Case study

Making critical health information easier to find and understand for millions of people.

Auditing, restructuring, and building scalable content systems for CDC.gov so the general public, healthcare providers, and public health professionals could all find what they needed.

ClientCenters for Disease Control and Prevention (via Peraton)
RoleInformation Architect
TimelineApril 2023 to July 2024
ResultScalable content system used across hundreds of CDC sites

The challenge

CDC.gov serves the general public, healthcare providers, and public health professionals. Before the redesign, the site had accumulated years of duplicative, outdated, and inconsistently structured content across hundreds of subsites. Visitors struggled to find clear, current information, and content creators across the CDC lacked a shared framework for how to organize and write health content.

100s
CDC subsites audited
3
Distinct audience types to serve simultaneously
1
Scalable content system to replace it all

The scale of the problem made one-off page fixes impossible. The only viable solution was a systems approach.

The modernization initiative aimed to streamline the site so people could find what they need faster, written in language they could understand. The challenge was doing this at scale across an enormous, decentralized organization where scientists, writers, and subject matter experts all contributed content independently.

What I did

I conducted a comprehensive content audit across hundreds of CDC sites, identifying what to keep, consolidate, archive, or rewrite. But the audit was just the starting point.

1
Comprehensive content audit
Reviewed content across hundreds of CDC subsites to identify what to keep, consolidate, archive, or rewrite. Established a baseline for what existed and what was missing.
2
Content models and content types
Created reusable models that defined how different types of health information should be organized, what fields they needed, and how they related to each other. Any new health topic added to the site would automatically inherit a clear, tested structure.
3
Training writers and scientists
Created training materials, ran workshops, and provided ongoing support as teams migrated their content into the new system. Scientists write for other scientists. My job was to help them see their content through the eyes of a parent searching for answers at 2 AM.

The deeper work was creating content models and content types that gave migrated content a consistent structure. These models defined how different types of health information should be organized, what fields they needed, and how they related to each other.

Key decisions

Three decisions shaped how the system was designed.

One topic, three audiences — each with different needs
General public
  • Plain language
  • Practical actions to take
  • What this means for me
  • When to see a doctor
Healthcare providers
  • Clinical guidance
  • Treatment protocols
  • Diagnostic criteria
  • Reporting requirements
Public health professionals
  • Surveillance data
  • Field guidance
  • Policy frameworks
  • Outbreak response

The same health topic had to serve three audiences without diluting the information for any of them.

Content model: how health topics were structured
Condition overview
  • What it is (plain language)
  • Symptoms
  • Who is at risk
  • What to do
  • Related resources
Clinical guidance
  • Diagnostic criteria
  • Treatment protocols
  • Reporting requirements
  • Evidence base
  • Last reviewed date
Data summary
  • Key statistics
  • Trend data
  • Geographic breakdown
  • Data source
  • Methodology notes
Professional resource
  • Purpose and audience
  • Field guidance
  • Tools and templates
  • Training materials
  • Contact for questions

Reusable content models meant any new health topic added to the site automatically inherited a clear, tested structure.

Audience breakouts
Structured content so the same health topic could serve three different audiences without diluting the information for any of them. Each audience got what they needed in the language they used.
Models over one-off pages
Rather than redesigning individual pages, focused on creating reusable models that would scale. The long-term value comes from systems, not one-off fixes.
Plain language without losing accuracy
Content had to be scientifically accurate and accessible to non-experts. Worked with subject matter experts to focus on practical actions people could take rather than clinical abstractions.

The outcome

The redesigned CDC.gov launched with streamlined information, new navigation, improved readability, and audience-specific content paths. The CDC's public announcement noted that the site was redesigned based on testing with real users and that the goal was to provide information that is easier to find and understand.

The content models I created continue to be used as new health topics are added to the site, giving CDC teams a scalable framework rather than a one-time redesign.

The most important outcome was not the launch. It was what happened after. Any new health topic added to CDC.gov now inherits a clear, tested structure automatically. The system does the work that used to require a project.

What this shows

I can take a massive, complex health content ecosystem and give it structure, clarity, and consistency, while training the people inside the organization to sustain it without me. The work is not just about making content better. It is about making the conditions for good content permanent.

← Back to all work