‘Helpful Quality Measures’ on infrastructure for cycling

This article presents a set of ‘measures’ – really just a small set of five very simple questions – which can be used to quickly provide a top level assessment of the quality of an environment for cycling. These measures have already been tested in a number of scenarios, and I’m happy that they are useful, as are many of those who have used them – therefore the main aim of this article is to set them free into the wider world.

(Skip ahead to the measures)


I’ve spent time (previously) in a professional role where I was often asked about the quality of cycle routes. It turned out that these questions were very difficult to answer – even when I knew a route well. To put it simply my ‘good’ might be ‘bad’ to someone else (or the other way around).

Then later, in conversation about auditing routes, the same question re-emerged, but from another angle. What should a surveyor record in order that someone sat in an office, with only maps and data to go on, could answer these kinds of questions?

I know that many readers of this blog will at this point be wanting to tell me about a tool they know of, which tries to measure quality more objectively.  I know many of these tools – and there are a large number of them. My contention is this: many of these tools are useful, but they generally do not get around the subjective nature of these judgements about quality – they just hide it away. Look carefully, and it’s still there.

This is no different to discussions about the quality of hotels, or books, or films. What one person likes another may dislike. Quality can be discussed, and we recognise it, but it cannot easily ‘measured’ in an objective way. Of course measurements of some features can be taken, but this approach can end up obscuring an understanding of overall quality with details about things only of interest to specialists.

Quality in this respect is ‘subjective’ not ‘objective’. Once we accept this it allows us to embrace the subjectivity – to look for a way to work with that subjectivity rather than trying to hide it. I think the results of doing this are very helpful. People might be afraid that this would lead to ‘poor’ being classed as ‘good’, but I think there are several ways we can avoid that.

So I set out to look for a method to work with the subjective nature of the judgements that would be required, but without losing a clear definition of what ‘good’ looks like. So we’re not talking about an unfocused 5 point scale or similar here…

There is already one fairly well established idea in the UK, providing a rough guide to the level of traffic safety available on a street. That idea is that ‘safe’ in broad terms can be judged as being the level at which a street is suitable for cycling by an unaccompanied twelve year old child. This is sometimes misunderstood to be about designing for children, but (to my understanding) the idea has always been that the level of safety needed for an unaccompanied twelve year old is also representative of the level of safety needed for a whole range of other bicycle users.

This idea is useful – but of course the quality of cycling experience relies on much more than traffic related safety. Surface quality also matters (not least in the UK where people sometimes think that cycling has been adequately provided for by putting signs on unsurfaced and indirect paths).

However there are also other very important factors. The Helpful Quality Measures include important questions about ‘social safety’ and ‘flow’. These are in recognition that everyday cycling can’t be hidden away in places where people are afraid to go at night, and that a certain basic level of speed and comfort must be facilitated.

But even this isn’t the whole story. It is very important to note that these Helpful Quality Measures are only intended for helping with judgements about those elements of quality which are inherently subjective. There are other elements of quality which can be objectively measured – and which must also be taken into account in a proper assessment. Directness, gradient, appropriate prioritisation of different users, and accessibility (around disability) are all at least as important as the factors assessed using the Helpful Quality Measures. Of course there may be no harm in also creating simple questions to help with assessing these if that approach fits a particular situation.

I’m indebted to a small group of staff from the organisation Sustrans Scotland who, mostly outside their normal roles, helped me with the very earliest discussions about this issue, the development of the measures, their initial testing for auditing purposes, then the further refining of the wording used.

The remainder of this article presents and explains the measures (and is written to be used without this introduction, so briefly repeats a few of the points already made above):

The Helpful Quality Measures

The ‘Helpful Quality Measures’ are the five simple questions below:

Looking only at traffic-related safety, would most people allow an unaccompanied 12 year old to cycle here?
Looking only at surface quality, would a road (racing) bike be used here?
Looking only at issues of social safety, would most people feel comfortable walking here after dark?
Looking at ‘flow’, can a relaxed 8mph (12kph) be continually and safely maintained here?
Looking only at route signage, can this cycle route be followed confidently without a map?
[This measure is ONLY FOR SIGNED ROUTES]

Each question can be answered with one of the following ratings:

Probably not

Yes+‘ may also be helpful as a rating for surface quality


How the measures work

People don’t have any issue forming opinions about the quality of infrastructure they encounter when cycling, but what makes it difficult to measure quality is that personal opinions vary, and people disagree about what ‘good’ looks like. These measures make it easier to communicate about a relatively well defined level of quality.

