How to Audit a Public Space at Night: Lighting, Mobility and Safety

The short answer

A night audit is a structured walkthrough between roughly 6pm and 6am that reads a public space the way users experience it, not the way its design drawings show it. Assess lighting as perceived effect rather than fixture counts, check whether transport stops and the last mile to homes are lit and overlooked, and map who is actually using the space hour by hour before you invest in fixes.

Key takeaways

Perceptions collapse between day and night: what feels like a friendly park at noon can read as a dead, unsafe corner after dark, so conclusions from daytime inspections rarely transfer
Lighting must be assessed as perceived effect, not fixture count — brighter is not automatically safer, and vertical light on faces at eye level matters more than horizontal road luminance for reassurance
Safety has two layers — actual risk and felt fear — and both respond to eyes on the space, active frontages, sightlines, maintenance and programme, not only to cameras
Transport safety does not end at the stop: the last mile between station, bus stop or parking and the front door is where fear curtails mobility for many users
Use patterns shift across the night in predictable bands, so an audit must be repeated across evening, late night and early morning to avoid one-time snapshot bias
Local governments can translate findings into concrete, fundable interventions such as relighting stops, replacing underpasses with street-level crossings or adding request stops
No single blueprint fixes night safety — evidence gathering, local listening and budget follow-up matter as much as any single light or crossing

Why the night walkthrough is a different discipline

A public space is not the same object at 2pm and at 11pm. Research and practice consistently show that the transition from daylight to darkness sharply increases fear of crime and changes which routes people choose, how long they stay and whether they leave the house at all. What reads as a lively square by day can become a dead, intimidating pocket after dark, often precisely because it was designed and inspected only in daylight.

A night audit exists to close that gap. It treats the space as a sequence of hours, users and journeys rather than as a single static site. It also widens the definition of a good outcome: a space is not merely lit or safe on paper, but feels usable to the people — especially women, older adults, night workers and younger users — who have to rely on it after sunset. UN Women's fifteen years of Safe Cities work across dozens of cities makes the same point: a safe city is measured by small daily decisions, such as which route feels safest after dark, not by the presence of infrastructure alone.

Daylight inspections miss the dominant sensory conditions of night use: darkness, silence, empty frontages and longer reaction distances
Fear changes behaviour measurably: in UK polling, roughly 9 in 10 women reported feeling unsafe walking after dark and many said they alter routes in darker months
Good night design combines lighting, sightlines, activity, transport and maintenance — auditing all of them together avoids isolated fixes

Timing and method: running the audit itself

A credible night audit is repeated, not a single excursion. Split the window from early evening to early morning into bands — roughly 18:00–21:00, 21:00–00:00 and 00:00–06:00 — because each band has different users, different lighting regimes and different risks. Run the same route on a weekday and on a weekend, and again at the start and near the end of the dark season so you capture seasonal darkness rather than only summer nights.

Send small mixed groups with at least one facilitator rather than a single inspector, because individual perception is personal and a group surfaces more diverse observations. Plan the route as a set of journeys that mirror real use: arriving at a transit stop, walking the last mile home, crossing a park, passing a dark underpass. Record observations at fixed points so that repeat passes are comparable, and interview or shadow actual users where possible instead of relying only on your own impressions.

Repeat the walk across evening, late night and early morning bands, and again across weekdays, weekends and the dark season
Walk journeys, not just locations: stop-to-door, crossing, park route and waiting point
Use small teams with a facilitator and fix recording points so passes can be compared
Pair physical observation with user interviews or walkshops to capture lived experience

Lighting: read the effect, not the fixture

Auditors too often count poles and confirm a compliance level, then declare the space lit. That misses the point. People navigate by vertical, semi-cylindrical light — light that falls on faces and objects at about eye height — rather than by the horizontal luminance measured for traffic. A path can meet a road-lighting norm and still leave faces in shadow, which is precisely what erodes reassurance. In Russian and international practice alike, standards such as the GOST 55706 series describe this distinction, using semi-cylindrical illuminance measured around 1.5 metres up as the criterion for telling whether people can recognise one another.

Judge light by its effect along the whole journey: even distribution with no dark pools between pools of light, no glare that blinds an approaching person, warm enough colour to read a face naturally, and light that reaches the space users occupy rather than only the carriageway. Keep vegetation and street furniture out of the light path. Just as important, resist the instinct to simply add more: brighter light does not automatically mean safer and can create glare, hard shadows and light pollution. The Arup night-time vulnerability assessment is explicit that lighting that merely exceeds minimum compliance, or simply gets brighter, often fails to address how people actually feel.

