Cybersecurity researched Daylam Tayari dug deep into Twitch's API (via The Verge) to uncover the Brand Safety Score, a grade given to streamers to inform a brand of that streamer's content. Think of an ESRB rating or star-grading system a la Lyft and Uber but for streamers.
The Brand Safety Score is calculated by using a number of metrics. This includes the streamer’s age, a rating given by Twitch staff itself, the streamer's ban history, the relationship the streamer has with Twitch, the streamer's automod settings, their partnership status, the ESRB rating of the game the streamer is playing, and whether the stream is set to mature.
Twitch a couple days ago added to their GQL (internal) API endpoints (keyword to query the API) for a "brandSafetyScore" which is sent to advertisers for ads, sponsors and also probably for bounty purposes.— Daylam 'tayari' Tayari (@tayariCS) March 9, 2021
Check this tweet's images for screenshots of the API changes.
It doesn't seem like the Brand Safety Score is built for streamers, if Tayari's research into Twitch's internal API is accurate. The rating is an indicator for brands almost exclusively, giving them a sense of whether or not a particular streamer is brand-friendly enough to strike a deal with. It will also likely come into play when brands want to set up sponsorships with streamers.
Twitch has not publicly commented on the Brand Safety Score feature yet.
The Brand Safety Score would also shift the way brands interact with streamers. Right now, the main way for brands and streamers to connect on Twitch is through the platform's bounty board, a listing of paid opportunities for affiliates and partners around the world to choose from. There's speculation that if the Brand Safety Score comes to fruition it could be used to expand the bounty board feature, giving brands another way to judge potential partners.
Related Forum: Gaming Discussion