The measures aren’t complicated, and are designed so that they can be understood and used by ordinary members of the public, but it is helpful to understand the following points if you’re going to work with them.

The measures are specifically about cycling.

These measures are only about the quality of the infrastructure in regard to cycling – not about other issues like the quality of place, or about how well designed a street is for those on foot or using a wheelchair (etc). Cities and towns which are genuinely good for cycling are also designed to be good places to live. These measures tackle the one important issue only, with an approach that works specifically for this issue. The measures should be used alongside consideration of the wider issues.

The measures are about ALL infrastructure which might reasonably be used for cycling for everyday journeys.

These measures are specifically designed to be as applicable to ordinary roads as to anything built specifically to support cycling. Only the measure about signage has a more limited usage (applying specifically to signed routes).

The measures might not be applicable at all in regard to other types of cycling. For example they are clearly not of any use in assessing mountain bike routes.

A complete assessment of infrastructure for cycling also must consider other factors.

These measures, on their own, will never completely define the difference between ‘good’ and ‘poor’ infrastructure – and they aren’t intended to do so. Many other factors have to be taken into account.

I’ve created these measures to fill a gap in the tools and measures which currently exist. They don’t replace these. They address qualitative/subjective factors which aren’t easily measured by existing tools/methods.

Among many important issues which are not addressed by these measures are directness, gradient, appropriate prioritisation of different users, and accessibility for a variety of bicycles (including those designed for use by people with disabilities).

Each measure is a simple, memorable, question.

I think that the simplicity and memorability of the measures is important. I think that these questions can be easily understood both by ordinary members of the public and by specialists.

The questions are carefully thought out so that, while remaining simple and memorable, they are focused on specific quality factors. The approach provides much more useful information than simpler ratings would.

The ratings which can be given are a set of memorable words or phrases.

The quality scale provided by the ratings is very simple and is memorable. This makes the measures easier to use than if they used a numerical scale, and this makes them more practical for the purposes I’ve designed them for.

In this article I’ve explained the measures in detail, but in practice it ought to be enough to remember that the questions can be answered with ‘yes’, ‘perhaps’, ‘probably not’, and ‘no’.

The rating ‘yes’ is the minimum standard, indicating basic, decent, usable infrastructure, and relatively civilised streets which can be used for everyday cycling.

The measures are phrased so that even the rating ‘perhaps’ indicates a failure to provide decent infrastructure which will adequately support everyday cycling. ‘Yes’ is the minimum standard for ‘adequate’. People may learn to cycle safely in hostile conditions, where their needs are disregarded, and may keep cycling even when this is an unpleasant experience. Most people who currently cycle do so simply because it’s a good way to get around. But the measures lay out the conditions which we’re going to need to create if we’re to make cycling just a transport choice – rather than a test of character.

You can use these measures however you like. They are useful in discussing or planning new infrastructure, informal assessments of existing roads, formal and informal assessments and audits by groups of people (including the public), and for communicating with the public (perhaps with mapping).

I’d strongly discourage attempts to formalise the measures or to make them less subjective, or to change the wording – this has been carefully thought out and tested.

You are encouraged to use the measures for any purpose you desire, but if writing about them, using them in presentations, or for paid work please provide a link to this article and an acknowledgement of my work.

For measuring quality in more detailed ways look to tools like the London Cycling Campaign ‘Cycling Level of Service’ tool (or a local equivalent) which is designed for this purpose.

Appendix 1 provides additional notes about the practical usage of the measures for auditing purposes.

Appendix 2 answers some additional ‘frequently asked questions’ about the structure and wording of the measures for people interested in how they have been developed.

Details about individual measures

This section provides more detail about each individual measure. When working with the measures most people only need to read this section once, or only need to refer to it to answer specific questions as they occur.

Measure 1: Traffic related safety

Looking only at traffic-related safety, would most people allow an unaccompanied 12 year old to cycle here?

The potential ratings, ‘yes’, ‘perhaps’, ‘probably not’, and ‘no’, are simply answers to the question.

Key information

This measure is specifically about traffic-related safety.

