At the time of Legible London’s inception in 2010, Slate magazine highly commended the new signage, but stated that “comprehensive systems are more commonly found in smaller cities”, calling the original goal to complete city-wide for the London 2012 Olympic Games “with budget shortages and a time crunch … extremely tough to meet”.

This post will examine how T-Kartor managed to roll out the Legible London basemap cost-effectively for the whole of Greater London in time for the Olympics and has created over 10,000 map based products during the system’s first ten years.

By the end of 2009 Legible London pilots consisting of 60-80 signs had been created for the South Bank and Bankside, Covent Garden; Bloomsbury Clear Zone and Richmond & Twickenham and were undergoing evaluation. The signs proved very popular with the public and it was decided to continue the roll-out of Legible London wayfinding.


Illustration 1: the three original pilot areas for Legible London 

The goal was to cover the Cycle Hire area: 400 docking stations spread over an area of 66 km2, for its planned launch in June 2010. Two years later, the complete wayfinding system would need to cover the whole of Greater London, an area of 1,600 km2, for the London 2012 Olympics.

Budget calculations based on the pilot methodology and an area the size of the Congestion Charge Zone (below) were seen as extremely prohibitive, without even considering that the whole of Greater London would cover 72 times this area (albeit less densely detailed).

T-Kartor were able to offer cost-efficient development methods combined with a mix of economies of scale, design pragmatism and lessons learned from the pilot to bring costs within a reasonable level. In fact the city wide basemap was created for less than a third of the original quote for the Congestion Charging Zone.


Illustration 2: The Congestion Charge Zone (22 km2), Cycle Hire (66 km2) area and Greater London (1,600 km2)

The problem was solved by creating a seamless GIS database instead of individual map files. This had the advantages that we were able to start with data which already existed, editing rather than creating it from scratch. In addition a system could be offered with the ability to create any map, rotated, overlapping, varying in size and format, virtually “on-the-fly” and highly automated.

A GIS database vs a cartographic product

In its data form, a GIS database consists of separate sources of lines, shapes and points, with geographic coordinates placing them in their correct position. This data has no colour, font, text size or scale. A map, or cartographic product, is created by selecting the data, colouring it according to a design specification, and ordering it in layers so that a realistic, legible interpretation is created.


Illustration 3: GIS data has no styles applied, no fonts and no scale. The application of these graphic styles creates a map product

The use of existing data meant that rather than manually drawing map objects and typing road and object names (both very labour intensive), the work involved selecting and validating existing objects, then enhancing their graphic appearance (tidying) as styles are added at the appropriate scale. In this way, large areas can be processed at the same time, layer by layer.

It should be noted that no dataset is suitable for use without a large amount of work, as it will have been created for different purposes than pedestrian wayfinding. City owned cadastral mapping, for example, was created as a record of the extent, value and ownership of land. It includes highly accurate building shapes (useful for our building layers) and details of land use, but may be less complete for depicting footpaths through public parks. For wayfinding purposes accuracy in footpaths is more important than exact building edges, and this is one of the layers which requires extra attention, complementary data / reference material, and graphic enhancement.

Content selection and field surveys

Besides graphic legibility in land use, footpaths and other topographic features, a pedestrian wayfinding basemap needs to be selective with how much information it portrays. A hierarchy of information was developed to allow landmark buildings to stand out and large area names to be visible but not distracting.

City municipal GIS datasets include a lot of information superfluous for wayfinding. A considerable effort is required to manage content, layer by layer, including carrying out field surveys to check accuracy. Although time consuming and a significant proportion of total cost, this step is essential to guarantee the accuracy of the highest priority layers, such as landmark buildings (names, entrances and positions) and public transport infrastructure.


Illustration 4: This field survey plot shows all available data, with useful data for validation highlighted for clarity in the field

The final cartographic database is an accurate, consistent, legible basemap (which is constantly maintained) served from a mapping platform which enables automated, cost effective outputs of thousands of rotated, overlapping map products. Each output comes from fully maintained, updated data and follows TfL’s rigorous content and design guidelines.

Illustration 5: This film shows the complexity of the complete database, being drawn “on-the-fly” by the system software

Systems for smaller cities

The sheer scale of London, New York and Toronto can lead one to believe that a data-driven system is only for large cities where economies of scale are greatest. However, its benefits can be enjoyed at a much smaller scale.

Birmingham is the UK’s second city with a population of over a million and the main metropolitan area is 260 km2. By 2013 the city had just completed installation of a new wayfinding system, comprising mapping information totems for pedestrians and bus stop totems for public transport users.

Birmingham authorities had a larger vision however; a city of interconnected journeys served by a comprehensive information system. This system would serve different transport modes and local contexts via various communication channels from a core data engine.

In order to realise the Interconnect West Midlands vision, T-Kartor created a cartographic GIS database from the existing wayfinding basemap, originally a hand-drawn Adobe Illustrator file. In this way the area of coverage was doubled and the number of map products increased from around 70 to over 200.

At the time, the full future scope of the Interconnect wayfinding system was not known, (it was initiated in stages based on funding: see link). However, as a data-driven core system, T-Kartor were able to add new products, deliver via new channels and connect nearby small towns as part of an incremental, staged growth. Importantly, a maintenance system was established including an online content management portal in time for the installation of the West Midlands Metro tram network. This enabled a controlled update of all affected on-street signage as the tram system reached completion.

There is a threshold for a data-driven wayfinding system to enjoy economies of scale, but it is not concerned with the size or population of a city. What matters is the ability to expand with new information types, added local areas, innovative distribution channels and above all, to create a maintenance routine for the master basemap and all information assets.



Follow our coverage of the world of City Wayfinding...