How Town Planning Works | TPW-0096 — The Prototype City: How Tactical Urbanism Uses Temporary Change to Test Permanent Planning Decisions

A permanent planning decision can cost millions, take years to deliver and still discover the most important mistake only after the concrete has cured.

A prototype asks a cheaper question first.

What if the new crossing were painted and protected for three months? What if three parking spaces became a tiny public space for one season? What if a neglected plaza received shade, seating, play and programming before the city commits to rebuilding it? What if the curb changed on Tuesday and planners spent Wednesday watching what people actually did?

This is the useful core of tactical urbanism. It is not merely colourful paint, planters and temporary furniture. It is a planning method for turning an argument about a possible place into an observable place that can be tested.

The idea remains very current. In 2026, UN-Habitat has been commissioning tactical and practical urbanism demonstration projects in Colombian cities as part of public-space programmes, while the United States Department of Transportation continues to recognise quick-build demonstration activities using temporary materials within road-safety planning. On 10 September 2026, C40 Cities and Greater Amman Municipality also unveiled a tactical pilot on Al-Taj Street designed to test heat protection and safer, more welcoming public space.

The interesting part is not that temporary projects are fashionable. It is that a city can use reversible change as a form of evidence.

The reader job: learn before the permanent build

This article has one planning job: explain how to use a temporary intervention as a disciplined prototype for a permanent decision.

That boundary matters because several neighbouring subjects already have owners in this series. The Public Realm explains why the space between buildings matters. The Listening Town owns public participation. The Learning Town owns the broader feedback logic of adaptive planning. The Safe System Map owns the transport-safety mechanism.

The Prototype City asks a narrower question: how do we turn a proposed spatial change into a bounded real-world trial, learn from it, and decide whether to keep, revise or remove it?

Temporary does not mean casual

A weak pilot is simply a cheap project.

A strong pilot has a hypothesis.

Perhaps the city believes that narrowing a turning radius will slow vehicles without preventing service access. Perhaps a parklet will reveal whether there is enough demand for public seating. Perhaps a shaded play zone will show whether families stay longer in a plaza. Perhaps a temporary loading arrangement will show whether deliveries can occur without blocking a pedestrian route.

The hypothesis should be written before installation. Otherwise every outcome can be described as success after the fact.

Temporary materials make the intervention reversible. They do not remove the need for design, safety, accessibility, authorization, monitoring or a decision rule.

Tactical urbanism can come from government, communities or both

The term covers several traditions.

Some interventions begin informally, when residents demonstrate a need by painting, planting, seating or temporarily occupying underused space. Others are formally authorised city pilots, designed and installed by agencies or partners. Still others are co-created: government provides permission and technical support while communities shape the use, programme and local details.

These models should not be confused.

An unsanctioned intervention can reveal a real problem but may not satisfy safety or liability requirements. A formal municipal pilot can satisfy technical standards yet fail if it merely consults residents after the design is already fixed. A good planning process understands where the idea came from and what legitimacy, evidence and technical review it still needs.

The prototype should test the thing that is actually uncertain

There is no value in prototyping what is already known while leaving the real uncertainty untouched.

If the dispute is whether emergency vehicles can pass, test the clear width and turning movement. If the dispute is whether people will use a plaza, measure occupancy at different times and weather conditions. If the dispute is whether a bus stop can coexist with a protected cycle route, test boarding behaviour and conflict points. If the dispute is whether businesses will lose customers, collect sales or footfall evidence where feasible instead of relying only on predictions.

A prototype is strongest when it isolates a decision variable.

Otherwise the city changes five things at once, observes an outcome and cannot tell which change caused it.

A prototype needs a baseline

“People seem happier” is not enough.

Before installation, record the condition that the pilot is meant to change. Depending on the project, that could include pedestrian counts, crossing distance, speed, near-miss observations, dwell time, seating occupancy, wheelchair access, delivery conflicts, temperature, shade coverage, noise, business footfall, parking turnover, public-space gender balance or the number of children using the site.

The baseline does not need to become a research thesis. It needs to be good enough to prevent memory from rewriting the old condition once everyone gets used to the new one.

Photographs from fixed viewpoints, timed observations and a few carefully selected measures can sometimes be more useful than a large dashboard nobody will revisit.

