一个即插即用的滥用保护工具包在Stripe Radar之外是否也会有用？
支付是在线业务中的一个问题，尽管有不少投诉，Stripe迅速成为主要的支付系统。Marc Louvion发布ByeDispute后，我对Stripe未能覆盖这一问题感到好奇，因此深入研究了信用卡欺诈和Stripe的运作方式。
确实，Stripe Radar存在并覆盖了一些欺诈案例，但并不能涵盖所有情况，尽管有此功能，仍然有关于账户被标记的投诉，或者欺诈检测算法的修改导致所有进来的交易都被阻止，且没有任何停止的可能。此外，还有虚假注册、试用/退款循环、数据抓取或促销代码滥用等问题。
企业级工具显得过于复杂，而自助解决方案又消耗了开发时间。因此，我在想是否可以开发一个更通用的产品，能够检查每个用户的单次试用、检测一次性邮箱和数据抓取、进行行为机器人检查、预防促销/推荐滥用以及识别退款/退单模式……这可能会更有趣。在标记时，用户可以看到原因，并且解决方案可以随时停用。也许还可以建立一个社区，提供关于欺诈支付或争议的共同黑名单。此外，还可以提供一个仪表板来跟踪所有这些信息。
这样的产品会有帮助，还是只是增加噪音？我很好奇其他人是否也曾为此开发自己的系统。
查看原文
Payment is one of the problems in online business and Stripe quick emerged as the main payment system despite seen a fair amount of complains.
After Marc Louvion released ByeDispute, I was intrigued that Stripe was not covering that and so ended up having a tunnel on card fraud and how Stripe works.
Yes Stripe Radar exists and cover some fraud cases but does not cover everything and there have been complains of account flagging despite it or a modification of the fraud detection algorithms that blocks all in coming transactions without any possibilities to stop that. But also fake signups, trial/refund cycling, scraping, or promo code abuse.
Enterprise tools are overkill, and DIY solutions eat up dev time. So I wonder if a more general product that check One-trial-per-user, detect disposable email and scraping, have behavioral bot checks, prevent promo/referral abuse and chargeback/refund patterns, ... Would actually be more interesting. When flagging you would get the reasons and the solution can be disactivated at any time. Maybe even a community side with common ban list on fraudulent payments or disputes. On top of that a dashboard to follow all of this.
Would something like this be helpful or just more noise? Curious if others have had to roll their own systems for this.