If somewhere is (or will be) safe from traffic it should be given a positive rating even if you wouldn’t allow an unaccompanied 12 year old to cycle there for other reasons.

Frequently asked questions about this measure:

Q. I don’t know any 12 year old children. How do I know whether people would allow a 12 year old to cycle somewhere?

A. In reality 12 year old children vary hugely in their skills. The idea of an unaccompanied 12 year old conveys a very rough idea of a level of safety. I consider that a 12 year old child will have the physical skills to be in complete control of a bicycle, able to cycle predictably and according to any way they have been taught. I also consider a 12 year old as a child, with a child’s understanding of risk and a lower ability to predict and avoid this. I consider that ‘most people’ would be much more protective of a 12 year old child than of an adult.

Q. Does ‘most people’ mean everyone, or only those already cycling?

A. ‘Most people’ means everyone, not just people who currently cycle.

Q. Two of us disagree about whether we’d let a 12 year old cycle in a particular place. How do we decide whether to record ‘yes’ or ‘perhaps’.

A. That’s the nature of these measures. Talk about it. Use the discussion to learn from each other about why you disagree. Ask other people for their opinions. Record ratings from several people and work out what the median is (the rating that half the people think is too low and half think too high) but also record that people couldn’t agree because that’s useful to know too.

Q. 12 year old children can be very experienced on a bike. Am I judging on the basis of a trained and experienced 12 year old, or one lacking experience?

A. Judge on the basis of a typical 12 year old for the area/place you’re rating. The fact that it would be possible to train a 12 year old to be safe, despite poor traffic safety, should not be taken into account. In countries with good quality infrastructure much younger children can safely get around unaccompanied and with minimal or no training.

Q. It’s possible for 12 year old to cycle safely here if they use the pavement, and lots of people don’t seem to mind if this happens. Should we rate the route according to its legal use or on the basis of how people actually use it?

A. We’re looking for a rating for ‘proper’ use of the infrastructure – as it is designed to be used by people following any rules laws and signs. If people don’t follow the rules because of feeling unsafe then you almost certainly need to be recording the lowest level of rating for safety here.

Q. The level of traffic-related safety on this road varies during the day. During commuting hours it’s poor, but on weekends it’s pretty good. How should we record the level of safety?

A. This isn’t a scientific measure. Do what seems sensible. Is the route used primarily during commuting hours? If so then record what it’s like at this time. Is it used primarily for leisure at weekends? If so record what it’s like at the weekend. Or give it two ratings. You’re only using these measures to communicate about the route.


Measure 2: Surface quality

Looking only at surface quality, would a road (racing) bike be used here?

The potential ratings ‘yes’, ‘perhaps’, ‘probably not’ and ‘no’ are simply answers to the question.

It may also be useful to allow ‘yes+’ to indicate the very highest quality surface:  motorway standard tarmac/bitmac, which is very smooth, with no cracks or potholes at all. This rating is helpful because it indicates a surface which is so good that it can also be used for small wheeled devices like skates.

The rating ‘yes’ would tend to be used for normal UK roads which are in relatively good condition – covering a wide range of conditions.

Key information

This measure is specifically about surface quality.

Although the question asks about ‘road’ bicycles the measure is not used to judge how fast it is possible to cycle. Issues around speed are dealt with in the ‘flow’ measure – which penalises designs which don’t adequately account for the wellbeing of those on foot (etc).

The measure asks about ‘road bikes’ because these aren’t normally used on rougher surfaces. Skilled bike users can do amazing things but the measure asks if a road bike would be a sensible choice on a particular surface.

If a surface is (or will be) smooth then rate it highly for this measure even if nobody would actually use a road bike there for other reasons.

Frequently asked questions about this measure:

Q. The quality of this path surface varies during the year. It’s really good in dry weather but unusable in the wet. How should we record the surface quality?

A. This isn’t a scientific measure. Do what works. Perhaps record two values, for summer and winter, or consider whether the path is normally wet or dry. Simplest may be to record it at its (normal) worst.

Q. What exactly is a ‘road bike’ and how do I make judgements if I’ve never ridden one?

A. ‘Road bike’ is the proper title for what some people call a ‘racing bike’. It’s what someone in the Tour de France rides and a simple Google image search will provide examples. The measure asks about road bikes because they have very skinny tyres, which are pumped up to be very hard, and are therefore normally only used on relatively smooth sealed surfaces. Remember that this isn’t a scientific measure so a guess about what it is like to ride such a bike is still useful.

