Rescheduling events by timezone heatmap: a participation case

Rescheduling events by timezone heatmap: a participation case

A 2025 case from a 35,000-member global Discord pulled three months of message-timestamp data to build an activity heatmap, then moved its weekly community event from a fixed US-evening slot to the data-identified global peak. Measured 8 weeks.

What the data shows
— Event attendance rose 41% after the time shift.
— The new slot captured a previously-underserved European/Asian overlap window the founder hadn't realized existed.
— Post-event channel activity (a retention proxy) ran ~25% higher on event days.

Why it happens
Founders schedule on their own timezone intuition, which systematically underweights the rest of a global membership. Timestamp heatmaps reveal the actual concentration of active members, which often sits in non-obvious overlap windows. Aligning a synchronous event to real availability is a mechanical, measurable win.

The caveat
Attendance is partly zero-sum across timezones — the +41% may have pulled from members who'd previously attended at the old time, netting less than it appears. Eight weeks won't show whether the new slot's novelty fades. And message timestamps measure existing-member activity, not the latent audience that never engaged because timing never suited them.

Open question: should you schedule for where your active members already are, or for the dormant timezone you're currently failing to serve?