Desire lines: what the paths people choose reveal about a place

There is a park in Helsinki where the landscape architects, faced with designing the footpath network, made an unusual decision: they did not design one. Instead, they waited for the first snowfall of winter, walked the site, and mapped the tracks that visitors had already made in the fresh snow, the natural routes, chosen without signage or instruction. Those tracks became the paths.

Urban planners call them desire lines: the routes people actually take, as opposed to the routes designers intended them to take. The worn diagonal across a corner of grass is one of the most legible pieces of data a place can generate.

The geometry of least effort

The shortest distance between two points is a straight line, and people are good at finding it. Space syntax research, the academic discipline that analyses how street networks shape movement, describes this as least-angle routing: most pedestrians, consciously or not, minimise the number and severity of turns between origin and destination, even when a slightly longer route exists. Geometry, in other words, largely predicts where people will walk and where they won’t.

Desire lines are what happens when the designed geometry and the natural geometry diverge. The formal path goes around the edge; the desire line cuts the corner. The official entrance faces the main road; the desire line leads straight to the car park. Neither is wrong, but only one of them reflects what people actually do.

From parks to city squares

Cities that treat desire lines as feedback rather than trespass tend to end up with infrastructure people actually use. Finnish planners have formalised the snowfall method, visiting parks after the first dusting so that footprints reveal flows before any official paths are marked.

Rem Koolhaas reportedly took a similar approach when redesigning the Illinois Institute of Technology campus in Chicago, letting student movement patterns shape the pathways before anything was paved.

There is even a historical case that Broadway, New York City’s diagonal slash through its grid-bound street plan, follows the Wickquasgeck Path, a Native American trail threading the least-cost route between settlements on Manhattan, avoiding swamps and hills.

The desire line outlasted every subsequent plan laid over it.

For city centres and smart city planners, the same logic applies at district scale. Where do pedestrians actually cross? Which entrances do visitors use? Which public squares are traversed and which are skirted? Read across a whole neighbourhood, these routes are the footfall pulse of the district. Answering those questions accurately requires data, not observation from a single vantage point at one time of day.

The indoor version

Step inside a retail store or a transport hub and the dynamic is identical, just with a roof on top. Shoppers do not follow the gondola layout the planners drew; they follow sightlines, floor-level cues, and the path of least resistance. Some zones receive far more traffic than their position on a floor plan would suggest; others sit chronically under-visited despite being physically central.

Retail heatmaps make this visible. Built from anonymous, aggregated position data rather than cameras watching named individuals, a zone-level heatmap shows the dominant flows through a space, the desire lines of the indoor environment, so that fixtures, signage, and category placement can follow what people actually do. For cities and municipalities managing large public buildings or transit spaces, the same data informs everything from wayfinding placement to cleaning schedules. The Wi-Fi-based people counting approach does this without identifying a single person in the building.

Why planners resist them, and why they shouldn’t

The instinct to block a desire line with a bollard is understandable. The designed path exists for reasons: drainage, safety, sight lines. But a desire line is not a complaint; it is evidence of a mismatch between the designed environment and human behaviour, and that mismatch has costs. Retail dead zones lose revenue. City squares that pedestrians skirt become underused assets. Public spaces with confusing flow patterns create congestion and resentment.

The smarter response is to treat the informal path as a signal. Is the designed route too long? Is the entrance not obvious? Is the corner more convenient than the official route because of what it connects? Those are design questions, and answering them well requires knowing where people actually go.

What data reveals that observation cannot

Standing in a space and watching is not the same as measuring it. A person at a doorway sees the next few visitors; movement data sees the pattern across tens of thousands of journeys, across every hour and weather condition. It surfaces things that feel counterintuitive until you see the numbers: the entrance nobody uses at lunchtime, the zone everyone traverses but nobody stops in, the sightline that pulls people reliably off the designed route.

Cellular network analytics does this at district scale, aggregating movement across a whole urban area from signals phones exchange with the mobile network , no app, no camera, no identification of individuals. The same data also shows where those visitors came from in the first place. The data deliverables from this kind of system amount to a desire-line map of the city: where people go, in what volumes, at what times.

Following the feet

The desire line is a democratic signal.

It records what thousands of people, acting independently, found to be the better route, without anyone asking them. Urban design and retail planning have long argued between top-down imposition and bottom-up emergence. Desire lines cut through that argument: the answer is iteration. Design, measure, adjust.

The worn patch of grass is already an answer to a question. The job of good measurement is to read it at scale, across all conditions, without waiting for the season to change. Anonymous movement data lets you do that (for the park, the shopping street, and the shop floor alike) and it is the only footfall method in Europe approved by a data protection authority.