The pilot period has to match the behaviour being tested

A one-day event can test installation logistics and generate attention. It may be poor evidence for everyday behaviour.

A weekday morning does not represent a weekend evening. A school-term trial does not show holiday patterns. A dry-season plaza does not reveal drainage during heavy rain. A summer shade pilot may not reveal winter wind. A new layout often produces a short learning period while drivers, cyclists, businesses and pedestrians adjust.

The duration should therefore be chosen from the decision question, not from how long the temporary materials happen to last.

Where seasonality matters, the planning team may need more than one observation window or an explicit limit on what the first trial can prove.

Reversibility changes the politics of trying

Permanent proposals often force people to take a position on an imagined future.

A shop owner imagines lost access. A resident imagines noise. A cyclist imagines safety. A parent imagines a child crossing. Each is arguing from a model in their head.

A reversible pilot changes the conversation. The city can say: this arrangement will operate for eight weeks; these are the measures; these are the safety conditions; these are the circumstances under which it will be revised or removed.

That does not eliminate disagreement. It makes disagreement testable.

Reversibility is therefore not only a construction property. It is a governance property that lowers the cost of learning.

But a pilot should not become a way to avoid due process

Temporary interventions can create a temptation.

If a city calls something a pilot, can it bypass the processes that would apply to permanent change?

The answer depends on local law, but the general principle is clear: temporary status should not be used to evade necessary authority, accessibility, fire access, traffic control, procurement, heritage, environmental or public-notice requirements.

A prototype can simplify what is appropriately simplified. It cannot make legal duties disappear.

The city should publish what is temporary, what approval allows it, how long it will remain, who maintains it and how complaints or safety issues will be handled.

Accessibility must be designed into the temporary condition

Temporary projects often rely on objects that are easy to move: planters, barriers, cones, tables, ramps and modular platforms.

Those same objects can accidentally create narrow passages, abrupt level changes, confusing edges, blocked tactile routes or inaccessible seating.

The temporary condition still has users with wheelchairs, walking aids, prams, visual impairments, fatigue, limited balance and different sensory needs.

If the pilot excludes them, the city is not testing the future public space. It is testing a narrower population.

Accessibility is therefore part of experimental validity as well as basic inclusion.

Temporary materials have different failure modes from permanent materials

Paint fades. Lightweight barriers move. Planters die. Timber swells. Plastic delineators are struck. Temporary signs disappear. Movable furniture migrates. A platform that is stable on installation day can loosen after repeated use.

Therefore the maintenance plan is part of the prototype design.

Who inspects it? How often? Who replaces damaged parts? What happens after a storm? Who waters the planting? What condition triggers immediate closure? How is night visibility maintained?

A pilot that deteriorates into a hazard does not prove the permanent idea is bad. It proves temporary materials require their own operating discipline.

Quick-build is not the same as permanent-lite

Some pilots evolve into a useful middle category often called quick-build.

Quick-build interventions use relatively inexpensive and adjustable materials but are intended to operate for longer than a weekend demonstration. They can provide meaningful service while a permanent capital project is designed, funded or evaluated.

The United States Department of Transportation’s current Safe Streets and Roads for All examples describe quick-build demonstration activities using temporary barriers, delineators, paint, planters and other materials to test measures such as refuge islands, visibility improvements and traffic-calming arrangements.

The important distinction is that quick-build still needs a defined life cycle. Otherwise “temporary” infrastructure can remain for years without being upgraded, removed or properly maintained.

The prototype can test public-space demand before capital spending

Imagine a wide, windswept paved area beside a civic building.

A permanent redesign might propose trees, shade structures, play, seating, kiosks and event infrastructure. Before committing to the entire scheme, a pilot can test where people choose to sit, whether food brings activity, which corners attract children, whether evening use creates conflict with nearby homes, and how much maintenance the programming requires.

Temporary use can expose the difference between a beautiful plan and a place people actually inhabit.

It can also reveal latent demand. Sometimes a space appears empty not because nobody wants public life there, but because it offers no reason to stop.

Measure who uses the pilot, not only how many

A busy space can still be exclusive.

If a new plaza attracts adults but almost no children, the design may not be serving families. If seating is occupied almost entirely by customers of one café, the public character may be weaker than the raw occupancy count suggests. If women avoid the site after dark, daytime totals conceal a time-specific problem. If older people pass through but rarely sit, seating geometry or shade may be wrong.

