
Testing changes in this way is important for several reasons:
由于以下几个原因，以这种方式测试更改很重要：
By testing changes on a small portion of your population, you minimize any negative impact of rolling out a change.
通过测试一小部分人口的变化，可以最大程度地降低实施变化的负面影响。
You can statistically prove the effects of your proposed change, instead of relying on gut feeling.
您可以从统计学上证明您提出的更改的效果，而不必依赖于直觉。
By deciding on what constitutes “significant” change before running the test, you can be objective in your assessment.
通过在运行测试之前确定什么构成“重大”更改，您可以客观地进行评估。
Testing is an iterative process. The diagram below breaks down A/B testing step-by-step.
耐心和纪律是成功进行A / B测试的关键。 由于您感觉到出现了明显的结果，因此提早结束测试不会产生最佳结果。 等待测试运行全过程。
保持一致性 (Maintain consistency)
You can only draw reliable conclusions from tests that action one change at a time. As soon as multiple variables are at work, you can’t accurately evaluate the precise impact of any of them. It’s one at a time, or not at all.
您只能从一次执行一个更改的测试中得出可靠的结论。 一旦有多个变量起作用，您就无法准确评估其中任何一个变量的精确影响。 一次只有一个，或者根本没有。
始终使用相同的受众 (Always use the same audience)
Your control and treatment groups must be made up of identical audiences, or the whole exercise is pointless. Geolocation, demographic, and play style are useful markers for segmentation because such things differentiate player profiles. If you want to know the true impact of a change, it must be tested on identical subjects.
您的控制和治疗小组必须由相同的受众组成，否则整个练习没有意义。 地理位置，人口统计和游戏风格是进行细分的有用标记，因为这样可以区分玩家个人资料。 如果您想知道更改的真正影响，则必须在相同的主题上进行测试。
We already have A/B testing functionality within the deltaDNA platform, and it is an integral part of optimizing everything – from game difficulty to in-app purchase (IAP) offer value. With cross-platform and rich data capability, this end-to-end solution enables publishers and developers to better understand different player behaviors and create personalized experiences, targeting individual players in real-time. Get a full-feature 30-day free trial of deltaDNA to discover what automated A/B testing can do for your game!
我们已经在deltaDNA平台中具有A / B测试功能，并且它是优化所有内容的组成部分-从游戏难度到应用内购买(IAP)提供价值。 凭借跨平台和丰富的数据功能，此端到端解决方案使发布者和开发人员可以更好地了解不同的玩家行为并创建个性化的体验，从而实时针对各个玩家。 获得 deltaDNA 的全功能30天 免费试用， 以发现自动A / B测试可以为您的游戏带来什么！
翻译自: https://blogs.unity3d.com/2020/07/13/how-a-b-testing-can-help-you-optimize-your-mobile-game/