INTERNAL TOOLING
User Support/ Trust & Safety
Knowledge Base
This is an internal knowledge management system which I designed from zero to one for the trust and safety - content moderation team at global social media company.
Timeline
From explorations to final designs in 4 months.
My Role
To design and deliver a new content management system to be created for global content moderation agents which will be the central resource for all policy, enforcement and performance support resources needed to provide a high quality, consistent scaled policy enforcement service to better serve online communities.
My 4 key areas of responsibility included user research, style guide, user flow and UX/UI design - which including creating the design system from scratch myself.
Partners
2 Engineers, Program Manager, Content Designer, Data Scientist and 2 Policy Specialists along with a range of external stakeholders who were critical to my user research and testing efforts.
This category details the step-by-step approach taken during the project, including research, planning, design, development, testing, and optimization phases.
Context
Lets start with some context here and to set the scene - content moderation is the process of reviewing user generated content that has been uploaded and reported on social media pages.
Research & Planning
My first task here was a deep user research study to best understand the current processes along with the target audience’s needs and issues before progressing to any solutionizing.
My research approach was thorough, incorporating 270 surveys, 19 user interviews, focus groups, and detailed data analysis. This methodology not only deepened our understanding of the issues but also helped us define key success metrics and establish methods for measuring impact.
This took alot of engagement of both internal operational teams but also with our many external vendor partners. It also took considerable investment in sourcing and analysing data - luckily I had great support from a really collaborative data scientist who I worked with to get the right insights where were super important as it guided many design outcomes.
Target User Profiling
The impacted population here was around 6k agents with an annual attrition rate of around 30% so onboarding is a regular process. These agents are responsible for reviewing content for policy violations, globally dispersed working both in office and remote with less than 20% with English as a first language. Agents are trained on content policies through a series of lengthy instructor led decks which define the kinds of abusive content that can’t be posted on the social media platform.
Note that were also secondary user groups here including the operational full-time employees providing policy, quality and training support but very much struggling with no foundational structure so everything was very dynamic and unstructured - simply put, the organisation was not setting up support teams for success which was negatively impacting the global online community.
Problem Definition
The user pain points discovered can be grouped into three key themes: Efficiency, Reliability and Consistency.
Centralized Content Management System consolidates all policy guidance, local cultural abuse context, training materials, and quality support features.
A newly implemented notification system delivers essential policy updates to relevant agents, requiring acknowledgment prior to accessing live content moderation queues.
Single Source of Truth
Centralizes all policy guidelines and training materials, creating a reliable, definitive source for information.
Interactive Engagement
Facilitating an active feedback loop, enhancing the learning experience and adaptability of content.
Self-Guided Empowerment
Tools for self-assessment and performance improvement, coupled with easy self-service access to essential information for effective job performance.
Here, the outcomes and achievements of the project are highlighted, including user feedback, adoption rates, and industry recognition.
Increased Efficiency
Users report significant time savings and improved productivity through optimized scheduling recommendations.
Positive User Feedback
High user satisfaction ratings and positive reviews highlight the app's intuitive interface and powerful AI capabilities.
Growing User Base
The app quickly gained traction among individuals and businesses worldwide, with a steady increase in user adoption and engagement.