Where the data comes from
Nothing on this site is scraped from another website and no figure is entered by hand. Everything is drawn from published open data feeds. This page lists all of them, what each one contributes, how often it refreshes, and the licence it is used under.
Network Rail Open Data
The operational railway. Network Rail publishes these feeds to registered users through its open data platform; they are the source of the timetable, the live positions and the punctuality figures.
SCHEDULE — the timetable
The full national rail timetable in the industry's CIF format: every scheduled service, its calling points, timings, platforms, operator and the days it runs, including short-term plan variations and overlays. Reloaded daily in the early hours.
VSTP — very short term plan
Schedules created or altered at short notice — diversions, extras, and services retimed after the daily timetable was published. Streamed continuously and applied to the stored timetable as it arrives.
TRUST — train movements
Reports of trains passing reporting points, with the delay against schedule, plus activation, cancellation, reinstatement and termination. This is the source of "last reported at" and of the delay attached to a running service. Continuous.
TD — train describer
Signalling berth steps: the messages that move a headcode from one section of track to the next. This is what puts trains on the live map. Continuous, at roughly twenty to sixty messages a second nationally.
RTPPM — real-time public performance measure
Punctuality by operator and by sector, updated about once a minute, and the source of everything on the train performance page.
TSR — temporary speed restrictions
The weekly national list of temporary speed restrictions, published on Friday mornings. Restrictions are matched geometrically onto route legs and shown as a warning line beneath the affected leg.
SMART and CORPUS — reference data
SMART maps signalling berths to location codes; CORPUS maps location codes to timing point codes, station names and the various industry identifiers. Together they are what turns a berth step into a point on a map. Re-downloaded automatically when more than a week old.
Used under Network Rail's open data licence. Registration is free at publicdatafeeds.networkrail.co.uk.
National Rail Enquiries
Where Network Rail publishes what the railway is doing, National Rail publishes what it is expected to do, and the passenger-facing reference information.
Darwin Push Port
The national forecasting system: per-stop estimated and actual times, platform numbers and platform changes, cancellation and delay reasons in plain English, and whether a service carries passengers at all. This drives the departure and arrival boards and the expected times on route popups. Streamed continuously, with a full timetable snapshot loaded periodically so the boards are complete rather than only covering services seen since startup.
KnowledgeBase
Two separate datasets. Incidents gives current disruption — what has happened, which operators and routes are affected, and the advice being given; refreshed every ten minutes. Stations gives the station directory: postal addresses, facilities, step-free access and accessibility information, behind the station lookup.
Historic Service Performance
Punctuality history for a specific origin, destination and time of day over the last four weeks, used to put a reliability figure on a journey leg. Queried on demand and cached, because it is a slow and rate-limited service.
Used under the National Rail Enquiries open data licence; accounts are free at opendata.nationalrail.co.uk.
Bus Open Data Service
Operated by the Department for Transport, BODS is the statutory publication route for bus information in England, and carries a good deal of Scottish and Welsh data too.
Location data (SIRI-VM)
Live vehicle positions with line, operator, origin and destination, journey reference and, where the operator provides it, bearing and occupancy. This is the source of the buses on the map and of the punctuality estimate. Note that it carries no delay figure — see how it works for how one is derived.
Timetable data (GTFS)
The national bus timetable — around 300,000 stops and some 53 million connections between them. Used for stop departure boards outside London and for bus and door-to-door journey planning. Rebuilt automatically each week.
Disruptions (SIRI-SX)
Planned and unplanned bus disruption — roadworks, diversions, cancellations — shown on the map and in the disruption list alongside the rail equivalent.
Stop, place and operator reference
NaPTAN
The national register of public transport access points: every bus stop, coach stop and station entrance in Great Britain, with its ATCO code, position and the locality it belongs to. Around 344,000 stops.
NPTG
The national gazetteer of localities — the towns, villages and suburbs that stops belong to, and the administrative areas above them. This is what makes it possible to browse bus stops by place rather than by code, and to tell six different Kingswoods apart.
NOC
The national operator codes database, which turns an operator reference in a live feed into a company and a public brand name.
Transport for London Unified API
Used only for live bus arrival predictions at London stops, where TfL's own countdown predictions are better than anything derivable from the national feeds.
NaPTAN, NPTG, NOC and BODS data are published under the Open Government Licence.
Geography and engineering reference
OpenStreetMap
The track geometry itself. Every route line drawn on this site follows a graph built from OpenStreetMap's railway data, including the Engineer's Line Reference recorded against each stretch of line. © OpenStreetMap contributors, available under the Open Database Licence.
Network Rail engineering exports
Two published datasets add industry detail the map alone cannot give: track position data, which yields the ELR and miles-and-chains mileage shown against waypoints; and track gradient data, which gives the climb, descent and steepest sustained gradient reported for each leg of a route.
Accuracy, and what to trust
These feeds are the same ones the industry runs on, but that does not make this site authoritative. Reference data can be stale, a berth can be misconfigured, an operator can publish a bad position, and our own processing can be wrong. Anything derived rather than reported — bus punctuality, estimated bus arrival times, the interpolated position of a train between calling points — is labelled as an estimate where it appears.
For anything that matters, the station's own screens and the operator's own app are the sources of record. If you spot something on this site that is plainly wrong, please tell us — feed-level bugs are usually reproducible and usually fixable.