Campaign: International Best Practice

This article was originally published in the December 2014 edition of London Cyclist

What can London learn about cycle infrastructure design from other cities? John Dales from Urban Movement hit the road to find out

This article is based on a study undertaken in 2013, on behalf of Transport for London, by a team led jointly by myself and Phil Jones (of Phil Jones Associates). The outcomes of the study have already been used to influence the new London Cycling Design Standards (LCDS) and our study report will be released once the final LCDS document is published.

The study cities were chosen to allow us to learn a variety of lessons: from what works best in cities where mass cycling is established to how cities lower down the curve have applied learning from those further up; from physical techniques to systems of governance; and from the different challenges and responses found in cities of different sizes and population densities.

We needed to get answers not just to the question ‘what have the best cities done?’, but also to the question ‘how can we do what they’ve done?’. And that’s why we didn’t only visit Dutch, Danish and Swedish exemplar cities but also others who are ahead of London in translating that best practice to their cities and streets.

Most study tours were of two day’s duration, during which we rode 40-50km on average and also interviewed local practitioners, as well as some campaigners. Wherever we went, the practitioners we met were keen to emphasise that we should focus less on specific street profiles or junction layouts, and more on working out how best to apply the design principles and techniques that lie behind those arrangements.




Amsterdam (Netherlands)

Berlin (Germany)

Brighton & Hove (UK)

Cambridge (UK)

Christchurch (New Zealand)

Copenhagen (Denmark)

Dublin (Ireland)

Malmo & Lund (Sweden)

Minneapolis (USA)

Munich (Germany)

Nantes (France)

New York (USA)

Seville (Spain)

Stockholm (Sweden)

Utrecht (Netherlands)

Washington DC (USA)



1.            There is strong, clear political and technical pro-cycling leadership at the city level.

2.            Cycling is considered an entirely legitimate and desirable everyday mode of transport.

3.            Growing cycling is part of an integrated approach to decreasing car mode share.

4.            Loss of traffic capacity or parking to create better cycling facilities, while often a considerable challenge, is not a veto on such action.

5.            There is dedicated, fit-for-purpose space for cycling — typically of one of these three types:

a)      Paths/tracks/lanes on busier streets which provide a degree of separation from motor vehicles that is appropriate to motor traffic flows/speeds and the demand for cycling.

b)     Quiet streets or ‘bicycle streets’ with 30kph/20mph or lower speed limits and often restrictions on motor vehicle access, particularly for through movements.

c)      Motor traffic-free cycleways or ‘greenways’ away from the main highway, but still well connected to the rest of the network at frequent intervals.

6.            Where the aim is to grow cycling rapidly, simple, cheap and effective means of securing space for cycling have been used as first steps.

7.            There is clarity about the overall cycling network (including planned future development), with connectedness, continuity, directness and legibility all being key attributes.

8.            There is clear, widely-accepted and routinely-used guidance on the design of cycling infrastructure.

9.            The frequency of occasions when cyclists need to give way or stop is minimised. This means that people cycling are able to make steady progress at a comfortable speed.

10.          The driving culture (and indeed city culture generally) is respectful of the needs of cyclists; and local traffic laws often play a part in this.

11.          Making better provision for cycling is understood as an ongoing challenge; with growth in cycling, and of city populations as a whole, requiring clear

forward planning.



Having finished the study tours, our task was to distil what’s needed to create comfortable cycling conditions in UK cities. We quickly agreed that the most important actions relate less to design techniques than to the political, funding, policy and legal environment.

If this is not conducive, then the simple fact is that opportunities to apply best infrastructure practice will be extremely limited. The accompanying panel (left) summarises the conditions that we found to be common in most cities with mature cycling cultures, with recent significant growth in cycling, or with a serious commitment to getting more people travelling by bike. Together, these conditions comprise what could be considered an ideal basis for growing cycling.



Where the above conditions are found, a widerange of cycle-friendly techniques have commonly been adopted. In our report, we describe 35 of these under five themes: Links, Junctions & Crossings, Network/Traffic Management, Interaction with Other Users, and the catch-all ‘Miscellaneous’. The report also assigns each technique a ‘degree of difficulty’ as regards their application in a UK context.

For most, there are no UK legal or regulatory constraints on their use; for others, there are some concerns over the current UK legal or regulatory position and/or their operational and safety record; and a few cannot currently be widely adopted in the UK due to legal, regulatory or other obstacles.

Since there’s no chance of doing justice to all of these techniques within this article — our report runs to 170 pages — we’ve chosen to highlight and illustrate five that we trust are of particular interest to Londoners as well as being, in our view, particularly important.




Enabling cyclists to feel safe while travelling alongside relatively fast and heavy flows of motor traffic is at the heart of Space for Cycling. We found numerous ways in which cycle tracks are separated from the main carriageway, from painted lines to vertical features that would damage any over-running motor vehicle.

