The data layer, alert channels and your existing stack

Trafficure sits between two things a city already has: Google's traffic data on one side, and the tools the team works in on the other. Google Maps Roads Management Insights is the data layer, the plumbing that delivers anonymized speed and movement signals per road segment. Trafficure is the activation layer that turns that plumbing into a working command centre, then sends what it produces back out across the channels and formats the agency already uses.

The principle through the whole stack is that it stays open at every edge. The data comes from Google. The alerts go to channels you choose. The analytics export in standard formats you own. Nothing about the stack locks a city in.

The data layer is Google Maps Roads Management Insights, launched in August 2025 and sourced from aggregated, anonymized movement across the more than one billion smartphones running Google Maps and Waze. Lepton is an authorized RMI reseller, so the data subscription and the activation layer can be bought together. RMI arrives through three interfaces, each doing a distinct job.

The data layer keeps widening. As of May 2026, Google has added experimental RMI feeds that Trafficure can build on: anonymized vehicle counts, harsh-braking event clusters, and crowdsourced incident reports.

The biggest unlock ahead is simulation, the ability to model an intervention before it is built. These are early feeds, named here so the stack is honest about where it is heading rather than overstated about where it is today.

When a corridor begins to deviate from its baseline, Trafficure raises an alert and delivers it to the team where they already are. There are six channels, and a city configures which ones carry which alerts. Thresholds, priority levels and auto-resolve are all set per agency, so a minor build-up and a major incident do not arrive the same way.

Between the data layer and the channels sits the analytics pipeline. It normalizes more than 15,000 road segments per city against the road graph, builds baselines from roughly ninety days of history, and runs the full cycle in under 120 seconds.

The output feeds the dashboards and alerts, and it ships out as PDF reports and standard-format exports. Because exports are standard formats, the data a city produces stays the city's to keep and to move.

The data layer is built on aggregated, anonymized movement with k-anonymity applied at the source. There is no individual-level view in the stack. The posture is aligned with the GDPR and with India's DPDP Act.

The same property that protects privacy is also an honest limit, because k-anonymity means very sparse roads can drop out and there is no commercial-fleet-only view, which is why Trafficure stays complementary to a camera ITMS that keeps enforcement and number-plate recognition.