To Compstat or not to Compstat

Compstat is a police administration and management model. This paper was a response to a question of whether or not it should be implemented in a given agency as well as how to go about implementing it. I like many aspects of Compstat as it clarifies expectations within the organization. I’ve used it in non profit and NGO contexts with quite a bit of success since I wrote this paper and encourage the reader to go find out more about Compstat.

Compstat Analysis and Implementation

Since the dawn of civilization crime has been a constant concern for societies. Crime led Hamurabi to issue the first set of written codes for his subjects; however, he was far from the last to issue codes and laws for a populace (Samaha, 2008). As we fast forward to the present, we observe that the United States of America has a complex system of laws that are in effect.

While most citizens may not be terribly concerned with every law enacted or the subsequent criminal activity that is the violation of those laws, the average citizen does desire to be free of the fear of victimization which may include theft, robbery, burglary, assault, battery, rape, or murder (Worrall, 2007). The protection of the public and enforcement of laws logically leads to the commissioning of police departments; however, great debate erupts regarding how best to operationalize police departments. Fortunately, the many years of debate and study have revealed several operational models from which police departments can base their community protection strategies (Cordner & Scarborough, 2007, McDonald, 2002).

One such police management system or strategy that has garnered a great deal of attention lately is Compstat. The following sections will overview compstat and analyze it’s implementability within small and medium sized organizations.

Historical Police Management Systems

Compstat is an evolutionary outgrowth of police management systems and strategies that came before it, as such we must first understand the historical legacy of police management systems in order to understand the deficiencies in prior models which compstat seeks to address. The initial system was a bureaucratic and centralized system as conceived by Sir Robert Peel which can be further broken down into two historical phases (Cordner & Scarborough, 2007, McDonald, 2002).

As the traditional system was designed based upon the management structures of the military and the Roman Catholic and Anglican churches, it focused primarily around a centralized power base from which to direct police activities. At the time police activities largely surrounded addressing public order offenses (Cordner & Scarborough, 2007).

During the 1950′s a major police reform movement occurred which had the emphasis of professionalizing the police force. One of the leading factors in the charge toward reform had been abuses of power and other corrupt activities being committed by police officers. Major shifts in technology such as patrol vehicles and personal radios also caused police to be less tied to geographical locations (Cordner & Scarborough, 2007).

While many positive effects were achieved by the implementation of the professional model, it remained a highly centralized bureaucratic method. It also caused a general disconnect between the police and the community it served (Cordner & Scarborough, 2007, McDonald, 2002).

In recognition of the disconnectedness facing the police and the community, new theories in community policing were established. The basic emphasis was on shifting problem solving from a centralized and bureaucratic police activity into a grassroots partnership in addressing crime causation that is owned by the community (Cordner & Scarborough, 2007).

Unfortunately, while police departments are eager and quick to accept the community policing paradigm, in practice it rarely becomes operationalized at the line level. The typical implementation only utilizes some programs which are viewed as community policing programs; however, the majority of the staff, decision making, and strategies remain in the domain of the centralized and bureaucratic professional model (Magers, 2004).

The community policing model lost popularity during the 1980′s crack cocaine epidemic; however, has seen a resurgence as it remains a non controversial and media friendly policing model (Magers, 2004, McDonald, 2002).

The Compstat System

The compstat crime control model is the brainchild of William Bratton who during tenures in executive positions in New York and Boston developed a synthesis of police strategies and managerial concepts aimed toward crime reduction. Bratton rejected notions that the police were incapable of deterring or preventing crime; as such, rapid results in crime reduction is a focal point in the compstat system (McDonald, 2002).

Beyond the conviction that police have the ability to reduce and prevent crime, the compstat system includes five operating principles: specific objectives; timely and accurate intelligence; effective strategies and tactics; rapid deployment of personnel and resources; and evaluation, monitoring, and reassessment (McDonald, 2002). It should also be noted that compstat is a geographically linked management system in which captains or commanders are assigned precincts or districts which they are accountable for.

Specific objectives are an early step in any strategy and are not unique to compstat. Strategies must be based upon objectives and objectives must be based upon mission, in this way the mission and vision of police agencies are realized. As compstat focuses on the community protection mission to which veritably all police agencies ascribe, the objectives set forth must be the reduction of crimes, specifically the reduction of particularly problematic crimes in the community. This objective can be the reduction of robberies or auto thefts or whatever crime the community has cause for concern (McDonald, 2002).