While the former provide very little physical protection we nevertheless found the design of — and respect for — simple painted lanes in places like Berlin made cycling feel far more comfortable than in the UK.

Appropriately-wide tracks separated by kerb steps from the carriageway and footway to either side are very common in Dutch cities, Copenhagen and elsewhere, and are considered by many to be the standard of provision that the UK should aim for on busier streets. They not only provide a high degree of subjective safety, but also make for a simple, rational streetscape that pedestrians find easy to negotiate.

 Nevertheless, a kerb step can be over-run more or less as easily as other ‘low-profile’ separators — like the ‘armadillos’ which are common throughout Barcelona, along with other forms of intermittent feature (such armadillos are already used on Camden’s Royal College Street).



Outside the UK, it is very common practice to run signals according to a very simple two-stage pattern. So, at a typical crossroads, the north and south arms receive a green light, then the east and west arms; and repeat. Giving simultaneous green lights to all types of traffic (including pedestrians) on a given junction arm is an extremely efficient arrangement in terms of the amount of green time that all users get. But it relies on turning motor traffic giving way to pedestrians, and cycles crossing traffic — and that makes its application in the UK highly difficult.

To be clear, as things stand in this country, whenever pedestrians get a green light to cross, no other traffic gets a green light enabling it to go across the pedestrians’ path. To change this arrangement would require legislative change, a major shift in our street use culture, and the acceptance of losing a well-established pedestrian priority principle. To pursue such change would involve a very long process of decision-making and action at the national level. Bear in mind that when the Department for Transport’s new Traffic Signs, Regulations and General Directions (TSRGD) document is published in 2015, it will have taken five years of work to make a small number of relatively modest changes in how the UK’s streets are controlled. We will not be holding our breath while waiting for simultaneous greens to come along.

That said, trialling the use of flashing amber signals to warn drivers of the need, when turning, to give way to pedestrian and cycle traffic could be a first step in exploring the potential benefits of new forms of signal control.

Such flashing amber signals are used in Munich, Seville and Dublin. Although we didn’t see such an arrangement on our study, we would also like to see trials of simultaneous green signal stages for cycles on all arms of a junction, such as those found in Groningen in the Netherlands



Related to the above point is the specific problem at signalised junctions of the conflict between motor traffic making nearside turns (left in UK/Ireland, right elsewhere) and cyclists going straight ahead. In many study cities, this problem is mitigated by the fact that, by law, the turning traffic has to (and does) give way to cyclists on the nearside going ahead at the same time. This provides a good level of actual and subjective safety for cyclists, but the arrangement can cause delays to turning motor traffic. If this is a problem, the green time given to cyclists can be cut short, so as to give the turning traffic some green time of its own.

Another way of addressing the hooking problem while maintaining motor traffic capacity is to merge cycling and vehicle flows on the approach to a junction. The preferred arrangement in New York is to provide ‘mixing zones’ shared by cycles and turning motor traffic, with markings used to make drivers give way to cyclists when entering the zone. A similar but less formal arrangement is found at some junctions in Copenhagen.

In Munich, and increasingly Berlin, cyclists wanting to travel ahead are given a painted lane in which to do so (pictured above). Right-turning general traffic has to cross the painted cycle lane to access the mixed turning lane. Washington DC uses this arrangement, but increases protection for the ahead cyclists by using ‘wands’ to limit the length of the crossing zone for motor traffic.

While we found a very high degree of courtesy on the part of cyclists, with almost all adapting appropriately to the presence of people boarding and alighting, the concern often expressed in the UK is that unacceptable pedestrian-cycle conflicts will inevitably ensue with either arrangement.

The year-old ‘floating’ bus stops on the Lewes Road in Brighton indicate that this is by no means the case. While a study we have recently undertaken on Royal College Street in Camden (where the cycle track is also the bus boarding/ alighting area) also indicates that London cyclists can be trusted to ride responsibly.


In the Netherlands, and increasingly in Germany, some streets are designated as ‘bicycle streets’. Dutch Fietsstraten have no legal status, they are simply streets where bicycles are accorded priority over motor vehicles, and many are clearly signed as such.  Their success essentially depends on compliance by drivers — the vast majority of whom will be local residents.

German Fahrradstrassen have a clearer status in law (being designated by traffic regulation orders), but are essentially similar in character.

Key to the success of a bicycle street, is for traffic speeds to be a maximum of 30kph (just under 20mph), for traffic volumes to be low (less than 2,000 per day in the Netherlands; less than 3,000 per day in Germany) and ideally for cycle volumes to exceed those of general traffic. Measures associated with bicycle streets typically include filtered permeability (to exclude through traffic) and allowing two-way cycling on otherwise one-way streets.