Public Safety9 min
Many municipal police departments already have crime maps, yet patrols are still assigned by neighborhood and by habit. This article explains what a map is missing before it becomes a plan: the right scale, the time of day, the dosage, someone accountable and verification of what happened.
Public Safety9 min
Wallets, phones, keys and cash pass into the authorities' hands in every administrative detention, yet in many holding cells they end up in a bag with a name scrawled on it. This article explains why safeguarding belongings is a recordkeeping duty and how to control it from the case file.
José Campillo
Public Safety8 min
Mexico's Constitution caps arrest for administrative offenses at 36 hours, yet in many municipal holding cells nobody records when a person came in or when they must leave. This article explains why tracking arrest time is a recordkeeping obligation and how to control it from the case file.
José Campillo
Public Safety9 min
Every person who enters a municipal civic court arrives with a health condition that someone must examine and document. This article explains why the intake medical certificate is a safeguard for the person and for the institution, and why it must live in the same case file as the booking and the ruling.
José Campillo
Public Safety10 min
Every traffic crash a municipality handles ends up in three places: the parties' proceedings, official statistics and road planning. This article explains why the expert crash report needs structured capture, review with approval and a locked record, not just a well-drawn sketch.
José Campillo
Public Safety9 min
A fast lookup does not improve public safety if it returns isolated hits without the linked history needed to understand recurrence, associated events, and prior decisions. This article explains why public safety and civic justice need contextual search built on a shared operational record.
José Campillo
Public Safety7 min
When admission, medical review, personal property, custody handoffs, prior-record search, and release are tracked in separate records, institutions lose continuity exactly where control matters most. This article explains why detention has to be governed as an operational file rather than a booking log.
José CampilloMonday, 8:00 a.m. The command meeting opens with a map projected on the wall. Three red blotches cover downtown, a neighborhood on the east side and the bar district along the main avenue. Someone says what everyone already knew: "that's where it happens." The director asks to reinforce those areas. The shift commander nods, leaves the meeting and assigns the same units to the same sectors as last week.
A month later, the map looks almost the same. No one can say whether the patrols were on the red streets, at what time, for how long or what they did there. Nor whether the blotch grew or moved because of something the police did or because of something unrelated to them.
The map did not fail. It showed where the records piled up. What was missing was everything between the map and the street: routes, schedules, accountable people and a review of whether it worked.
Mexico's National Model of Policing and Civic Justice (Modelo Nacional de Policía y Justicia Cívica), published in the Official Gazette of the Federation in January 2021, frames the shift clearly. In its comparison between reactive policing and problem-oriented policing, it puts "random patrol" on one side and "strategic patrol" on the other. On one side, the emphasis on response capabilities; on the other, the emphasis on analytical capabilities.
The same document asks the police, with support from their analysis units, to identify the environmental conditions that enable crimes and administrative offenses, and to do so through systematic pattern analysis: the type of offenders, the characteristics of places, the times and the property being targeted. The Model also sets the unit of organization: the quadrant. Its size should allow officers to cross it from one end to the other in no more than 5 minutes, and the officer in charge of each quadrant must answer for what happens in it and justify its productivity indicators.
There are two of the pieces a map does not bring on its own: an analysis that explains the pattern and a person who answers for a territory. A third piece is missing, one the Model does not detail and the international evidence does: scale.
A heat map usually shows soft blotches covering an entire neighborhood. That image is misleading about how crime is distributed.
In 2019, Spencer Chainey and a group of analysts from the region published in the journal Crime Science the first comparative study of crime concentration at micro-places in Latin America. They analyzed data geocoded by street segment (the stretch between two intersections) in cities in Argentina, Brazil, Colombia, Mexico, Uruguay and Venezuela. On average, across all crime types and all cities, 0.8% of street segments accounted for 25% of crime, and 2.5% of segments accounted for half. For homicide, concentration was even higher: in every case, half of all homicides occurred on no more than 1.4% of segments.
The authors conclude that crime is concentrated in Latin American cities at higher levels than in the industrialized countries where this line of research emerged, and that this opens opportunities for programs precisely targeted at those micro-places.
The operational consequence is direct. If half of a city's robberies happen on a small fraction of its streets, assigning a patrol to "the neighborhood" dilutes the effort across hundreds of street segments where almost nothing happens. A red blotch over a neighborhood does not tell you which ten blocks sustain the problem. A list of segments does.
