Keep on cycling: The network effect

This article was published in 2018, in Magazine 141.

Seville didn’t build bike lanes. Seville built a cycling network. It was one project.

Manuel Calvo, cycle network designer, speaking at the UCI Bike City Forum in Milan

The value of the telephone network increases with each line because each new line can call all the other numbers and all the other numbers can call it

Image as described adjacent

The value of a network is a complex calculation. To understand this let us consider three different types of network: the telephone network, the road network, and a cycleway network.

The value of a telephone network is not that you have a connection to it. The value is derived from the very many other people and businesses who are connected to it. If you have the one and only telephone then you have nobody to call, and nobody can call you, and therefore the telephone itself is valueless. The first telephone call was made between two phones, from Alexander Graham Bell to Thomas Watson. As more phones are added to the phone network the value of that network increases.

The value of the telephone network is in proportion to the number of lines. Each line costs about the same to install. The value of that network increases because that new line can call all the other numbers, and all the other numbers can now call this new line. This is a non-linear relationship.

The road network has a similar value. Before road maintenance was largely centralised, each local parish had to maintain its own roads, and they just cared about the roads to their fields and perhaps the local market town, and pretty much ignored most other roads. There was little value in increasing the quality of a road to a nearby village because people rarely went in that direction. The value in visiting the next village could never justify the cost of maintenance.

To be a cycling city and not have the assurance that you can cycle to other places is the equivalent of having that single telephone line with nobody to call

Once the bicycle was invented, and then became popular, this road network became more valuable as people could travel further in less time. This was further expanded once motor vehicles started using the roads, and their maintenance became more difficult as these vehicles became heavier and travelled faster.

Many people assume today that you can drive anywhere. If you wanted to drive to Stiklestad in northern Norway, it might take a couple of days but the road network would allow you do it. This is the value of the road network. Most of the stuff you buy has been moved along this road network at some point.

The cycle network is a different story. Can you even cycle safely to the Harwich Ferry? Are there good cycleways all the way? To be a cycling city and not have the assurance that you can cycle to other places is the equivalent of having that single telephone line with nobody to call.

Within Cambridge there is a mostly reasonable secondary network, with the highlights being places like Riverside and a few strategic links like The Tins that provide a good network, if you know where to go. Cambridge has a very poor primary network. In South Cambridgeshire the primary network is mostly non-existent.

To understand this, we will first explain the difference between a secondary network and a primary network. Then we will explain why there are different approaches to network expansion in Cambridge and South Cambridgeshire.

Secondary network

A secondary cycle network is a set of links that provide connectivity between neighbourhoods. In a cycling country, the secondary network would connect the primary network to the places you actually want to reach. Typically, secondary routes do not take the most direct route, but instead take the easy way around the problems.

The secondary network is the first network that any place has to install to start expanding the number of people who feel safe enough to cycle. Easy wins can include closing a few strategic streets to motor traffic, as was done on St Andrews Road at Elizabeth Way, or Hooper Street between Kingston Street and Sturton Street. Other simple wins include providing crossings over main roads, such as the Abbey Street crossing of Newmarket Road, or the Chesterton Hall Crescent to Ascham Road crossing of Milton Road.

It is then easy to draw a few lines on the map in the hope that people can navigate through the city along streets they are unfamiliar with, and through areas they don’t know. This does nothing to show people who don’t currently cycle how popular moving around on a cycle is. When you hide the people cycling away from the people stuck in the traffic jam, you don’t attract people out of their cars.

Primary network

In a cycling country, the secondary network would connect the primary network to the places you actually want to reach

Easy wins can include closing a strategic street to motor traffic as was done on Hooper Street between Kingston Street and Sturton Street.
Image as described adjacent

The primary cycle network creates cycle routes that are attractive, direct, and efficient. When we say efficient, we don’t mean routes that encourage people to cycle furiously at high speed, we mean routes that don’t require you to stop at every side road to give way to others, or stop at every junction and press a button and wait several minutes to get a few seconds to cross the road.

The directness of the primary network is the key to making it work. When people ask where should the primary cycle network go, we could just point to a map of Cambridge and say ‘the red roads are where the primary cycle network should go’. These main roads are already as straight as we can get.

If we had the primary cycle network along the main roads then many people stuck in traffic jams would see thousands of people pass them everyday on cycles. Eventually they might even think that they could try it one day a week, or two days a week, or perhaps for a full week when it is sunny and dry.

For the benefits of the primary network to be maximised we need to rebuild these routes to a very high standard. Most of the main roads are wide enough to build high-quality cycleways, with segregated footpaths, motor traffic lanes, and still have plenty of space for avenues of trees and bus-stop islands.

Where two primary routes meet, the junction must be of the best possible design

The really key part of this is at the junctions. Where two primary routes meet, the junction must be of the best possible design. It should prioritise people first, people walking, people cycling, and people on buses. Only then should the needs of those in private motor vehicles be considered, using any space left. This means that a junction’s capacity for cars might slightly decrease in the short term, but the people capacity of a junction would significantly increase, and we need to move people around the city, not just cars.

Network effect

Unfortunately, we have not reached consensus for building a primary cycle network, and there appears to be little political leadership to drive such a network design.

Whilst there have been some recent improvements to roads like Hills Road and Huntingdon Road, they haven’t really tackled the junction issue.