Observation should therefore consider age, activity, time, group size and accessibility where it can be done ethically without intrusive identification.

The question is not “did people come?” but “did the place become more usable for the people the project was supposed to serve?”

A survey is useful, but behaviour is also evidence

Ask people what they think. Then watch what they do.

Someone may say they dislike losing parking but still visit the street more often after the change. A resident may support new seating in principle but avoid the exact seats because they are exposed to afternoon sun. A business may fear reduced loading access while delivery observations show that a timed bay works better than the previous informal arrangement.

Neither stated preference nor observed behaviour is automatically superior. They answer different questions.

A strong pilot uses both, and treats disagreement between them as something to investigate rather than something to hide.

The loudest feedback is not necessarily the representative feedback

Temporary changes can generate intense reactions because people encounter them immediately.

Email inboxes and public meetings can therefore overrepresent people with time, confidence and strong opinions. Passers-by, shift workers, teenagers, visitors, children, people with limited language proficiency and people who simply accept a better condition may be less likely to file comments.

The evaluation should combine complaint logs with intercept surveys, direct observation, stakeholder interviews and usage data where appropriate.

Participation is evidence, but it needs sampling awareness.

A prototype can expose hidden operational conflicts

Plans often look clean because operations are drawn as symbols.

Reality contains rubbish collection, school drop-off, taxis, deliveries, emergency access, street cleaning, construction vehicles, mobility devices, informal vendors, rainwater, queues and people who stop exactly where the drawing assumed they would keep moving.

A live prototype brings these conflicts into the same space.

This can be frustrating, but it is valuable. A temporary curb arrangement that repeatedly fails at 6 a.m. is telling the planning team something that a noon site visit never could.

Operational friction is not noise in the experiment. Often it is the experiment.

Do not test a permanent project with an unfairly weak prototype

A pilot can create a false negative.

Suppose the permanent proposal includes high-quality shade, drainage, accessible paving and comfortable seating, but the temporary version uses hot black asphalt, two plastic chairs and no shade. If people avoid it, the city has not fairly tested the permanent concept.

The prototype does not need to imitate final materials exactly. It does need to reproduce the functional feature being evaluated well enough that the result is meaningful.

Experimental fidelity matters. A bad model can discredit a good idea.

Do not use a successful pilot to overclaim

The opposite error also occurs.

A popular festival weekend does not prove a plaza will work every day. A three-week trial does not prove long-term maintenance. Temporary planters do not prove mature trees can be established in the same location. Volunteers staffing a project do not prove the city has an operating budget for permanent programming.

Every pilot has an evidence boundary.

The final report should state what the trial demonstrated, what it suggested and what it could not test. This protects the permanent decision from enthusiasm as much as from opposition.

The decision rule should exist before the results arrive

At the beginning, define the possible endings.

Keep: the pilot meets safety, access and performance thresholds and can advance toward permanent design.
Revise: the core concept works, but geometry, operations, materials or programme need another iteration.
Extend: evidence is promising but the trial period or seasonal coverage is insufficient.
Remove: the intervention creates unacceptable safety, access or operational problems that cannot be repaired within the trial.
Split: some components succeed while others should be abandoned.

A predeclared decision rule makes it harder for the city to move the goalposts after political pressure arrives.

Removal is not necessarily failure

A prototype that reveals a bad idea before permanent construction can be extremely successful.

The city has learned cheaply.

This is psychologically difficult because agencies and community partners can become attached to visible projects. Photographs have been published. Politicians have visited. Volunteers have contributed. Removal can feel embarrassing.

But the entire point of a prototype is that revision is allowed.

A planning culture that permits only successful pilots will eventually produce dishonest evaluation or timid experiments.

Temporary projects need an exit plan even when they succeed

If the pilot works, what happens next?

Permanent design may require capital funding, engineering, utility coordination, procurement and years of lead time. The temporary installation may not safely last that long.

The city should decide whether there will be a quick-build interim stage, whether temporary materials will be renewed, or whether the site returns to its previous condition until permanent construction begins.

Success without a transition plan can create the strange result of proving a better arrangement and then allowing it to decay.