Q. Why are there 5 possible rating values for this measure (‘yes+’ is allowed) when there are only 4 for the other measures?

A. The measure is intended to record a very wide range of surfaces, and four levels weren’t enough. The top level (‘yes+’) is for recording sections of tarmac which aren’t just good enough for use on a road bike but also good for other uses – for example on a small wheeled scooter, with less robust wheelchairs, or on skates. Such surfaces also make for comfortable and efficient/relaxed cycling on any normal city bike too.


Measure 3: Social safety

Looking only at issues of social safety, would most people feel comfortable walking here after dark?

The ratings, ‘yes’, ‘perhaps’, ‘probably not’, and ‘no’ are simply answers to the question.

Key information

This measure is intended to record issues of social safety.

If a place feels safe in social terms you should give it a high score even if there are other reasons you might not walk or cycle there.

Although this measure asks about walking it’s designed to record information for people cycling. The measure asks about walking because people are generally more cautious when walking than when cycling, making the measure is more sensitive.

Frequently asked questions about this measure:

Q. This measure asks about walking after dark. Should I record a rating on the basis of what the area is like in the middle of the night, at rush hour during the winter, or at another time?

A. Do what’s sensible. This isn’t a scientific measure but just an attempt to capture something about the character of an area. It may be helpful to record two or more ratings for defined situations. One simple system can be to record one measure for busy times (e.g. winter rush-hour) and one for quiet times (e.g. winter late evening). It’s unlikely to be useful to provide a rating for use in the middle of the night. If you want to record only one measure judge according to conditions during a late winter evening – i.e. after dark, after rush-hour, but not in the middle of the night.

Q. Why does the question refer to ‘walking’ when these measures are about cycling?

A. The measure asks about walking after dark because that makes it more sensitive. In areas where social safety is perceived to be lower people are less willing to walk than to cycle. It’s also clear that some people are used to cycling in darkness in environments which others find intimidating. We want this measure to support judgements about everyday cycling.

Q. Why does the question ask about ‘most people’?

A. The question asks people to ignore whether they personally would walk in an environment after dark, instead judging whether it would be a normal thing to do. This also makes the question more objective.


Measure 4: Flow

Can a relaxed 8mph be continually and safely maintained here?

(use 12kph instead of 8mph in this question if you wish)

This question can sometimes be a little more difficult to answer, so the following additional explanations may help.

The rating ‘yes’ indicates that it is never normally necessary to drop below 8 mph – little concentration is required to cycle above 8 mph – cycling above 8mph is safe and will not alarm anyone on foot (or using wheelchairs, etc).

The rating ‘perhaps’ indicates that it isn’t normally necessary to drop below 8 mph – but that if this is necessary then it is only slightly below 8mph and it isn’t for long – or that cycling above 8mph might perhaps risk alarming people on foot (but might not) – and might (or might not) require some increased concentration.

The rating ‘probably not’ indicates that it is probably not possible to remain above 8mph – or that at some times this is impossible – or that it is not a relaxed experience doing so (it requires concentration) – or that it is likely to cause risk (or alarm).

The rating ‘no’ simply records that it is not possible to remain above 8mph, or that doing so causes alarm or risk, or requires significant concentration or skill.

Key information

This measure is about quality of design, and usability.

Good infrastructure is easy to use, meaning that cycling can be an efficient and relaxed experience, and that at the same time the wellbeing of others – such as those on foot or using wheelchairs or similar – is fully taken into account. This measure is designed to take each of these issues into account. At first sight it may seem slightly more complex, but it’s been found to be fairly simple to apply in practice.

Note that a full picture of ‘flow’ also requires that length/chance of delay is recorded for locations where it may be necessary to stop cycling. This can be measured objectively (so details are not discussed as part of these helpful quality measures).

Frequently asked questions about this measure:

Q. How fast is 8mph/12kph? I don’t have a speedometer on my bike.

A. This is about the maximum speed that most healthy adults would reach running for a bus, and only fit adults running on a flat straight path would maintain this speed for any significant time. It’s a speed which on a bike probably won’t feel any more energetic than walking does. While this measure doesn’t ask for precision it can nevertheless be helpful to use a speedometer to check what 8mph/12kph feels like before trying to use this measure (even using a speedometer smartphone app, provided the phone is suitably mounted for safety of course).

