Sift > 实例探究 > How Skillshare keeps its platform free of spam and fraud

How Skillshare keeps its platform free of spam and fraud

Sift Logo
公司规模
1,000+
地区
  • America
国家
  • United States
产品
  • Sift’s Content Integrity product
技术栈
  • Sift
实施规模
  • Enterprise-wide Deployment
影响指标
  • Cost Savings
  • Productivity Improvements
技术
  • 网络安全和隐私 - 应用安全
适用行业
  • 教育
适用功能
  • 商业运营
用例
  • 欺诈识别
服务
  • 数据科学服务
关于客户
Skillshare is an online learning community for creators, where teachers and students gather to learn and teach anything they’re passionate about. Teachers can make money based on the minutes students have spent watching their classes, and students have the opportunity to pick up new skills – just look up a subject and start watching a class. In addition to the desktop site, on-the-go students can watch classes remotely using Skillshare’s iOS app. With over eight million students and thousands of classes in art, design, animation, photography, creative writing and more, Skillshare, based in New York City, is a global community with students and teachers from around the world.
挑战
Skillshare, an online learning community, was facing issues with fraud, fake accounts, and spam. Teachers were creating fake student accounts and watching their own classes to increase their earnings. The company also discovered collusion between teachers and students, with fraudsters using stolen credit cards to create many fake student accounts, and then redeeming the same teacher referral code across those accounts to get the fraudulent teacher referral bonuses. Fraudsters were also using Skillshare to engage in SEO spam by creating landing pages on the platform for products they were selling. This was risky as the landing pages could take users off-platform to questionable sites, and it was also detrimental to Skillshare’s reputation. Skillshare’s fraud management was primarily via SQL queries, and these schemes were only discovered after they had already happened. They needed a way to proactively detect and remove networks of colluding users, and to keep SEO spammers off of the platform.
解决方案
Skillshare turned to Sift’s Content Integrity product to get ahead of the fraudsters and their schemes. Trust and Safety Manager Susannah Page-Katz implemented Sift and, within two weeks, saw significant, accurate results from the machine learning model. Susannah used Workflows to automate blocking and deleting risky accounts based on Sift Score (risk score based on behavioral attributes), eliminating the need to review accounts manually, and stopping problem users at sign up before they even made it onto the platform. She was also able to automate the removal of spammy SEO advertisement pages via Workflows, without having to spend valuable time tracking down these pages and manually removing them. One of the most powerful Sift features for Skillshare has been the Network view, which was a game-changer for the company. It was a difficult, time-consuming process to try and unearth connected users via SQL queries; once Skillshare was able to visualize the entire web of students and teachers that were colluding to commit fraud, they not only better understood the scope of the problem, but were able to quickly remove those users from the platform and stop them from returning.
运营影响
  • Skillshare has prevented thousands of dollars in fraud losses annually.
  • They’ve also shaved 8-10 hours a week off of manual review time, which is critical for Susannah as a team of one.
  • She’s been able to rely on Sift – even during seasonal spikes in activity – without having to hire an additional team member.
  • As Skillshare continues to grow, Sift scales in tandem, leaving Susannah free to focus on things other than manual review and playing catch up to fraud.
数量效益
  • 8-10 Hours a week in manual review eliminated
  • Thousands of dollars in fraud losses prevented annually

Case Study missing?

Start adding your own!

Register with your work email and create a new case study profile for your business.

Add New Record

相关案例.

联系我们

欢迎与我们交流!
* Required
* Required
* Required
* Invalid email address
提交此表单,即表示您同意 IoT ONE 可以与您联系并分享洞察和营销信息。
不,谢谢,我不想收到来自 IoT ONE 的任何营销电子邮件。
提交

感谢您的信息!
我们会很快与你取得联系。