Check vertical and semi-cylindrical light at around 1.5 metres, not just road luminance, because face recognition drives reassurance
Look for uniformity and glare: dark gaps, blinding poles and hard shadows are failures even when averages look acceptable
Verify that light reaches paths, stops and waiting areas users occupy, and that vegetation and furniture do not cast hiding shadows
Treat added brightness as a trade-off with glare, ecology and light pollution — more lux is not an automatic win

Mobility and transport: the last mile after dark

Fear does not stop at the platform. For many users the vulnerable stretch is the journey between a station, bus stop or parked car and the front door, or the wait at an unlit stop. A night audit must therefore cover transit as a connected sequence: the illuminated stop, the visible and overlooked route away from it, crossings and refuges, and the final approach to housing.

Audit whether stops have their own lighting and seating, whether waiting areas are visible from surrounding windows and passing traffic, and whether the walking route is genuinely overlooked rather than merely present on a map. Cities are also testing demand-responsive fixes for the last mile: in Vigo, Spain, night-time bus request stops let passengers alight closer to home after dark, shortening the feared stretch. Elsewhere, replacing underpasses with street-level crossings has removed the most feared single point on a route. Assess step-free access, tactile cues and crossing refuge as mobility issues in their own right — accessibility and fear compound after dark.

Check transit stops, shelters and waiting areas for their own lighting, visibility and seating
Walk the last mile from each stop and crossing to nearby housing and record where light and eyes-on-the-street drop off
Consider route-level fixes such as night request stops or closing underpasses in favour of street-level crossings
Include step-free, tactile and crossing-refuge checks, since physical access fears multiply at night

Safety: perceived and actual, and the eyes on the space

Treat perceived safety and actual safety as two linked but distinct measures. Actual risk is about incident data and known hazards; perceived safety is about whether a person feels able to use the space, which for many women and marginalised users is the binding constraint. Both respond to the built environment, and good audits assess them separately rather than assuming one follows the other.

The recurring principle is surveillance by design: active frontages and windows that look onto a route, a mix of uses that keeps activity flowing into the evening, routes that follow roads busy and overlooked by other people, and maintenance that removes the signals of neglect — broken lamps, litter, overgrown planting that creates hiding places. UN Women's evidence underlines that infrastructure alone is not enough and that behaviour, local voices and coordinated services matter. Audits should therefore record the 'eyes on the space' at each band: how many frontages are lit and active, whether anyone can see a user from a window or passing vehicle, and where blind spots, dead-ends and recessed entrances concentrate fear.

Score perceived safety and actual incident risk separately and combine both in your findings
Record active frontages, windows and passing traffic that keep a route overlooked at each time band
Flag recesses, dense planting, dead-ends and underpasses as sites where sightlines collapse
Note maintenance as a safety signal — a clean, repaired, well-kept space changes how safe people feel
Temper environmental fixes with evidence from local voices and behaviour, not assumptions

Reading use scenarios through the night

Different hours have different publics, and the space must serve each. Early evening brings commuters, leisure and families; the late band brings night workers, students and the hospitality economy; the small hours serve those whose work genuinely runs overnight. Design each band for its users rather than assuming one configuration covers everything. Mixed and active uses — shops, cafés, gyms, services that stay open — extend natural surveillance and give people a reason to be present.

Map actual activity by hour with counts and observations: who is present, which routes they take, where they pause and where they avoid. This reveals the gap between the planned use of the space and its real night-time life, and it points to cheap activation fixes — a temporary market, a lighting upgrade around a specific corner, a programme that keeps a frontage alive — that change perception faster than expensive rebuilds.

Segment users by time band: commuters and leisure, night economy and workers, and around-the-clock essential staff
Count who is present, which paths they take and where they avoid to expose planned-versus-real use
Use mixed and active frontages as surveillance infrastructure that extends into the evening
Prioritise low-cost activation and relighting where programme, not geometry, is the blocker

Turning findings into interventions and repeat measurement

Close the audit with a shortlist that ranks interventions by leverage, cost and the users most affected. A few targeted changes often outperform broad ones: relighting one feared corner, relocating a bench, trimming vegetation that destroys a sightline, replacing a single underpass with a street-level crossing, or adding a request stop. Budget and political follow-through matter — in England, central guidance and a multi-year funding pot explicitly steer local authorities toward street safety improvements such as better lighting and overlooked walking routes.

Every finding should carry an owner, a cost band and a verification step. Then re-run the audit on the same route, in the same time bands, after the change to prove whether perception actually shifted. Because fear is slow to respond, plan measurement across seasons rather than expecting an instant result. Treat the audit as an ongoing practice, not a one-off report, and keep local users in the loop so that the space is adjusted to their lives rather than the other way round.

