What Is the SARA Model in Problem-Oriented Policing?

The SARA model in problem-oriented policing is a four-step framework — Scanning, Analysis, Response, and Assessment — that officers use to identify recurring crime problems, figure out what sustains them, design targeted solutions, and then check whether those solutions worked. It replaces the cycle of answering the same calls over and over with a structured way to address why the calls keep coming. A meta-analysis of 34 studies found that agencies using problem-oriented policing saw roughly a 34 percent reduction in crime and disorder compared with areas receiving standard police services.¹

Where the Model Came From

Researchers John Eck and William Spelman built SARA in 1987 during a project with the Newport News, Virginia police department. It was their attempt to turn Herman Goldstein’s 1979 concept of problem-oriented policing into something officers could actually do on a shift.² The model was originally envisioned as a department-wide strategy rather than a side project for a few specially assigned officers.³ In practice it gets used at every scale, from a beat officer tackling thefts at a shopping center to a multi-agency task force addressing drug markets citywide.

Scanning: Spotting the Problem

Scanning finds the patterns that individual incident reports hide. Dispatchers may log dozens of calls about fights at the same bar, but unless someone steps back and sees the cluster, each call gets treated as a one-off. During scanning, officers review crime data, call-for-service logs, community complaints, and their own observations to pick out recurring issues that share something in common — a location, a time, a type of victim, a method.⁴

Not every cluster is a “problem” worth a full project. The ASU Center for Problem-Oriented Policing uses the CHEERS test, six criteria a real problem should meet:⁵

Community: real people, businesses, or institutions are affected.
Harm: someone suffers actual loss, injury, fear, or damage to quality of life. Illegality alone is not enough.
Expectation: community members expect police to act, as shown through calls, meetings, or coverage.
Events: the problem is made up of identifiable, discrete events, not a vague sense of unease.
Recurring: the events happen repeatedly.
Similarity: the events share at least one common factor that ties them together.

A rash of catalytic converter thefts in a particular neighborhood might pass all six. A single dramatic robbery would not, because it is not recurring.

Analysis: Understanding What Drives It

Analysis is where officers dig into why the problem exists and what keeps it going. It is also where most SARA projects fall short. Research consistently finds that officers either skip the stage or run a surface-level review of official data before jumping to a response.⁶ A response built on a shallow understanding usually hits the wrong pressure points.

Good analysis reaches beyond police reports. Officers interview victims, offenders, witnesses, business owners, and other residents. They pull data from code enforcement, hospitals, schools, business licensing, and social service agencies. The goal is to understand who is involved, what is happening, where and when it occurs, and what conditions make it possible.⁴

The Problem Analysis Triangle

The main analytical tool at this stage is the Problem Analysis Triangle, sometimes called the crime triangle. It breaks every problem into three elements: an offender, a target or victim, and a location where the two come together.⁷ Crime happens when all three converge and nobody is exercising effective control over any of them.

The triangle’s outer ring names the people who can supply that control:

Handlers, who can influence a potential offender’s behavior: parents, teachers, employers, parole officers.
Guardians, whose presence protects targets and victims: security staff, attentive neighbors, friends walking together.
Managers, who control places: landlords, bar owners, school administrators, parking lot attendants.

The triangle forces officers to think past arrest. A problem might persist because a motel manager ignores illegal activity on the property, or because victims routinely walk through an unlit area with no natural surveillance.⁸ Addressing the weak side of the triangle often produces a more durable solution than arresting the same people again and again.

Response: Designing a Tailored Solution

The response phase turns analysis into action. The common mistake here mirrors the one in analysis: officers default to what they already know. The National Institute of Justice found that responses in many projects end up as variations on conventional tactics — crackdowns, surveillance, arrests — which defeats the point of studying the problem first.⁶

Effective responses go after the specific weak points identified during analysis. If commercial burglary is clustering at a strip mall with poor lighting and no cameras, the response may involve working with the property owner on environmental changes rather than adding patrols. If repeat domestic violence is tied to one address, the response may mean coordinating with social services and housing agencies rather than answering each new call in isolation.

Situational Crime Prevention Options

One useful way to generate response options is situational crime prevention, which groups interventions into five categories:

Increase the effort: make the crime physically harder through target hardening, access control, or exit screening.
Increase the risk: make offenders more likely to be caught through better surveillance, reduced anonymity, or active place management.
Reduce the rewards: remove or conceal targets, disrupt stolen-goods markets, deny the benefits of offending.
Reduce provocations: address frustrations, disputes, or emotional triggers that lead to offending in that context.
Remove excuses: set clear rules, post instructions, control access to drugs and alcohol.

