Real-Time Graphics Infrastructure for Live Election Coverage

How I designed and documented a complete live graphics system that enabled distributed contractors to execute complex broadcast graphics in real-time during high-stakes election coverage—with zero margin for error.

The Washington Post Video Department | 2016-2025

The Challenge

Live election coverage is one of the most demanding environments in broadcast journalism. Results change every 30 seconds. Graphics need to update in real-time. There's no room for errors when millions of viewers are watching.

Major news organizations typically staff election night with 15-20 motion graphics artists working in shifts. We had a 3-person core team and needed to deliver the same quality across primary elections, midterms, and presidential races—often with simultaneous state and national coverage.

The traditional approach—designers manually creating each graphic—doesn't work when vote counts update continuously and winner calls happen unexpectedly. We needed a fundamentally different system.

The Scale of the Challenge

Election night requires dozens of simultaneous graphic elements: live vote counts, winner boards, breaking news banners, analyst name cards, county-by-county maps, delegate trackers, exit poll data, and real-time headlines. All updating dynamically. All needing to maintain brand consistency. All with zero tolerance for on-air mistakes.

The Solution: A Template-Based Live Graphics System

Rather than throwing more people at the problem, we redesigned the entire production workflow. I architected a complete live graphics infrastructure in Brainstorm (broadcast graphics software) that allowed a single graphics operator to manage 50+ live elements simultaneously—without design team intervention.

The System Architecture

I designed a hierarchical template system organized into four layers (CG1-CG4), allowing operators to stack multiple graphics without conflicts:

What Made This Different

Each template wasn't just a design—it was a parameterized system with:

  • Toggle controls (on/off options for various elements)

  • Text fields with character constraints

  • Image slots with positioning/cropping controls

  • Color selectors for party affiliation

  • Automated data integration for live vote counts

  • Built-in animation timing

The system prevented operator error by only exposing valid options. You couldn't accidentally create an off-brand graphic because the templates encoded all design decisions.

Building a System That Others Could Execute

The real challenge wasn't just designing the graphics—it was enabling a distributed team to execute them flawlessly during live broadcasts.

Comprehensive Technical Documentation

I created a complete operator guide with:

Visual Examples:

Every template shown with actual on-air examples and annotated screenshots showing editable parameters

Technical Specifications:

Detailed breakdown of every control, toggle, and text field for each graphic

Workflow Diagrams:

Decision trees showing which graphic to use in specific scenarios (speaker on-screen vs. voice-over, captioned vs. uncaptioned video, etc.)

Edge Case Handling:

Notes on special situations and future enhancements

Contextual Instructions:

When to use each graphic, how they layer together, timing considerations

Why Documentation Mattered

This system enabled a graphics operator in Spain to execute designs I created in DC during live election broadcasts—with zero latency for design approval. The templates encoded all brand standards upfront, and the documentation provided the operational knowledge needed to make real-time decisions under pressure.

Distributed Team Coordination

For each election, I coordinated contractors across four time zones:

  • DC (core team)

  • Austin, Texas

  • Guatemala

  • Spain

The documentation system allowed each operator to work independently while maintaining consistency. Clear handoff procedures, asset libraries, and timezone-optimized workflows meant we had 24/7 coverage when needed.

The Results

The system delivered across 8 major elections from 2016-2024:

  • Matched competitor output with 80% fewer people – Our 3-person core team + 2-4 contractors delivered what other organizations needed 15+ people to achieve

  • Real-time responsiveness – Vote counts and winner calls updated within seconds of data availability

  • Consistent quality across elections – The system worked identically for primaries, midterms, and presidential races

  • Knowledge transfer without training bottlenecks – New operators could learn the system through documentation alone

  • Cost efficiency – 70% reduction in personnel costs compared to traditional staffing models

What This System Required

Building infrastructure for live broadcast operations demanded different thinking than traditional design work:

Real-Time Data Integration

Graphics connected to live election APIs, automatically updating vote counts and winner calls as results came in. The system had to handle data refreshes every 30 seconds without disrupting on-air graphics.

Fail-Safe Design

Every template included fallback states and error handling. If data failed to load, graphics displayed placeholder content rather than breaking entirely.

Performance Under Pressure

Graphics had to render instantly when triggered. No loading delays, no lag, no buffering. When a winner was called, the graphic appeared on-screen within 2 seconds.

Multi-Output Coordination

Single graphics source fed multiple outputs: main program feed, video wall, touchscreen displays, streaming platforms. All synchronized perfectly.

Key Success Factors

1. Documentation as Operational Infrastructure

The comprehensive guide wasn't just nice-to-have—it was mission-critical infrastructure. Without it, distributed teams couldn't operate independently. The documentation transformed tacit knowledge into explicit systems.

2. Parameterized Design That Prevents Errors

By encoding design decisions into template parameters, we eliminated the “creative freedom to make mistakes.” Operators could work confidently because invalid choices weren't possible.

3. Hierarchical Organization That Scales

The CG1-CG4 layer system meant operators could manage complexity systematically. Graphics stacked predictably without conflicts, allowing dozens of simultaneous elements on-screen.

4. Testing Under Real Conditions

We didn't just build and hope. Every template was tested under actual broadcast conditions with real data feeds and live operators. We caught edge cases before they became on-air problems.

Key Insight

“At the end of the day, it’s about how we provide journalism that’s credible, reliable, and trustworthy in ways that are accessible to the audience”

Catherine Kim, NBC News executive vice president of editorial

Beyond Election Night

This wasn't just an election graphics system—it was a blueprint for how to architect operational infrastructure for high-stakes, real-time creative work.

The principles applied here translate directly to any environment where:

  • Multiple team members need to execute consistently

  • Real-time constraints don't allow for approval cycles

  • Quality standards are non-negotiable

  • Distributed teams must coordinate seamlessly

  • Documentation must enable independent operation

Whether it's live broadcasts, rapid-response social content, or real-time marketing campaigns, the same systems thinking applies: build infrastructure that encodes quality, document it comprehensively, and empower teams to execute without bottlenecks.

Continuous Improvement

The system wasn't static—it evolved across 8 elections:

  • 2016 Primaries & General: Initial system with basic templates

  • 2018 Midterms: Added touchscreen integration and expanded ticker

  • 2020 Primaries: Introduced pandemic-era remote workflows

  • 2020 General: Refined real-time data integration

  • 2022 Midterms: Streamlined multi-state coverage

  • 2024 Primaries: Enhanced distributed team coordination

  • 2024 General: Peak optimization with full automation

Each election taught us something new. Each refinement made the next election smoother. That's the nature of good operational systems—they compound over time.

Interested in how this approach to technical documentation and systems architecture could apply to your operations?

Contact Me
← Back to Case Studies