The most cited systematic review on the subject is by Anthony Braga and colleagues for the Campbell Collaboration, updated in 2019. It gathered 65 studies with 78 tests of police interventions at hot spots, including 27 randomized trials. Its conclusion: focusing police on high-activity crime places produces small but statistically significant reductions in crime and disorder, across drug offenses, property crimes and violent crimes.
The review also answers the most common objection, that crime simply moves around the corner. In the 40 tests that measured it, crime control benefits were more likely to diffuse into surrounding areas than crime was to be displaced.
A third finding matters for program design: problem-oriented interventions, which analyze why a place concentrates crime and act on those conditions, produced larger effects than those that only increased police presence.
The evidence from Latin America is more recent, and it carries a lesson the map cannot teach. In 2017, four Argentine cities (La Plata, Morón, Santa Fe and Tres de Febrero) implemented hot spot patrol programs with a very similar design, focused on robbery and theft against pedestrians.
The team identified the street segments that together accounted for 25% of those crimes, drew foot patrol routes of no more than 8 segments each and organized pairs of officers who spent one hour on each route before rotating to the next. The results were published in 2022 in The Police Journal.
In La Plata, robbery fell a significant 31% in the second half of the intervention and theft fell 39% in the first half. The other cities saw a mix of non-significant decreases and even increases. No displacement to assaults or vehicle crime was observed.
The authors' explanation is about management, not about the map. Santa Fe implemented only five routes, compared with 25 in La Plata, and covered them in long blocks spread throughout the day. La Plata covered each route in shorter periods concentrated on certain days and times, for example Thursday, Friday and Saturday from 10:00 p.m. to 2:00 a.m. All four cities started from the same kind of analysis. What changed was how many places were covered, when, and with what intensity.
Given that evidence, a data-driven patrol plan has at least six elements. The map provides the first one and part of the second. The institution has to build the rest.
The last point is the one most often skipped. Without a record of presence, a poor result cannot be told apart from a plan that was never carried out.
There is a caveat worth making before turning any map into a plan. Maps are fed by records, and not every record measures the same thing.
A map of complaints and calls shows, within its limits, where people report problems. A map of arrests shows something else: where the police were and acted. If the police patrol an area more, they detect more offenses there, the map turns it redder and the plan covers it again. The loop reinforces itself, even if the problem lies elsewhere.
That is no reason to discard arrest data. It is a reason to read each map layer knowing where it comes from and to check whether the concentration also shows up in what residents report. The National Model points in that direction when it calls for strengthening how municipal police receive crime reports.
Tribuna does not decide where to patrol. That decision belongs to command staff. What it does is make sure the decision starts from organized data and that operations leave a trail to review afterward.
It all starts at capture. In Tribuna, events are recorded by arrest, call or patrol round, with a georeferenced location from the tablet or the desktop. Having patrol rounds as located events is what makes it possible, over time, to compare where the police were with where incidents happened. And because the tablet app stores records on the device and syncs them when the signal returns, the location is not lost in areas with poor coverage.
On top of those records, the intelligence map offers layers for incidents, heat, neighborhoods and the last 7 days. It shows concentration by hour and a day versus night comparison, the top offenses, repeat offending and average resolution time, plus a timeline to see how the city changes between periods. With that, an analyst can move from "downtown is red" to "these streets concentrate these offenses between these hours," which is the input a plan needs.
Every read and every change is logged, with role-based permissions, so access to sensitive information used to plan operations also leaves a trail.
It is the same idea we developed in A command center is not a screen wall: seeing more information is not enough if it does not translate into shared, verifiable decisions.
Before buying another map layer or a bigger screen, the public safety department should be able to answer with data:
If several answers depend on what the shift commander remembers, the problem is not the lack of a map. It is that the map is not yet connected to operations.
A heat map is not a patrol plan. It is a starting point that shows where records piled up. The plan begins when someone translates that image into specific streets, schedules, time on location, accountable people and a way to verify what happened.
The Argentine experience makes it clear: with the same analysis, results depend on how execution is managed.
If your institution wants to move from the map on the wall to patrols that can be reviewed, explore Tribuna, see how Agora organizes the citizen reports that also inform how the territory is read, or request a demo.