Q. Can you come up with a better word than ‘flow’? Couldn’t you measure it differently? Shouldn’t you just talk about average speed?

A. This isn’t really a measure about speed at all, but of the character of infrastructure and of how speed relates to the wellbeing of others. I’m very happy to consider alternative terms, but so far this has been the simplest, and in testing it proved useful. This measure penalises designs which make cycling difficult, demand silly manoeuvres, or mix pedestrians and cycling on narrow paths.

Q. Often people can maintain speed by ignoring signs and instructions or by bending the rules. Should this rating of ‘flow’ be based on speeds in reality, or speeds if you stick strictly to all the rules?

A. This is a measure of infrastructure quality, not of actual speed. It should be judged on the basis of someone following all the rules, including any instruction given on signs and similar.

Q. The usability of a path will vary during the day. At times it will be full of people walking. Should I judge it’s ‘flow’ at busy commuting times or at quiet times?

A. This isn’t a scientific measure – do what’s sensible. I’ve written the measure to take account of this to some extent. The ratings ‘perhaps’ and ‘probably not’ can be seen to refer to changing conditions during the day. Remember that 8mph is a very relaxed speed on a bicycle and in cities which are good for cycling the answer to this question will be ‘yes’ in most places, even at rush hour.

Q. 12 kph isn’t quite the same as 8mph!

A. True. But it’s easier to remember rounded numbers and this isn’t a scientific measure, and memorability is more important than accuracy.

Q. When judging ‘flow’ on a hill, should I take account of the fact that it’s faster going down than up? Or should I disregard the hill in some way?

A. I think that the easiest way to deal with a hill, assuming that infrastructure is similar for uphill and downhill travellers, is to rate it for downhill use. This measure of ‘flow’ is intended to record design not geography. Gradient is important, but it can be measured separately should you wish – as an objective fact providing additional information about quality.


Measure 5: Route signage quality

Looking only at route signage, can this cycle route be followed confidently without a map?

This measure is only to be used for signed cycle routes. The following additional explanations may help to make it as useful as possible.

The rating ‘yes’ indicates that the route is obvious, and that users will be confident of being on the route.

The rating ‘perhaps’ indicates that it is necessary to look for signs carefully but they are present and a user will probably be confident they are on the route.

The rating ‘probably not’ indicates that most people will have some difficulty following the route and will often be unsure whether they are on the route or not.

The rating ‘no’ simply indicates that the route cannot be followed without a map.

Key information

This measure is specifically designed for cycle route signage.

In most locations (away from signed cycle routes) this measure can be ignored.

This measure is not useful for assessing other signage, even though this is often much more important in general wayfinding, and it is not useful as a wider measure of how easy it is to navigate.

Frequently asked questions about this measure:

Q. What counts as a ‘route’?

A. The word ‘route’ can be confusing, having many related meanings. Here I specifically mean a route for cycling which has been given a name, identifying number or other identifying code which (it is intended) means it can be followed using signage. This kind of signage is what is used for routes such as those on the UK National Cycle Network. I’m not meaning ‘routes’ used for cycling in the wider sense.

Q. What signs should be taken into account?

A. This is specifically about signs which use some kind of number, code, or route name identifying the route, to tell people how to follow it. It may be possible to navigate following other clues, including general road signage, but this shouldn’t be taken into account. The measure is specifically about the quality of dedicated route signage.

Q. Why does the question ask about route users being confident about being on the route?

A. The question is asking not just about whether following signs means people stay on the correct route, but also what it feels like to follow the signs. Good route signage doesn’t just point the way at junctions, but also ensures that route users remain confident that they are following the route correctly in-between junctions. There are route signs designed specifically for this purpose. This also tends to result in a robust route, which isn’t badly disrupted by vandalism, damage, or the loss of a few signs. A messier question might add the words  “and without feeling the need to check a map”.


Appendix 1:
Using the measures for auditing

For scenarios where the helpful quality measures are used for auditing/evaluation of existing infrastructure the following notes may be useful. I’ve included suggestions for understanding how to process auditing/evaluation ratings.

Using multiple auditors improves the value of the information.

