User Engagement Measurement with Plausible Custom Events

Petworth PorchFest 2026 (PPF) is only about a week away. The forecast looks good (knock on wood, fingers and toes crossed). With more than 400 bands slated to play in a single afternoon, it’s the biggest PPF yet, by far the biggest musical festival in the DC area, and among the very biggest music festivals in the country. In response, we considerably upgraded the festival’s tech stack this year by moving from a google sheet schedule to a proper web-based interactive version and by introducing PorchFinder, a web-app that allows advanced searching and custom itinerary building.

We built PorchFinder rapidly. The first version was built in days about a month before the big day. We’ve been steadily adding features (and squashing bugs) as the event approaches. The newest addition is an under the hood enhancement to give us user engagement data so that we can properly assess if people are indeed using PorchFinder. By default, we know how many people load the app, but that doesn’t say much. In particular, it doesn’t tell us anything about whether people are building their own custom itineraries, sharing them, or whether recipients are looking at them. Today we implemented custom event tracking for several key interactions. Our Plausible dashboard will start to report specific on-page actions users are taking such as creating itineraries and creating share links. We also receive a signal if/when a recipient of a shared itinerary opens it.

For our analytics, we use Plausible, an excellent privacy-friendly alternative to Google Analytics. This allows us to get some useful engagement data to measure, for example, usage levels without collecting any personal information. One important note is that Plausible’s business model depends on subscription fees (as opposed to being “free” like Google’s offering where they just harvest and monetize your data). The costs go up as your pageview and custom events go up. Unfortunately because PPF runs on a shoe-string budget, we can’t afford one of their higher-end plans. It’s very possible that with the limitations of our current plan, we won’t have a sufficient amount of pageviews to measure the full impact of our new tech stack. On the days leading up to the event plus event day, we typically serve 50k to 60k users. With event tracking place, that number will be even higher. But at the very least, it should help us differentiate between a fruitful product/feature and a dud, and give us some direction for Petworth PorchFest 2027!