How to Measure Urban Decay: A Practical Guide
I walked through a neighborhood in Malmö last March that the city classified as "stable." The data said fine. The street said something else entirely.
Forty percent of the streetlights were dead. The sidewalks looked like they'd been bombed by freeze-thaw artillery. Three buildings had boarded windows. A playground sat rusting behind a fence nobody had repaired. But the official neighborhood score? Stable. Because nobody had looked in two years.
Decay isn't a feeling. It's physical. You can photograph it, count it, score it. This is how.
What decay actually looks like on the ground
I've seen it everywhere. Cracks that start as hairlines and grow into canyons. Streetlights that die one by one until whole blocks go dark. Paint peeling off facades like sunburned skin. Graffiti that stays up for months because nobody cares enough to paint over it. Vegetation swallowing sidewalks. Litter accumulating in corners where the wind deposits it and nobody removes it.
These things cluster. One broken window is vandalism. A street of broken windows is a signal. The pattern matters more than any single defect.
The framework: categories, scales, coverage
We measure three things. What. How bad. How much of it.
Categories. Infrastructure: road surface, sidewalks, lighting, drainage, signage. Buildings: facade, windows, maintenance. Public space: cleanliness, vegetation, amenities. You can't manage what you don't name.
Scales. We use three buckets: good, degraded, failed. Good means no visible defects. Degraded means minor stuff — cracks, fading, small failures. Failed means major defects or complete non-function. You can get fancier with five-point scales, but three points gets you 80% of the value with 20% of the complexity. Start simple.
Coverage. Measure everything? Ideal. Impossible. We use stratified sampling — representative areas across neighborhood types. Quarterly for fast-changing stuff like roads and lighting. Annually for stable things like building facades.
Field collection — boots on the ground
The best data comes from people walking routes with eyes open and phones recording. Structured observation. Standardized criteria. Photos and GPS for every assessment.
Standardization isn't bureaucracy — it's survival. Two observers scoring the same asset should get the same result. If they don't, your data is noise. We train. We calibrate. We check. The app enforces structure, guides the sequence, validates inputs in real time.
Without this discipline, variation between observers swamps variation between neighborhoods. You end up measuring the measurers, not the streets.
From raw scores to indices that matter
Individual measurements are trivia. Aggregated into indices, they become intelligence. A neighborhood infrastructure index combining roads, sidewalks, lighting, and drainage tells you where the city is rotting fastest. A building condition index shows which blocks are sliding.
Indices answer the questions that matter. Which neighborhoods are improving? Which are declining? Where should the limited maintenance budget go? Without indices, these are political arguments. With indices, they're data conversations.
The Landvex City Health Index scales this to entire cities — infrastructure, buildings, public space, rolled into one score. Compare cities. Track trends. Target intervention where it matters.
Request a pilot →
The street doesn't lie. The register might. Go look.