Routes to Nowhere: The Crumbling Geospatial Foundations Beneath Britain's Public Transport Network

On a given weekday morning, several million people in Britain consult a journey planning application before leaving home. They trust the results they receive — the departure times, the walking distances, the platform numbers, the assurances about step-free access — because they have no practical means of verifying them independently. What most of those users do not know, and what the transport authorities responsible for those applications rarely discuss openly, is the degree to which the underlying geographic data is unreliable.

This is not a fringe concern. The geospatial foundations of British public transport — the databases of stop locations, route geometries, interchange points, accessibility attributes, and real-time positioning — are maintained by a patchwork of organisations operating under different standards, with different update cycles, and with limited coordination between them. The result is a system that presents a confident digital face while resting on data that is, in many places, quietly broken.

The Bus Stop That Isn't There

Bus infrastructure represents perhaps the most acute example of geospatial degradation in British transport. The National Public Transport Access Node (NaPTAN) database, which holds the coordinates and attributes of every bus stop in Great Britain, was conceived as a definitive national register. In practice, it is a palimpsest of additions, amendments, and deletions accumulated over two decades, with quality varying enormously by local authority.

Stops that were removed years ago remain active in the database. New stops installed as part of regeneration schemes or route changes are absent. Coordinates that were captured from paper maps rather than GPS surveys place stops on the wrong side of a road, or in the middle of a junction. For routing algorithms, which treat NaPTAN entries as ground truth, these errors are invisible — but their consequences are experienced by passengers every day.

A traveller whose journey planner routes them to a stop that was relocated eighteen months ago faces, at best, a short walk and a moment of confusion. At worst — in an unfamiliar area, in poor weather, or for a passenger with limited mobility — the consequence is a missed connection and a journey that fails entirely. The error rate across NaPTAN is difficult to quantify precisely, but independent audits conducted by transport researchers have consistently found inaccuracy rates in the range of ten to fifteen per cent in some authority areas.

Rail Accessibility: The Data Behind the Promise

The rail network presents a different but equally serious set of geospatial challenges. Following significant investment in station accessibility works under successive Access for All programmes, many stations have been substantially upgraded — new lifts installed, platform gaps reduced, tactile paving extended. The physical reality of these improvements is, in many cases, genuinely encouraging.

The data reality is considerably less so. The accessibility attributes held within journey planning systems — step-free status, lift availability, assistance service coverage — are not always updated to reflect completed works. A station that received a new lift two years ago may still appear in passenger-facing applications as inaccessible, deterring disabled travellers from routes they could now use. Conversely, stations listed as accessible may have lifts that are subject to frequent maintenance closures, a dynamic that static database entries cannot capture.

Network Rail and the Train Operating Companies maintain their own accessibility records, and these do not always reconcile with the data held by third-party journey planners or the Transport Focus accessibility tools that passengers consult. The absence of a single authoritative source, updated in near-real time and openly licensed, means that accessibility information is effectively a matter of probability rather than fact — a situation that falls well short of the legal obligations that operators carry under the Equality Act 2010.

Multi-Modal Interchange: The Data That Does Not Exist

If individual mode datasets are problematic, the geospatial representation of interchange — the moments at which a journey transitions from one mode to another — is, in many respects, simply absent. The walk from a bus stop to a railway station entrance, the path from a tube platform to a cycle hire dock, the ramp from a ferry terminal to a connecting bus bay: these are the connective tissues of multi-modal journeys, and they exist in no unified spatial database.

The Transport for London network is a partial exception, having invested substantially in spatially accurate interchange data within its operational area. But TfL's remit ends at the boundaries of Greater London, and the majority of Britain's interchange points — including some of the busiest transport hubs outside the capital — are represented in journey planning systems by approximate coordinates and assumed walking times that bear only a loose relationship to reality.

For the Aviemore visitor trying to connect a Caledonian Sleeper arrival with a local bus service, or the Birmingham commuter navigating a rail replacement service, the gap between the journey planner's confident instructions and the physical environment they encounter can be disorienting and costly. These are not edge cases. They are routine experiences for millions of travellers using a network whose spatial complexity far exceeds the data infrastructure built to describe it.

The Real-Time Illusion

The proliferation of real-time departure boards and live tracking applications has created a widespread impression that public transport data is, by definition, current. In fact, the live information overlaid on these platforms sits atop a static geographic layer that may be significantly out of date. A bus whose real-time position is tracked accurately may still be displayed against a route geometry that was last surveyed years ago, stopping at locations that NaPTAN records incorrectly.

The distinction matters for routing. Journey planning algorithms use static geographic data to calculate the sequence of stops, the walking legs, and the interchange points that constitute a recommended route. Real-time data updates the timing layer but does not correct the spatial layer. An error in the underlying geography propagates through every journey plan that depends on it, regardless of how accurately the live vehicle position is reported.

A National Spatial Transport Standard

The technical components required to address these failures are well understood. A mandatory refresh cycle for NaPTAN, enforced through local transport authority funding conditions, would progressively eliminate the accumulated inaccuracies in bus stop data. A unified rail accessibility register, maintained by Network Rail and openly published, would provide a single source of truth for disabled passenger journey planning. A national interchange geometry standard, modelled on the spatial data work already undertaken by TfL, would give multi-modal routing the geographic foundation it currently lacks.

What these measures require is not technological innovation but institutional will: the recognition, at the level of the Department for Transport and the devolved transport authorities, that geospatial data quality is not a technical footnote but a precondition for the network's basic functionality.

Britain's transport ambitions — decarbonisation, modal shift, inclusive access — are all, in the end, spatial ambitions. They depend on people being able to get from one place to another, reliably and with confidence. Until the geographic data underpinning that movement is held to the standard the ambitions demand, the gap between the network on the screen and the network on the ground will continue to widen.