The engineering team chose Uber’s H3 hexagonal grid system—it could handle millions of data points but created a new UX problem: how do you show relationships between hundreds of hexes without creating visual chaos?

I researched tools like Kepler.gl and Uber Movement. Most used connecting lines between hexes, but our use case was different: every hex could be both an origin AND destination. Traditional flow diagrams would create an unusable spaghetti map.

During discovery we explored using multi‑size hexes to let planners drill down into smaller areas, similar to other H3-based tools. We ultimately rejected this pattern. Allowing very small hexes in low-density regions would increase the risk of exposing an individual athlete’s ‘pattern of life’.

Keeping a single, coarser hex size protected athlete privacy and ensured the map only surfaced truly aggregate behavior, not individual routines.