基于ML的主动预防广告商流失的方法
Erika Sun ML Engineer | Advertiser Growth Modeling Team; Ogheneovo Dibie Engineering Manager | Advertiser Growth Modeling Team
Erika Sun ML工程师|广告主增长模型团队; Ogheneovo Dibie工程经理|广告主增长模型团队
Photo by Jason Blackeye on Unsplash
Summary
摘要
In this blog post, we describe a Machine Learning (ML) powered proactive churn prevention solution that was prototyped with our small & medium business (SMB) advertisers. Results from our initial experiment suggest that we can detect future churn with a high degree of predictive power and consequently empower our sales partners in mitigating churn. ML-powered proactive churn prevention can achieve better results than traditional reactive manual effort.
在这篇博文中，我们描述了一个由机器学习（ML）驱动的主动预防客户流失的解决方案，该方案的原型是我们的中小型企业（SMB）广告商。我们最初的实验结果表明，我们能够以高度的预测能力检测未来的客户流失，从而使我们的销售伙伴能够减轻客户的流失。由ML驱动的主动性流失预防可以比传统的反应性人工努力取得更好的效果。
Introduction
简介
Like many ads-based businesses, at Pinterest, we are intently focused on minimizing advertiser churn on our platform. Traditionally, advertiser churn is addressed reactively. Specifically, a sales person reaches out to an advertiser only after they have churned. This approach is challenging because it is incredibly difficult to “resurrect” a customer once they leave the platform. To address the challenges with addressing churn reactively, we present a ML-powered proactive approach to advertiser churn reduction. Specifically, we developed a model that can predict the likelihood of advertiser churn in the near future and empowered our sales team with insights from this model to prevent at risk accounts from churning.
像许多基于广告的企业一样，在Pinterest，我们专注于将我们平台上的广告商流失率降到最低。传统上，广告客户的流失是被动地解决的。具体来说，销售人员只有在广告客户流失后才会与之联系。这种方法是具有挑战性的，因为一旦客户离开平台，要 "复活 "他们是非常困难的。为了解决反应性地解决客户流失的挑战，我们提出了一种由ML驱动的主动性方法来减少广告商的流失。具体来说，我们开发了一个模型，可以预测在不久的将来广告客户流失的可能性，并使我们的销售团队有能力从这个模型中获得洞察力，以防止风险账户的流失。
In this blog, we cover the:
在这篇博客中，我们涵盖了以下内容：
- Churn prediction model’s design and implementation
- 流失预测模型的设计和实施
- Experimentation in the managed North America SMB segment
- 在管理的北美中小型企业领域进行的实验