GTFS Routes, Trips & Stops Explained

Routes, trips, stops, and shapes form the geographic backbone of a GTFS feed. They describe where transit vehicles go, what path they follow, and where passengers can board. This guide explains each file in detail and shows how they connect to form a complete picture of a transit network.

routes.txt: defining transit lines

The routes.txt file defines the transit lines that an agency operates. Each route represents a group of trips that are displayed to riders as a single service, like "Bus 38" or "Red Line."

Key fields

Route types

The route_type field classifies each route by vehicle mode. The most common values are:

The extended GTFS route types specification includes more granular values (100s for rail subtypes, 200s for coach subtypes, etc.), but the basic values above cover most feeds.

The GTFS Viewer includes a Routes table where you can browse every route with its short name, long name, type, and agency, all sortable and searchable.

Example

route_id,agency_id,route_short_name,route_long_name,route_type,route_color
R001,sfmta,38,Geary,3,FF6600
R002,sfmta,N,Judah,0,1B9AD1
R003,sfmta,F,Market & Wharves,0,F0C832

trips.txt: individual journeys

While routes define transit lines, trips.txt defines individual journeys along those lines. A single route typically has dozens or hundreds of trips per day, one for each departure time in each direction.

Key fields

The service_id field is crucial: it links each trip to the calendar files that define which days it operates. The shape_id links to shapes.txt to define the geographic path on the map.

Example

trip_id,route_id,service_id,trip_headsign,direction_id,shape_id
T38_001,R001,weekday,VA Medical Center,0,shape_38_out
T38_002,R001,weekday,Transbay Terminal,1,shape_38_in
T38_003,R001,weekend,VA Medical Center,0,shape_38_out

Notice how trips T38_001 and T38_002 share the same route (R001 = Bus 38) and service (weekday), but go in opposite directions with different headsigns and shapes.

stops.txt: where passengers board

The stops.txt file lists every physical location where passengers can board or exit a transit vehicle. Stops are shared across routes: a busy downtown stop might be served by a dozen different bus lines.

Key fields

Stop hierarchies

For complex stations (like subway systems), GTFS supports a hierarchy: a station (location_type=1) contains multiple platforms (location_type=0) and entrances (location_type=2). The parent_station field links child stops to their parent station.

Example

stop_id,stop_name,stop_lat,stop_lon,location_type,parent_station
STN001,Powell Street Station,37.7844,-122.4080,1,
STN001_P1,Powell Street - Westbound,37.7844,-122.4081,0,STN001
STN001_P2,Powell Street - Eastbound,37.7843,-122.4079,0,STN001
S200,Geary & 25th Ave,37.7812,-122.4844,0,

Powell Street Station is a parent station with two platforms. Geary & 25th Ave is a simple bus stop with no parent.

TransitLens includes a dedicated Stops table where you can sort and search through all stops in the feed, which is useful for large feeds with thousands of entries.

Connecting stops to trips

Stops connect to trips through stop_times.txt, which defines the sequence of stops each trip visits and the arrival/departure times at each one. This is the largest file in most GTFS feeds: a single trip visiting 30 stops generates 30 rows.

shapes.txt: geographic paths

While stops define where vehicles pause, shapes.txt defines the actual path vehicles follow between stops. Without shapes, a map can only show straight lines between stops. With shapes, you see the real road or track geometry.

Key fields

A shape is defined as a series of latitude/longitude points that, when connected in sequence, trace the vehicle's path. More points means more geographic detail: curves in the road, loops around blocks, and highway ramps are all captured through dense point sequences.

Example

shape_id,shape_pt_lat,shape_pt_lon,shape_pt_sequence
shape_38_out,37.7749,-122.4194,1
shape_38_out,37.7751,-122.4180,2
shape_38_out,37.7752,-122.4165,3
shape_38_out,37.7755,-122.4140,4
shape_38_out,37.7812,-122.4844,50

shapes.txt is optional, but most modern feeds include it. Without shapes, route visualization is limited to stop-to-stop straight lines, which gives a misleading picture of the actual network geometry.

How they all connect

The four files work together through ID references to describe the complete geographic structure of a transit network:

routes.txt → trips.txt → stop_times.txt → stops.txt
      join: route_id   join: trip_id   join: stop_id

trips.txt → shapes.txt
      join: shape_id

Reading this diagram from left to right:

1. A route has many trips (via route_id)
2. Each trip visits a sequence of stops (via stop_times.txt)
3. Each trip may follow a shape (via shape_id)
4. The stops define where vehicles pause; the shape defines the path between pauses

This relational structure means a single stop can appear in thousands of stop_times rows (served by many trips on many routes), and a single shape can be shared by multiple trips that follow the same path.

Why visualization tools matter

Looking at these CSV files in a spreadsheet, you see rows of IDs, coordinates, and times. But load the same data into a visual tool, and the transit network materializes:

• Routes appear as colored lines on the map, following their shapes through streets and neighborhoods. You can instantly see coverage, density, and overlap.
• Stops appear as points at their geographic locations. Clicking a stop reveals which routes serve it and how frequently vehicles arrive.
• Trips become visible journeys: you can trace a single vehicle's path from start to finish, seeing every stop it makes along the way.
• Shapes reveal actual paths: without them, you'd only see straight lines between stops. With shapes, you see the real road geometry, including one-way loops, highway routing, and detours.

TransitLens renders all of these relationships on an interactive map. Upload a GTFS feed and you can browse routes by name, click on stops to inspect them, and see how the entire network connects geographically. It's the fastest way to understand how routes, trips, stops, and shapes work together in practice.

Learn more about all exploration capabilities, including data tables and calendar analysis, on the GTFS Viewer page, or see the GTFS Map Viewer for spatial analysis workflows.