UN-Habitat’s 2026 demonstrations show the method travelling across purposes

UN-Habitat’s 2026 calls for tactical and practical urbanism interventions in Armenia, Quindío and other Colombian contexts are useful because they show that the method is not limited to one transport treatment.

The projects combine public-space quality, accessibility, safety, environmental performance, social inclusion, youth participation, early-childhood needs and community ownership. They also require implementation planning, materials, coordination, participation and documentation for learning and replication.

That is a much richer model than “paint something brightly and see what happens.”

The temporary intervention becomes a compressed planning cycle: co-create, install, observe, document, learn and decide what should scale.

The current Amman pilot adds climate adaptation to the prototype logic

On 10 September 2026, Greater Amman Municipality and the Jameel C40 Urban Planning Climate Labs announced a tactical pilot on Al-Taj Street in Amman’s Al-Yarmouk District.

The project is particularly interesting because it connects temporary spatial change with heat adaptation and everyday public life. Instead of treating tactical urbanism only as a mobility technique, the pilot uses rapid, relatively low-cost intervention to test how a street can become cooler, safer and more welcoming for children, students and families.

The broader lesson is that the prototype method can be applied wherever the uncertainty is spatial and observable: shade, seating, play, loading, crossings, public space, temporary uses and more.

A prototype can become a participation object

Traditional consultation often asks people to react to drawings.

A live pilot gives them something they can walk through, sit in, cross, avoid, use with a wheelchair, visit with a child or operate a business beside.

This changes the quality of feedback. People can point to an exact corner and say the turning vehicle still feels too fast. They can show where the shade ends at 3 p.m. They can explain that the table blocks a route when the lunch queue forms. They can demonstrate that the temporary ramp works from one direction but not another.

The prototype makes abstract planning discussable at human scale.

Do not let affluent neighbourhoods monopolise experimentation

Small pilots are often easiest where residents have strong organisations, professional networks, grant-writing capacity and political access.

That can produce an experimental inequality: already advantaged districts get parklets, play streets, shade pilots and public-space improvements while places with weaker institutional capacity wait for conventional capital programmes.

A citywide programme should therefore allocate staff, technical assistance and small implementation budgets to communities that cannot organise the entire process themselves.

Tactical urbanism should lower the threshold for trying an idea, not shift public planning responsibility onto the communities best able to volunteer.

A practical prototype brief

Question: What exact uncertainty is the pilot testing?
Site: Why is this location suitable and what neighbouring uses can be affected?
Users: Who must be able to use the temporary condition safely and comfortably?
Baseline: What will be measured before installation?
Intervention: Which temporary elements reproduce the function of the proposed permanent change?
Authority: What permits, approvals, traffic orders or property permissions are required?
Accessibility: How are continuous routes, ramps, tactile information and seating needs handled?
Operations: How do deliveries, emergency access, cleaning, waste, maintenance and drainage work?
Duration: How long must the pilot operate to answer the question?
Evidence: Which counts, observations, surveys or sensor measures are proportionate?
Maintenance: Who inspects and repairs the temporary installation?
Decision rule: What results lead to keep, revise, extend or remove?
Transition: What happens between pilot completion and any permanent project?
Archive: How will lessons be recorded so another neighbourhood does not repeat the same experiment from zero?

A pilot should shorten the distance between planning and reality

The normal planning sequence can place years between a first idea and the moment ordinary people experience the result.

During those years, drawings become more detailed, reports become thicker and investment becomes harder to reverse. Yet some of the most important truths remain simple: can a wheelchair pass, do children use it, can deliveries work, does shade reach the seats, do drivers slow, does the shop still receive customers, does the space feel better at 9 p.m.?

A prototype brings those questions forward.

It does not replace engineering, planning law, design skill or public deliberation. It gives them a real condition to learn from.

The cheapest permanent mistake is often the one a temporary project discovers in time.

Build the learning loop, not just the installation

The visible part of tactical urbanism is the intervention.

The valuable part is the loop around it.

Define the uncertainty. Record the baseline. Build a faithful enough temporary condition. Let real users encounter it. Observe ordinary operations. Seek feedback beyond the loudest voices. Check accessibility and safety. Compare the result with the starting condition. State the evidence limits. Then keep, revise or remove the design.

That is how a city turns temporary material into durable knowledge.