Signetic started as flu vaccination software. Then COVID hit, and suddenly we were the system running mass vaccination events for the City of Seattle — including Lumen Field.
The stakes were different from anything I’d worked on before. If our software failed, people didn’t get vaccinated. Missed appointments meant wasted doses. Confusing UX meant elderly and non-technical users couldn’t book. Reporting errors meant public health agencies lost confidence in us. It wasn’t a typical startup where a bug means someone’s dashboard looks wrong for an afternoon.
What I came to understand was that trust — not features — was the thing we were really building. Public health agencies needed to believe we wouldn’t fail them during the worst crisis of their careers.
The Speed Was Relentless
We shipped capabilities in days and weeks, not quarters. Policy changed constantly — eligibility rules, reporting requirements, equity mandates — and each change rippled through the product. Traffic spiked unpredictably whenever a new eligibility group opened up or supply arrived.
The question that actually mattered wasn’t “How do we build vaccination scheduling software?” It was more like: how do we build something that public health agencies can rely on at scale, under uncertainty, while serving people who are scared and confused?
That question drove most of the product decisions we made.
We Sold a Service, Not Just Software
One of the more important positioning decisions was framing Signetic as a managed vaccination platform, not a self-serve SaaS tool. We paired the software with operational support — real humans coordinating with public health partners during live events.
This mattered because municipalities didn’t want to learn a new tool in the middle of a crisis. They wanted accountability. They wanted to call someone. That framing reduced adoption friction in ways that better features probably never could have.
I think this is something worth remembering: how people behave under pressure is different from how they behave in a demo. A county health official managing a 10,000-person vaccination event doesn’t want a self-serve knowledge base. They want a phone number.
Knowing What Not to Build
I pushed to keep us focused on operational reliability. We chose stability over experimentation. Simple, repeatable workflows over customization. Proven flows over speculative features.
When you’re serving tens of thousands of people per day at a live vaccination site, the cost of a feature that doesn’t quite work isn’t “we’ll fix it next sprint.” So we were careful.
We also designed explicitly for access and equity. That meant accounting for low technical literacy, device and bandwidth constraints, and language needs. The goal wasn’t elegant design — it was broad and fair access. Some of our most important UX decisions were about what to remove, not what to add.
I found that the discipline of knowing when to stop adding was one of the hardest parts of the job. In a crisis, the temptation is to say yes to everything. But every new feature is a new potential failure mode, and we couldn’t afford many of those.
When the Pandemic Ended
This is the part that doesn’t fit neatly into a success story. As vaccination demand declined, we had to confront the fact that our addressable market was shrinking. We tested pharmacy models and subscription approaches. Some worked, some didn’t.
We tried to make those calls quickly and transparently. Not every expansion effort succeeded, but I think facing reality early mattered more than getting every bet right.
What I Took Away
The platform supported roughly 415,000 COVID vaccinations. We earned sustained trust from public health agencies during the peak of the crisis. We proved we could scale under extreme load.
But the deeper lesson for me was about the limits of software. Software alone isn’t enough without aligned incentives, organizational readiness, and real adoption. You can build the most reliable system in the world, and it still needs humans who trust it enough to use it.
I learned a lot about staying calm in high-stakes environments, and about how the best product decisions in a crisis tend to be about what you choose not to build. Automate the critical path. Keep the humans focused on the decisions only they can make. That’s about all you can do.