Sift is pioneering new methods for filtering user-generated content so you can effectively moderate toxic users while encouraging positive engagement.
The tools have an impressive ability to filter billions of messages daily using over a million signatures in multiple languages. Sift has also bypassed current limitations by adding reputation and context into the risk analysis. This gives you the big picture on who your users are, what they are saying, and why.
Users exposed to unmoderated toxicity are three times more likely to disengage from an online community. But only a small percentage of users are toxic, why treat all their messages the same? The problem isn’t something bad being said once, it's about a user repeatedly engaging in toxic behavior that disrupts a healthy community.
Identify Your Toxic Users
Sift has innovated a better way to manage unwanted toxic behavior and to reward positive behavior by evaluating a user's reputation and understanding the risk of a message in proper context. This creates a positive experience for your users by increasing user engagement, ensuring safety, and building trust in your community.
Reward Positive Behaviour
Sift provides the most advanced functionality in the industry, all wrapped up in a simple HTTPS call. You just send your UGC and Sift filters it through your policies for that page, channel, or user. It will return “true” if the content passes or “false” if it doesn’t. It will then generate any additional parameters you have set, such as whether to h#sh out certain words, display warning messages, and so on. Sift’s backend is loaded with webhooks that can make calls to your server or generate a human review.Chat Check API Example
Behind Sift are full-time native speakers for each of its growing set of supported languages. In addition, we have developed a means to quickly identify core toxic-language patterns in additional languages to rapidly build a contextual database of problematic communications. Once we do all that work, we then tune it to your environment. Sift uses multiple algorithms in both natural and unnatural language to perform quality control on our list to form an effective defense against toxic users trying to bypass your system.
"Sift has made it possible for us to introduce chat safely, cost-effectively and with an extremely high degree of confidence that inappropriate language and behaviour will be identified and stopped before it ever reaches the world."
"Sift has an elegant and straightforward approach which streamlines the moderation task significantly. The additional benefit is an added dimension of analysis of user behaviour."
"We recommend it."
"The success of our daily operations depends upon our relationship with Sift. They're like fine waiters, there when you need them with exactly what you need!"
"Their leadership, advice, and counsel have helped JBF become a trusted provider for children’s digital content for Professional Sporting teams in the NHL, NFL, and the NBA"
"Millions of writers, readers, educators, and parents depend on Storybird to be safe and secure. And we depend on Chris and his team to deliver that security. They’re baked into our business."
Find out what we’re up to, what we’re thinking, and what we find intriguing. Check back for new posts both from us and people we find fascinating (and loosely relevant). We share thoughts on the online safety industry with topics as: chat filters, online risk management, language moderation, profanity, and behaviour influence through rewards.
Community Sift has been selected as one of Canada’s top 20 startup companies...
I'm a bit of an experience junky, when thrust into difficult situations I don't shy away from them...
She made rash decisions and seemed to enjoy giving out pink slips...
Learn more about the Community Sift chat filter and chat moderation technology and how it can work for your community. Contact us, we'd love to chat.
#204-1405 St. Paul Street
V1Y 2E4 Canada