Clearly these measures are subjective. A key learning point in testing was that while one rating from one auditor can provide useful information, more helpful information arises from ratings being independently provided by several different auditors.

To put it another way; the more people rate something the better. In testing we found that good infrastructure is given a consistently high rating, and very poor infrastructure is consistently given a poor rating. For situations in-between the ratings can vary a lot from auditor to auditor. This is useful in itself. The fact that ratings vary a lot in these situations gives important information (mainly indicating that the quality is too low to be considered satisfactory, but that it is not really bad).

Numerical scores can be used for processing, but shouldn’t be used for assessment or presentation.

For those who are involved in processing ratings from different auditors – probably to create an overall rating – numerical values can be assigned to the quality scale after an audit has taken place. However, even if results are processed in this way then any final presentation of results should rely on the words and ideas. The measures were not created in order to score quality numerically, and attempts to use them in this way seem likely to provide misleading results.

In testing we had success by converting auditor ratings on a scale from 1 (poor) to 4/5 (good) and by calculating median values from this for a particular section – then indicating the result with a suitable colour scale related to the original ratings.

Despite its statistical limits as a measure of variance we found standard distribution can be used to work out roughly where people agreed and where they didn’t. We think that it may be useful to present this information as part of results – once again not trying to assign a score, but just a general indication that people did or did not agree over their ratings.

Auditors should learn the questions and rating values (answers).

The individual measures are very simple, but it is important that auditors understand how the measures are interrelated, and how they each focus on something very specific.

In particular auditors should understand how not to confuse the measures – for example when encountering a path which is completely separated from traffic, but which passes through an area which feels unsafe for other reasons, they should be clear that it deserves a good rating for traffic related safety – therefore answering the question about whether “most people would allow an unaccompanied 12 year old to cycle here” with ‘yes’ (despite the fact that people would not actually allow this).

Different lengths of road/path can be audited.

There is no correct way in which to divide up a road or path or route into sections for auditing.

There are many possible approaches. For example it could be useful to come up with a rating, based on these measures, for a whole route. At the same time it could be useful to allow auditors to come up with ratings which vary even every few metres.

Different approaches result in different challenges for anyone processing audit results.

Preparation can be helpful.

The measures have been designed to make them easy to use, even by ordinary members of the public, and they are inherently subjective. However, a small amount of preparation may increase consistency between auditors.

In particular:

  • If an auditor has never ridden a road (racing) bike they may find it helpful to speak to someone who has, to try one out, or even simply to look at pictures or videos of them in use.
  • If an auditor has not been a parent of school age children they you may find it helpful to speak to someone who is before trying to judge how ‘most people’ think about a child’s safety on the road.
  • It may be helpful for an auditor to do some work to establish what 8mph (12kph) feels like on a bicycle before trying to provide a rating for ‘flow’.

Audit signage separately.

In general it seems easiest to work with the measure about route signage separately from the other measures.

This measure is about something quite different to the others, and it requires observation of very different features.

You can code answers for quick recording

In testing these measures many of those involved found that it was helpful to have a very simple way to write their observations quickly. The following codes are a very obvious way to do this.

‘y’ stands for ‘yes’
‘per’ for ‘perhaps’
‘pn’ for ‘probably not’
‘n’ for ‘no’

These codes work for all the measures. Remember that the surface quality measure also has a ‘yes+’ rating (for extra-smooth surfaces).

The list below illustrates how these letters can be combined with a letter for each question:

Looking only at traffic-related safety, would most people allow an unaccompanied
12 year old to cycle here?
12y | 12per | 12pn | 12n

Looking only at surface quality, would a road (racing) bike be used here?
Ry+ | Ry | Rper | Rpn | Rn

Looking only at issues of social safety, would most people feel comfortable walking here after dark?
Dy | Dper | Dpn | Dn

Flow: Can a relaxed 8mph be continually and safely maintained here?
Fy | Fper | Fpn | Fn

Looking only at route signage, can this cycle route be followed confidently without a map? [ONLY FOR SIGNED ROUTES]
Sy | Sper | Spn | Sn


Appendix 2:
FAQ about wording

As the helpful quality measures first were used and tested those involved or who were informed about the work inevitably asked questions about their wording. These notes explain why the wording is as it is. I’m quite happy to hear ideas for better wording – but please read this section before getting in touch. The current wording results from some careful thought and development, and real-world testing.

