RIZZASTER

Rizzaster Staff Handbook Incident Review

/archive/votes/bots/misc/ · last edited 01 Apr 2021 · Support

Last Tuesday evening a player in the main lobby began spamming all-caps messages with repeated slurs targeting another user's avatar style. The initial response from the on-duty helper was to issue a standard three-day mute, which is the default for first-time verbal abuse. This felt appropriate at the time because the helper was managing two other simultaneous reports and wanted to keep the queue moving. However, within an hour, the muted player returned on an alternate account and continued the harassment in private messages, causing significant distress to the victim.

The core issue was that the moderation policy for repeated harassment was not applied consistently. The initial mute was treated as a standalone event rather than part of a pattern. We failed to check the player's recent message history before assigning the penalty, which meant we missed the context that this was the third complaint against this account in a week. The volunteer team is small, so we cannot manually review every history log, but we established a quick checklist for high-severity keywords that triggers a manual review before auto-moderation takes over.

The policy has been updated to reflect that any account flagged for targeted harassment now receives a temporary ban rather than a mute, regardless of whether it is the first offense if the severity is high. The definition of severity now includes repeated targeting of specific users or avatar styles, not just general toxicity. All staff members must now verify the severity level using the provided keyword filter before applying any penalty. This adds about thirty seconds to the response time but ensures that repeat offenders are handled correctly from the start.

Going forward, the weekly training session will include a role-play exercise focused on recognizing patterns in user behavior rather than just reacting to individual messages. We will also implement a shared log for staff to note suspicious accounts immediately, so that if a user returns on a new account, the next helper sees the history instantly. This reduces the cognitive load on individual volunteers and ensures that our response is based on the full picture rather than isolated incidents.

Related pages

Up to /archive/votes/bots/misc/ · Archive home