The second facet of compstat, timely and accurate intelligence, draws primarily upon the prevalence of modern computing solutions. By leveraging the back end databases of modern report management systems (RMS) or management information systems (MIS), a crime analyst can directly connect a geographical information system (GIS) which is subsequently used to visually represent areas being affected by crime trends. Proper analysis can also begin to explore solutions as additional data types are aggregated such as the locations of abandoned buildings or schools in relation to crime hot spots (McDonald, 2002). Proper use of technology can make what would be a particularly time consuming activity of creating and analyzing crime statistics into a nearly instantaneous affair, which gives early warning to police managers as to developing crime trends.

The third element of compstat, effective strategies and tactics, centers around the idea of utilizing all potential sources of prevention at the police agency’s disposal, including patrol, specialized units, and other public agencies. With full cooperation between police sub groups and other public agencies, comprehensive strategies can be implemented to form a gestalt upon a specific crime pattern (McDonald, 2002).

Further elements in attaining effective strategies is the analysis of information attained via the timely and accurate intelligence facet of compstat. Furthermore this information needs to be readily available and easily communicated amongst precincts (McDonald, 2002).

The fourth aspect of compstat, rapid deployment of personnel and resources, involves two primary factors, that being command accountability and cooperation amongst precinct commanders and other command staff. For compstat to be effective, command staff are expected to be enthusiastic and ready to commit various resources in ways which will affect crime. Command staff are then held accountable toward their results in deploying resources in such a way that crime rates drop or crime trends end (McDonald, 2002).

Finally, evaluation, monitoring, and reassessment, entails the reporting and analysis of results obtained from a specific strategy and resources committed to a given problem. The New York compstat model is highly aggressive in this regard, so much so that Phyllis Parshall McDonald (2002) refers to the system as “relentless.” When problems persist despite strategic initiatives, problems are reevaluated and coaching is provided to command staff or barring teachability, staff are replaced outright (McDonald, 2002).

Compstat Benefits

Compstat is a proven and effective police management system when addressing the criminality of places (McDonald, 2002, Schmalleger, 2009). According to Dorriety (2005) the implementation of compstat in New York resulted in a 27% reduction in crime over a two year period from 1993 to 1995, while the national average dropped a mere 2%.

Compstat is generally well received by officers and command staff as they are empowered to reduce crime and address social problems. Furthermore, compstat offers a larger range of management participation from all levels of the police organization which has a positive effect on problem solving, information gathering, and personnel retention (Cordner & Scarborough, 2007, McDonald, 2002).

One of the advantages to compstat is that it more efficiently leverages various specialist units to work alongside patrol in addressing problems and crime trends. Traditional police management is inefficient as it essentially fractures the police agency into competing sub groups each with it’s own group culture. Compstat effectively places every group onto the same team with a singular purpose of working together in order to reduce crime (Cordner & Scarborough, 2007, McDonald, 2002).

Finally, compstat protects against group thinking and encourages out of the box solutions that are more effective than the status quo (Forrester & O’Connell, 2009).

Compstat Risks

Compstat is not without it’s risks, the first of which being that it regresses somewhat back into a bureaucratic police model; however, it is still highly geographically focused so does not fall back into an overly centralized model. While this does not preclude the use of community policing, it does require special attention if a police executive wishes to utilize both concepts in tandem (Magers, 2004, McDonald, 2002).

As compstat tactics have a tendency of focusing on criminality of places, massive short term crime reduction is common; however, crime and perpetrators can be highly transitory and move from area to area. It is not uncommon for criminals in the light of increased police pressure in one neighborhood to begin committing crimes in another; therefore, it is important that trends be analyzed amongst neighboring districts to ensure that crime isn’t merely being shuffled amongst precincts and districts (Magers, 2004, McDonald, 2002, Schmalleger, 2009).

A wise police administrator must remain cognizant to employing or eliciting cooperation in addressing other criminalogical theories. Examples would be working with mental health or social services to prevent crimes committed by the mentally ill or others committing crimes due to biological or psychological factors, or partnerships with education departments in order to reduce anomie and other social structure causes of crime (Schmalleger, 2009).

Finally, compstat has a reputation as an aggressive policing system which has the potential of causing concern amongst politicians, the media, and citizens. Some studies accused compstat of increasing complaints levied against the police; however, this accusation has since been shown to be unfounded (Davis, Mateau-Gelabert, & Miller, 2005).

