Freedom & Democracy AI Volunteer

Election monitoring is high-stakes, high-pressure work. You're trying to build systems that will process critical data, often in challenging technical environments, with users who are under stress and may have limited technical training. Here's what I've learned from several projects in this space.

Lesson 1: Design for Failure

Networks will be unreliable. Devices will run out of battery. Users will make mistakes under pressure. Your system needs to handle all of this gracefully.

This means:

Extensive offline capabilities with smart syncing

Automatic saving and recovery from interruptions

Clear feedback about what's been successfully submitted

Multiple fallback submission methods (app, SMS, web)

I've seen incidents where observers lost hours of data because an app crashed and had no local persistence. Don't let this happen to your users.

Lesson 2: Simplicity Over Features

It's tempting to build comprehensive tools with every feature an organization might want. Resist this temptation. In the field, under pressure, users need tools that are immediately understandable.

Minimize the number of screens and steps

Use clear, unambiguous language (avoid jargon)

Make the most common actions the most prominent

Test with actual users in realistic conditions

One of my projects initially had a sophisticated categorization system for incident reports with dozens of options. In practice, observers just selected "Other" for everything because they didn't have time to find the right category. We simplified to five broad categories and got much better data.

Lesson 3: Plan for Scale at Critical Moments

Election systems experience extreme load patterns—low activity for months, then massive spikes on election day. Your infrastructure needs to handle these spikes reliably.

This isn't just about server capacity. Think about:

Human capacity to review incoming reports

Dashboard performance with many concurrent users

Alert systems that don't overwhelm recipients

Triage processes for when volume exceeds capacity

We typically run load tests simulating 10x expected peak traffic and still get surprised sometimes. Always have a degradation plan for when things exceed expectations.

Lesson 4: Build Trust Through Transparency

Election monitoring organizations need to trust your systems, and so do the broader public who will use the results. This means being transparent about how data is collected, processed, and analyzed.

Document your methodology clearly

Make it possible to audit data and decisions

Be upfront about limitations and uncertainties

Separate facts from analysis in your presentations

One effective approach is publishing detailed methodology documents before elections. This sets expectations, allows for feedback, and establishes credibility.

Lesson 5: Security is Non-Negotiable

Election monitoring systems are targets. Governments that want to suppress findings will try to compromise your systems. Organized disinformation campaigns may try to flood you with false reports.

Essential security measures:

End-to-end encryption for all sensitive data

Strong authentication for observers and staff

Rate limiting and anomaly detection for submissions

Secure communication channels for coordination

Physical security for any central infrastructure

Also plan for personnel security. Your team and observers may face intimidation. Have protocols for handling threats and know when to involve security experts.

Lesson 6: The Report Matters Most

All the data collection and analysis in the world is useless if it doesn't lead to clear, credible reporting. Invest heavily in how findings are presented:

Clear visualizations that tell a story

Plain-language summaries for non-technical audiences

Detailed technical annexes for those who want them

Multiple formats for different stakeholders (press, government, international observers)

I've seen excellent data analysis undermined by confusing presentations. The report is your product—treat it accordingly.

Final Thoughts

Election monitoring technology is a small piece of a much larger democratic infrastructure. The real work is done by thousands of observers, analysts, and advocates on the ground. Our job as technologists is to amplify their efforts and help their findings reach the people who need to see them.

If you're considering working in this space, reach out to established organizations first. Learn from their experience. And remember that humility is essential—we're supporting experts, not replacing them.