Q. Couldn’t you make the questions simpler?

A. They could be made simpler, but I think that this would make the measures less useful. I worked hard, with those who helped me to create them, to make these questions as simple as possible – while maintaining a balance between this simplicity and their technical focus. Despite their subjective nature the measures are designed to collectively capture and communicate some very specific information. We tested the measures (and others did in several different situations) and generally they seem to work well.

Q. Lots of people haven’t ridden a ‘road’ bike or don’t know what this means. Would it be simpler to refer to a ‘city’ bike, a ‘hybrid’ or ‘a normal bike’?

A. See the point above. Most bicycles can be used in most circumstances and we want an idea of surface quality which is as clear as possible. Road bikes are much more sensitive to surface quality than other bicycles.

Crucially, I also consider that the design of a road bike is well known – races like the Tour de France are watched by many people who don’t cycle, and even those who don’t cycle can envisage that narrow wheels/tyres set limits. Actually it is difficult to find any other bicycle ‘type’ which is so commonly understood or which has been so stable over time.

Some people are worried that this question makes it sound like all cycle tracks are being designed for people to travel fast, even if this disadvantages others. This is a risk, but I believe that most people will recognise that these are technical measures and that the ‘flow’ measure specifically penalises designs which disadvantage others.

Q. Lots of really enjoyable leisure cycling is on more isolated paths and tracks which aren’t really designed to be used after dark. The question about social safety would rate these poorly. Shouldn’t it be re-written to account for that?

A. I’m enthusiastic about good quality infrastructure which is designed for leisure cycling, but I’m also keen to see bicycles used as everyday transport, including after dark. This measure helps to clearly distinguish infrastructure which is well designed for leisure purposes – from infrastructure which is well designed to support everyday bicycle usage for transport. Remember that ‘after dark’ includes, in the UK, everyday trips around working hours. Rather than re-writing the question, it might be easier to ignore it for leisure cycling routes (or to accept a lower rating, like ‘perhaps’ as the minimum level of quality).


Acknowledgements: I’m indebted to a small group of staff members from Sustrans Scotland for their support in helping me take these these measures from an initial set of conversations though to testing and refining – and for trying out their use more widely. Particular mention goes to Simon Philips (for the very earliest discussions about whether a set of measures could be created at all and about what they might measure), and to Chiquita Elvin (for suggesting the ‘after dark’ measure).

See also…


  • Scroll below for comments. As always, one of the aims of publishing here is to seek comment, ideas and criticism. Please let me know what you think.
    If you landed here from a Twitter link on a mobile device you may need to press ‘Leave a comment’ below to see the comments on this article or you can reload the page to see them.


  1. Well thought through. Rather than specify a type of bicycle it is possible to describe features, for instance the Australian guides talk about a reference bicycle without suspension and tyres no wider than 32mm (rather generous!).

    From a mapping and higher level overview what is lacking are plots of turning and reference vehicle dimensions. They are easy to overlay onto a drawing or map and verify that road vehicles will fit within the space, but I have not seen similar for bicycle access. A collection of them would also highlight the other wheeled vehicles such as tricycles, recumbents, trailers etc which would not be familiar to designers or auditors as “bicycles”.


    1. Thanks Meltblog. I thought about more detailed descriptions, but there was a balance to meet – and I wanted questions that people who don’t currently cycle could understand and relate to – so they need to be simple. There is probably a role for more specific detailed measures too (for different purposes).


  2. Eh, being Dutch there are just two types of surfaces that should be mentioned, rideable on not rideable. Surface condition is the main consideration. A took both my car and my road cycle with me and I was appalled by your tarmac. The disrepair of major roads is, eh, beyond any standard.


    1. We have a very strange system I think. In the Netherlands you plan for when a road surface will be replaced, then allow any work on services (pipes and electricity) which are under the ground at the same time. Here we spend money on laying a permanent surface, but we allow any company that needs to access the services to dig up the road and then mend it. Nobody does work to the services when the surface is being replaced. Nobody seems to have realised that it would be a good idea to do both things at the same time. A Dutch engineer told me that the main difference is that when a road is built in the Netherlands money is spent to make sure it’s done properly – so it lasts. But here we just keep repairing things, never actually re-building the street properly.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s