Rank interventions by leverage, cost and the demographic most affected rather than by convenience
Assign each action an owner, a cost band and a measurable outcome
Re-audit the same route and time bands after change, and measure across seasons since fear responds slowly
Keep users engaged so fixes respond to lived experience rather than assumptions

Put it into practice

Night Audit Field Scorecard

A reusable field sheet for a route-based night audit. For each fixed observation point, walk the route once per time band and score every line 0 (fail) to 3 (strong), noting evidence in one line per item. Sum by dimension to spot where the weakest links are.

Uniform light along the full route with no dark pools between fixtures, at band and seasonal settings
Faces readable at ~1.5 metres on paths and waiting areas (vertical and semi-cylindrical light), not only road luminance
No blinding glare or harsh shadows from any fixture along the line of travel
Transit stops and shelters have their own light, seating, timetable clarity and visibility from surrounding uses
Waiting areas and the last mile from stop to housing are overlooked by lit windows or passing traffic
Step-free, tactile and crossing-refuge continuity holds along the whole route in the dark
Active frontages or mixed uses keep at least one band of the route overlooked in the small hours
Sightlines clear: no overgrown vegetation, recessed corners, dead-ends or unlit underpasses creating hiding places
Maintenance evident: lamps working, no litter or vandalism signals that read as neglect
Real users present and counted per band, and at least one user journey verified on foot
Intervention shortlist written with owner, cost band and a re-audit date for each fix

Questions people ask

Why is horizontal road illuminance not enough to judge whether a space feels safe at night?

Road-lighting norms usually express horizontal luminance or illuminance on the carriageway, which serves drivers. Pedestrians, however, navigate by light falling on vertical surfaces and faces. Semi-cylindrical or vertical illuminance at around 1.5 metres determines whether two approaching people can recognise each other's faces, and that recognition is what drives reassurance. A route can fully meet a horizontal road standard while still leaving faces in shadow, which is why night audits should check vertical light at eye height rather than relying on compliance figures alone.

What are the typical time bands to audit and why should I repeat the walk?

Split the night roughly into 18:00–21:00, 21:00–00:00 and 00:00–06:00. Each band has different users, lighting regimes and risk profiles: commuters and leisure, then the night economy and workers, then around-the-clock essential users. Repeat each route on a weekday and weekend and again across the dark season, because seasonal darkness, weekend crowds and the small hours change perceptions dramatically. A single walk is a snapshot and will mislead you about a space that shifts hour by hour.

If an area is already 'bright', why might it still feel unsafe?

Brightness is not the same as reassurance. Excess light can produce glare that blinds approaching people, harsh shadows that create hiding places, and pools of dark between bright spots. What matters is even, well-directed light that illuminates faces and the space users occupy, together with the feeling that someone could see you — lit windows, active frontages, passing traffic. Expert night assessments note that simply adding brighter light often fails to improve perceived safety and can harm ecology and the night environment, so audits should judge light by effect rather than lumen count.

What are the highest-leverage transport fixes for the last mile after dark?

Start where fear concentrates: the unlit bus stop, the underpass, the dead-end and the route from stop to front door. Cost-effective fixes include dedicated lighting and seating at stops, making waiting areas visible from surrounding windows and traffic, and shortening the feared stretch through demand-responsive options such as night request stops that let passengers alight closer to home. Replacing underpasses with street-level crossings removes the single most feared point on many routes. Each should be verified by re-auditing the same route at the same hour after the change.

How do I measure perceived safety rather than only relying on incident statistics?

Use observation plus lived experience rather than crime data alone. Have small mixed groups walk the route and score perceived safety at fixed points in each time band; combine this with counts of who is present and where they avoid, and interview or walk with actual users — especially women, older adults and night workers. Research and municipal practice emphasise that felt fear, not only recorded incidents, drives whether people use a space, so capture both and treat them as separate evidence streams.

Sources and further reading

Sources were checked when this page was generated. Confirm changing dates, rules and prices with the original publisher.

Nationwide plans announced to design safer streets as 9 in 10 women report feeling unsafe walking at nightGOV.UK / Active Travel England
How to make a city safer for women and girlsUN Women
Night-time Vulnerability AssessmentArup
Designing for the night: why 12-hour cities can't sustain a 24-hour economyArup
ГОСТ Р 55706-2023. Освещение наружное утилитарное. Классификация и нормыФГБУ «Институт стандартизации» (текст документа на meganorm.ru)
Requirements of megapolities to city lightingQPRO / Construction News Agency