Getting the consensus required for a primary cycle network is not something that will happen given the secondary network we have today. In places that didn’t have a secondary network and jumped directly to a primary network, like London or Seville, the growth in cycling has been dramatic.

We could radically transform the ability for everybody to get around the city if we built a primary cycle network with people-friendly junctions.

Cambridge ridership

While there have been some recent improvements to roads like Hills Road, they haven’t tackled the junction issue.
Image as described adjacent

So what’s needed in Cambridge is to campaign for better cycle infrastructure along the main roads, consisting of segregated and protected cycleways. A primary cycle network is not just some painted lines at the edge of the carriageway. We need to build that primary cycle network, which will encourage more people onto cycles, and show the people stuck in cars that there is a viable alternative that people just like them are using right next to them today.

This is about building a ridership-based network. We need to plan this network for the next 10,000 or 20,000 people cycling. It is about creating an environment where people feel safe and can get where they are going by following routes they are already familiar with, and without having to worry about crossing junctions in multiple little stages.

South Cambridgeshire coverage

The approach needed for South Cambridgeshire is different. Outside Cambridge we first need to build a network. There are many places in South Cambridgeshire that have no cycleways. The Wilbrahams, for example, are as close as Bottisham to Cambridge yet the only way to get there is to cycle around blind corners on a 60mph road.

It is shocking that Cambourne was allowed to be built without a cycleway connecting it to Cambridge. Remember that Cambourne is closer than Over and Melbourn, yet Over and Melbourn are both now connected with cycleways to Cambridge.

The quality of the existing network also sometimes leaves something to be desired. The cycleway to Waterbeach is a 0.9 metre-wide strip of tarmac immediately next to a 50mph main road. Alternatively, you could cycle along the isolated towpath next to the river trying to avoid the puddles. Neither could be described as attractive.

The target for South Cambridgeshire should be to connect every village with every neighbouring village, and every village to the nearest strategic Greenway into Cambridge. This means that even though Harlton is a tiny little village and would on its own not justify a cycleway, it still needs to be connected by a cycleway to Haslingfield and the Eversdens. The value is in the network, and people in the Eversdens may one day be able to cycle safely to Haslingfield and beyond.

Ring roads

If the value of a network is in how easy it is to access something else on that network, then one of the key aspects of network design is to reduce congestion. If we continue building more cycleways and the Greenway network, and if these are all very successful, then we will reach a point where the cycle network itself is congested. Before this happens, we need to think about the big and bold solutions to this.

The solution is not that difficult, and has been created in many locations many times over. From the Beijing rings to the M25, from the Boulevard Périphérique in Paris to the I95 and I495 in Boston, MA, there are plenty of examples of car congestion being managed by creating orbital routes around cities and whole regions.

There are three obvious cycle rings that are necessary in the Greater Cambridge area: the inner ring, the middle ring, and the outer ring.

If you want to make people change their habits, you’ve got to give them a viable alternative.

Chris Boardman, champion of Greater Manchester’s Beelines network

Inner ring

Map of Cambridge overlaid with proposed primary network and concentric ring roads.
Image as described adjacent

The inner ring would provide a cycle route around the city centre so that people who are cycling from one side to the other don’t have to go through the city centre, especially during peak tourist season. This has the advantage of decreasing the number of people who need to cycle through the centre, whilst increasing the network effect of all the radial routes that are being built, or should be built, from the city centre outwards. This route would go from Queen’s Road, via Northampton Street to Chesterton Road, over to East Road via Elizabeth Way bridge, and then via Gonville Place and Lensfield Road to the Fen Causeway.

All this is possible given enough political leadership and sufficient funds

Middle ring

The middle ring would provide a cycle route around the city of Cambridge and would connect up so many places that are expanding or providing significant employment. The route would go from West Cambridge via Eddington, through what should become Darwin Green, to the south edge of Orchard Park (where a cycleway already exists), down to Cambridge North station, over the new Chisholm Bridge and then out via Wadloes Road to Barnwell Road, and via Brooks Road, Perne Road, Mowbray Road to Addenbrooke’s and the Biomedical Campus.

Outer ring

The outer ring would connect up all the new development that has leapfrogged over the greenbelt. This would link places like Cambourne and Bourn Airfield with Northstowe and Waterbeach New Town. This ring would typically be about 12km out from the centre of Cambridge. Again the value of this route is not just in the connections between these places, but if we have a full network each additional connected place amplifies the value of the whole network.


For Cambridge city, we need to create the primary cycle network. This will require redesigning the main roads to incorporate cycle infrastructure and providing safe and efficient junctions. These will originate from the cycle inner ring around the historic city centre, out to the middle ring that connects the main areas of development in the city. The secondary network would be maintained, but not significantly expanded.

For South Cambridgeshire, we need first to create a network of routes that can be cycled safely. Some will be radial routes towards Cambridge, while others would form orbital connections from these radial routes to nearby villages. Each of the villages should eventually connect with its nearest neighbours. An outer ring would tie all of this together, and allow families to go exploring the beautiful countryside we live in.

All this is possible, given enough political leadership and sufficient funds. We need to continue to campaign for the highest quality infrastructure, the safest junction designs, and the most connected network of cycle routes possible. We need to explain that the benefit of connecting a few hundred people in a village to the cycle network is not just that those people can go somewhere else, but that everybody else can now visit them and their pubs and shops too.

Robin Heydon