Case Study · Content Strategy Tooling

Help Center Gap Analysis Dashboard

A full-stack tool that turned scattered support signals into a ranked, evidence-backed list of what to fix first — across all five of Meta's Help Centers.

5

Help Centers unified into one view

4

Independent signals weighted per gap

P0–P2

Priority tiers auto-assigned by score

Project overview

Context: Meta's distributed content strategy teams manage information across five separate Help Center platforms. Identifying underperforming content or missing information required teams to manually hunt for data.

Challenge: Critical user-friction data points were scattered across siloed ticket queues, bug reports, and surveys. This manual auditing overhead delayed content deployments and obscured top-priority data gaps.

Approach: Build a functional, automated dashboard to centralize cross-surface data inputs, instantly surfacing and prioritizing critical content gaps for operational execution.

Users can type to filter gaps by title or help center name in real time. See the fully functional, sanitized mock-up →

Explainable scoring

Every score shows its work

Clicking into any gap breaks the composite score down signal by signal — support case volume, AI agent misses, in-product friction, and verbatim feedback — each with the specific reasoning behind it, like "10 bug reports linked to this gap" or "negative sentiment trending upward over the last 30 days."

My role

Interactive Interface Prototyping

Designed and built a functional frontend dashboard featuring an aggregate ecosystem health panel and expandable signal breakdown panels to make complex data instantly readable.

Signal Weighting Engine

Created the logic for a custom recommendations matrix that automatically calculates priority scores based on volume and topic clustering across four independent signal sources.

Accessible Data Architecture

Structured the system to run on an accessible matrix layer, allowing non-technical content owners to safely adjust signal weights without needing engineering support.

Impact Projection & Prioritization

Built a predictive layout component that displays estimated impact per gap — helpfulness lift, case reduction, users reached — so partners can see the value before any editing begins.

This is a sanitized, publicly shareable recreation of an internal Meta tool. The layout, scoring logic, and interactions are real; the underlying data has been replaced with illustrative figures so nothing confidential ships in a public link.