The Election Platform I Never Planned to Build

Before Comitia had a name, Donald Kwaw walked up to me in the library with a problem. His department needed a cheaper way to run an election, and he thought I might be able to build it.

He had seen some of my earlier work, including the UCC SRC app, AZ Learner and the Oguaa Hall Army Cadet app. Still, there was no contract, no roadmap and no certainty that the department would even choose me. I began researching anyway.

Today, Comitia is a digital election platform for student groups, associations and small organisations that want elections to be simple, affordable and trustworthy. It supports on-site and remote voting, role-based election management, voter authentication, live monitoring and audit records. But none of that was the original plan.

The original plan was simply to help one department vote.

Poll 1: The first election

Donald and I had served the Oguaa Hall Army Cadet, Oguaa Hall and the University of Cape Coast together. He was also a member of the Agri-Business Students' Association of Ghana, or ABSAG, in UCC's Department of Agricultural Extension and Economics. One day during the second-semester election season of the 2025/2026 academic year, he told me that ABSAG was looking for someone to design an election portal and that he had recommended me.

I was excited, obviously. It was a chance to learn something new and perhaps change how a departmental election could be run.

This is not to say I invented digital elections. Hell, the university had been using computers for voting before I even walked through its gates. AI had also made it easier for me to learn how to build working systems. But I had never built an election platform, and ABSAG did not have a digital election workspace of its own.

Before I received the final go-ahead, I researched election security, privacy, transparency and ways to prevent duplicate voting. I also investigated whether Firebase could support the infrastructure. Then the confirmation came.

I did not assume I already knew what ABSAG wanted. I created a form and sent it to Donald and his team, asking what they needed and what they did not. Their answer was refreshingly simple: a cheap, secure and transparent way to vote.

The system I built - originally called ABSAG-UCC - was tiny compared with Comitia today. The admin side had only three main screens. The dashboard opened, paused and closed the election. The candidates screen allowed us to add candidates individually or import their information from a CSV file. The voters screen registered voters using their index numbers and dates of birth.

At the registration desk, voters also had to present a student ID or Ghana Card to prove that the details belonged to them. Once someone voted, the same record could not be used again. It was a one-vote-per-voter system, not a maze of features pretending to be sophisticated.

The polling-agent dashboard showed how the election was progressing. The voter journey was equally direct: sign in, select candidates, review the ballot and submit it.

Election day was simple too. Voters arrived at 8:00 a.m., and voting began at exactly 8:30 a.m. Three administrators registered voters before they moved to the voting station. They entered their details, cast their votes and left. The system prevented a second vote from the same voter record. There were no errors, no delays and no long queues.

That first election proved something to me: an election system did not have to be complicated to be effective.

Poll 2: The second customer changed everything

Comitia could easily have ended with the ABSAG election. Believe me, I thought it had. We held the vote, published the results and moved on. A week later, I even called Donald to ask whether I could take the website down.

Then an aspirant in my own department's election called me. The moment he explained what they needed, I knew what to show him. I described the platform I had built for ABSAG and offered to adapt it for the Hospitality and Tourism Students' Association, HOTSA. They agreed.

That second election introduced the requests that became the foundation of Comitia. HOTSA wanted students to vote from anywhere. Other organisations wanted results published directly on the site. Administrators wanted to upload voter records in one batch instead of registering people one by one. They wanted clearer roles for administrators, polling agents and returning officers.

The platform was no longer being shaped by what I imagined organisations might want. It was being shaped by what actual organisations were asking for.

It also had its share of problems. Saying otherwise would be a lie.

During the ABSAG election, some students who had been registered still received a 'details not found' message when they tried to sign in. We discovered that a network failure had interrupted the registration, but the portal's generic confirmation message made it appear successful. The software had failed to explain what had really happened.

Later, while preparing for the HOTSA election, the volume of SMS messages delayed the delivery of some voters' unique codes. That meant a few people received their credentials late and could not vote on time. The response was practical: send credentials a day or two before the election, create a request page for registered voters who did not receive a code, and introduce clearer error messages so a failed action would no longer masquerade as a success.

Those were not glamorous improvements. They were more valuable than glamour. A polished button is useless if the voter behind it cannot enter the election.

As the system expanded beyond a single vote, I first renamed it Pollaris. I later changed the name to Comitia after discovering that another election platform was already known by the Pollaris name. The renaming was cosmetic; the real transformation was underneath. Comitia had become a reusable platform rather than a one-off departmental website.

