Waze makes its data available for you through a localized XML or JSON GeoRSS feed that's updated every 2 minutes.
View the Waze Data Feed Specification
To view this data:
- Log in to the Partner Hub
- Click Toolbox, in the sidebar
- Click Waze Data Feed
The feed links contain Waze traffic data for your polygons.
With these links, you can:
- Apply the data to your control center maps
- Collect the data for analysis
Reported incidents included in the Waze Data Feed
Data based on reports from Waze users:
- Traffic accidents
- Hazards
- Construction
- Potholes
- Roadkill
- Stopped vehicles
- Objects on the road
- Missing signs
Data automatically detected by Waze:
Waze automatically calculates traffic jams by comparing current road conditions with historical road data. When traffic on a segment is below average speed for that time of day or week, we identify it as a slowdown or jam. These are then compared with user GPS signals and user reports.
When parsing data across real-time streams, note the following configurations to prevent database mapping errors:
For jams- speedKMH: km/h
- speed: m/s (this is the exception, for historical reasons)
For irregularities
- speed: km/h
- regularSpeed: km/h
The maximum number of events in the Waze Data Feed is 5,000.
Got questions?
Why are my data feed links not working?
The URL format was updated on March 27th, 2023, to improve the data-sharing processes between Waze and our partners. Support for links created before this date ended in May 2023.
- New format: https://www.waze.com/row-partnerhub-api/partners/116773849/waze-feeds/646a7d93-6e34-42ca-b3aa-15487953abcd?format=1
- Previous format: https://www.waze.com/il-rtserver/web/TGeoRSS?tk=ccp_partner&ccp_partner_name=testPartner&bottom=32.06342712281313&left=34.75107670953731&ma=200&mj=200&mu=20&right=34.79582883050899&top=32.091877863016826&types=alerts%2Ctraffic%2Cuser
What is a Waze incident and why should I exchange incident data with Waze?
A Waze incident is any planned or unplanned event that disrupts a roadway. An incident could be a pothole, an object in the roadway, flooding or ice, construction, a traffic crash, or another disruption.
By sharing incidents with Waze, you help inform drivers and improve road safety in your community. By receiving incidents, you can better manage day-to-day traffic operations and view long-term trends to inform policy or infrastructure decisions.
How can I use Waze incident data to help my community?
Here’s how you can use Waze incident information to enhance traffic and roadway management in your area:
- Monitor and respond to real-time driver traffic reports from your operations center
- Identify problem areas reported by drivers, like potholes and other hazards, so that you can prioritize maintenance on certain road segments
- Feed larger studies, such as using reports of water on roadways to feed models that predict which roads might flood under certain weather conditions
How have other partners creatively used Waze incident data?
The city of Rio de Janeiro in Brazil created a heatmap of potholes using 42,000 incident reports, then cross-checked the map with traffic jam data to better plan and prioritize their road maintenance efforts.
The Miami-Dade Water and Sewer Department (WASD) used Waze data to update internal departmental construction and maintenance systems, improving quality control, research, and analysis.
Check out our case study gallery for other innovative ways partners have used Waze incident data
How do incidents appear on the Waze map?
How can I assess the accuracy of crowd-sourced incidents?
Reliability scores: Every alert receives a reliability score (0 to 10) indicating how trustworthy the report is. This is based on user reactions ("Thumbs up" or "Not there") and the experience tier of the reporter, who can scale from level 1 to level 6.
Every new alert automatically starts with a reliability score of 5. It drops below 5 only in rare instances where a highly ranked map editor submits a "Not there" report against the incident. This factors in how well Waze knows the reporter before other drivers have had a chance to verify it.
Confidence scores: Each alert gets a confidence score based purely on real-time driver reactions on the ground ("Thumbs up", "Not there", and comments). The score ranges between -1 and 5.
The system calculates this score dynamically using the following formula:
formula = Min( (# comments * 10 + # thumbs up * 10 + # not there * -10) / 20 ) , 5)
Automatic removal: If an incident receives exactly 2 "Not There" reports and has zero active "Thumbs Up" or "Comments", the confidence score drops to -1. At this point, the alert is automatically removed from the feed and Live Map.
A user's editor level reflects their history of constructive contributions to the Waze map. Learn more about a user's map editing level here.
Receiving Incident Data
When sharing Waze’s data, please follow attribution guidelines. Learn more.