These categories push officers toward options they would not otherwise consider, and most of them require partners. Lighting a parking lot is the property owner’s job. Changing a bar’s closing procedures requires the bar manager. Connecting at-risk individuals to treatment requires social service agencies. SARA treats those partnerships as essential, not optional.

Assessment: Did It Work?

Assessment asks a simple question with two parts. Process evaluation asks whether the response was actually implemented as planned — did the property owner install the lighting, did the partner agencies follow through? An intervention that was never carried out cannot be judged a failure of the idea. Outcome evaluation then measures whether the problem actually decreased, using the same data sources from scanning: crime reports, calls for service, community surveys, direct observation.⁴

This is where many projects quietly stop. Assessment is one of the most underused parts of the process.⁶ Officers implement a response, move on, and never check the results. Without that step, departments cannot learn from their work and may keep investing in approaches that are not helping.

Displacement and Diffusion of Benefits

Honest assessment also checks whether the response just pushed the problem somewhere else. Crime displacement occurs when offenders shift to different locations, times, targets, or methods in response to an intervention.⁹ A crackdown on street-level dealing at one intersection that moves the dealing two blocks over is not much of a win.

Research suggests displacement is less common than people assume. The opposite effect, diffusion of benefits, happens more often: areas near the targeted location also experience crime reductions, even though they were not directly targeted.¹

SARA is explicitly iterative. If assessment shows the response did not work, officers feed the findings back into analysis instead of trying harder with the same approach. Sometimes a failed response reveals something the original analysis missed.

SARA in Practice: Operation Waterside

The model is easier to see through a case. In England, the Cheshire Constabulary used SARA on what looked like a youth antisocial behavior problem and turned out to be child sexual exploitation.¹⁰

Scanning identified nearly 40 youths aged 12 to 18 involved in antisocial behavior, drug dealing, and carrying knives. Analysis using the Problem Analysis Triangle revealed something far more serious: an 18-year-old woman was sexually exploiting these young people, plying them with drugs and alcohol, and facilitating encounters with adult men. The location side pointed to a hotel whose layout let guests bypass staff using an elevator.

The response targeted all three sides at once. For the victims, officers prioritized safeguarding over prosecution and delivered education on exploitation. For the location, they worked with hotel staff to change elevator operations so reception could monitor who was coming and going. For the offender, they pursued arrest and bail conditions prohibiting contact with anyone under 16. Missing-person reports for the identified youths dropped by about 84 percent after the intervention, and both principal offenders were charged and convicted.

Why SARA Projects Struggle

Doing SARA well is harder than describing it. A systematic review of 34 studies found that about two-thirds reported significant implementation problems.¹ A few patterns recur.

Shallow analysis is the single most common failure. Officers glance at official crime data and stop, treating the stage as a box to check. Goldstein himself acknowledged the model was developed without fully appreciating how many people with research and analytical skills it would require.

Resource and leadership shortfalls are close behind. SARA projects compete with daily call volume. When training is thin and administrative support is missing, officers get pulled back to routine calls and the work stalls.

Internal resistance matters too. Not every officer or middle manager buys in. Some studies documented outright subversion of SARA projects, and one program saw limited implementation for its first nine months because of officer resistance.

Partners can also stall a project. Business owners have resisted ID-checking programs over customer inconvenience. Motel owners have refused to evict tenants involved in criminal activity because of the economic hit. Local government has blocked proposed responses when they conflicted with other priorities like redevelopment. And in neighborhoods with strained police relationships, residents may be unwilling to help with analysis or response.

These obstacles are real, but they do not argue against the model. Even with imperfect implementation, the meta-analytic evidence still shows crime reductions, which suggests that doing SARA imperfectly beats not doing it, and doing it well would produce stronger results still.¹

1
PMC (PubMed Central). Problem-Oriented Policing for Reducing Crime and Disorder: An Updated Systematic Review and Meta-Analysis
2
PMC (PubMed Central). Updated Protocol: The Effects of Problem-Oriented Policing on Crime and Disorder
3
National Academies of Sciences, Engineering, and Medicine. Proactive Policing Practices
4
ASU Center for Problem-Oriented Policing. The SARA Model
5
ASU Center for Problem-Oriented Policing. Use the CHEERS Test When Defining Problems
6
National Institute of Justice. Defining Police Strategies: Problem Solving, Problem-Oriented Policing and Community-Oriented Policing
7
ASU Center for Problem-Oriented Policing. The Problem Analysis Triangle
8
ASU Center for Problem-Oriented Policing. Step 8: Use the Problem Analysis Triangle
9
ASU Center for Problem-Oriented Policing. Analyzing Crime Displacement and Diffusion
10
Police Chief Magazine. Applying the SARA Model in Operation Waterside