Poll 3: What Comitia is today

Comitia now allows an organisation to manage the full election lifecycle from one place. Administrators can create elections and contested positions, register or import voters, add candidates, arrange ballot positions, invite election officials, personalise the voter interface, monitor election-day activity, publish results and export election records.

For remote elections, voters receive a unique eight-character alphanumeric code and an election link. The code is designed to make unauthorised guessing impractical and to give each registered voter one opportunity to submit a ballot. For on-site elections, the same platform keeps the process quick without forcing voters into a long queue.

Polling agents do not need a campaign scoreboard. Their interface focuses on operational information such as voter turnout and ballots cast, while allowing them to report incidents as the election progresses. Returning officers and administrators can follow the process without turning live monitoring into a source of unnecessary tension.

Some administrative powers are necessarily sensitive. If an authorised official must disable a voter record or reset a voting status to resolve a verified problem, that intervention should be exceptional and traceable. Comitia's audit trail exists so administrative actions are recorded rather than hidden. A platform cannot claim to make elections trustworthy while giving invisible power to the people running it.

“The online election system was simple, convenient and transparent, making the voting process smooth and stress-free. It was a great step towards ensuring a fair and efficient election.”

The platform now includes election controls, team roles, incident reporting, audit records, result certification, data exports, retention settings and branding options. That sounds like a feature list, but the important point is where those features came from.

There are many projects I started because I thought the world could use this or that. Comitia is not one of them. It was built and refined through real elections. Its strongest features did not arrive because I sat alone and imagined a perfect product. They arrived because voters, candidates and election officials exposed the gaps.

Poll 4: Where I want to take it

The future of Comitia is not about adding features for the pleasure of making the dashboard longer. It is about making credible digital elections accessible to organisations that cannot afford expensive enterprise systems.

That means improving voter verification, strengthening authentication, making audit records easier to understand and supporting remote and hybrid elections responsibly. Remote voting can include people who would otherwise miss an election because of distance, queues or conflicting schedules. It also makes identity verification more consequential, so convenience cannot be allowed to outrun trust.

Accessibility matters too. Voters, administrators and polling agents should be able to adjust contrast and motion without fighting the interface. The platform should work for universities, associations, unions and other membership organisations, while remaining understandable to the small teams that actually have to run their elections.

And yes, there is a pleasantly mundane benefit: nobody should have to burn fuel merely to put ink beside the face of their favourite candidate.

I am not claiming that software makes an election fair by magic, or that digital voting is free of risk. Governance, identity checks, clear procedures and accountable officials still matter. Comitia's job is to make those processes easier to operate and easier to audit - from voter registration to the publication of certified results.

Poll 5: What building it taught me

Start small, but solve a complete problem

The first ABSAG version was not a giant platform. It had three admin screens, one polling-agent view and a simple voting flow. But it solved the whole immediate problem. People could be registered, authenticated and allowed to vote once; officials could monitor the process; and results could be produced. Small did not mean unfinished.

Real users are a better roadmap than imagination

I did not originally plan remote voting. HOTSA asked for it. Batch voter imports were not central to the first build either. Organisations using the platform made that need obvious. Real users do not merely validate a product; they reveal the parts its creator was too far away to see.

Reliability is part of the product

The delayed SMS messages and misleading registration confirmations taught me that infrastructure is not background decoration. Delivery timing, network failures, recovery paths and useful error messages are the product when someone is trying to vote. In an election, a gorgeous interface cannot compensate for a credential that arrives after the polls close.

Being trusted changes how you build

A personal project can tolerate a broken corner while I promise myself I will fix it later. An election cannot. Candidates, voters and officials are trusting the system with something consequential. That trust forces different questions: What happens when the network fails? Who can change a voter record? Can an intervention be traced? How does someone recover if an SMS never arrives?

Looking back, there are decisions I would make differently. Some features arrived too early, others too late, and several assumptions disappeared the moment real voters touched the platform. That is not an embarrassment. It is the difference between designing a product in theory and building one in public.

Declaration of results

Comitia started because Donald Kwaw walked up to me with a problem. His department needed a cheaper way to run an election. There was no grand plan, no elaborate roadmap and certainly no expectation that the project would grow beyond that vote.

Sometimes products do not begin with an idea. They begin with someone trusting you with a problem.