Compstat Implementation in the Small or Medium Agency

In the sample police organization, a 150 sworn officer department currently utilizing a community oriented policing approach, are determining whether or not to implement compstat and the processes involved. In deciding whether or not to integrate compstat one must first ask “is there room for improvement,” “are we fully utilizing our resources in an effective and efficient manner,” and “are we actually fully implementing community policing or are we just implementing some practices of community policing?”

Just as Magers (2004) pointed out that very few police organizations are completely utilizing a community oriented policing approach other than some community based projects, we can assume that our sample police organization has room for improvement, is not fully utilizing resources in an effective and efficient manner, and is not fully implementing community policing.

Fortunately, compstat can be fused into other environments such as community policing environments. Furthermore, compstat can be accomplished on a rather low budget, thus alleviating budgetary and manpower concerns (Dorriety, 2005).

The first stage in implementing a compstat or compstat like program is a decision by the chief, sheriff, or commissioner to implement. From there he or she must bring all other stakeholders on board by evangelizing others toward the system. An important part of this process involves jointly naming the system (McDonald, 2002).

The next stage involves training commanders and captains to be accountable and responsible for their geographical area. As our department was already utilizing community policing, they should have been operating according to geographical zones already; in the event that they were not delineated into geographical sectors, then it was likely not particularly effective community policing prior to our system change, thereby rendering arguments pro community policing without compstat integration relatively moot (Cordner & Scarborough, 2007, McDonald, 2002).

During this phase it is also important to upgrade or establish computer systems which can analyze the crime data as well as any other data we wish to aggregate and analyze against crime data. While the New York Police Department and other massive police organizations outlay large sums of money to use database back end systems such as DB2 from IBM or 11G from Oracle (Lai, 2006), they have a need for stability during an incredibly high number of connections to the databases. Our department of 150 officers could expand well into 600 or more officers concurrently using databases from the free and open source software community (FOSS) such as MySQL and PostgreSQL. These databases are common back ends for RMS and CAD systems and integrate well with FOSS alternatives to GIS software such as ArcGIS. Future developments made possible by grants from the federal government will continue to see further developments in helping departments reduce costs of software systems and departmental interoperability by utilizing open source public safety programs such as ISES, Posse, OpenRMS, and Tickets.

It should be noted that the entire compstat database in 2006 still had not exceeded the 80 gigabytes of space that had been allotted at the program’s inception (Lai, 2006). This means that records can be easily stored on a virtualized infrastructure leading to no additional costs in hardware purchase and easy upgrades should the data needs expand.

The next stage in compstat implementation is the setting of annual objectives. These objectives should not be administrative objectives such as “hiring 5 new officers,” but rather objectives to reduce problematic crimes such as “to reduce burglary and auto theft” (McDonald, 2002). These objectives will guide commanders and captains in addressing strategies for the entire year and are thus vital to the organization.

Finally, compstat meetings should begin being held by an executive with the authority to force cooperation between specialized units and patrol. At meetings the executive is a coach and role model for commanders who have likely never been asked to be accountable in addressing crime problems within their areas. It may not be necessary to hold meetings on the schedule of the NYPD, some police agencies hold compstat meetings quarterly with success; however, this would depend largely on the task maturity level of commanders (Cordner & Scarborough, 2007, Dorriety, 2005, McDonald, 2002).

One need not feel locked into the measurements used by the NYPD or other compstat based organizations. It is reasonable in a community oriented policing organization that a monitoring and evaluation factor include community fear levels and numbers of contacts made with the community regarding problems identified.

Finally, as with any management system, continual review and revision will help streamline and integrate the workings of compstat which is problem and information oriented with the working of a community oriented model. Although McDonald (2002) does not recommend it, experimentation can even be used pertaining to the inclusion of community members in creating strategies for dealing with crime and public order problems at certain compstat meetings.


Police management has progressed immensely during the past past and current century; however, we can not presume that modern methods are the apex of police management. Police administrators should continuously evaluate new research in the areas of management and crime control that will enable their organizations to be more effective and efficient.

While wholesale adoption of compstat may not be ideal in every agency, it contains principles which can achieve immediate short term crime reduction which so long as activity levels of precincts are sustained, crime patterns can be regularly disrupted to the point of reducing overall crime. This and other factors warrant enough attention that every police administrator should glean at least some managerial practices from the experiences of compstat.


Cordner, G.W., & Scarborough, K.E. (2007). Police administration. Newark, NJ: Lexis Nexis.

Davis, R.C., Mateu-Gelabert, P., & Miller, J. (2005). Can Effective policing also be respectful? two examples in the south bronx. Police Quarterly, 8(2), Retrieved from doi: 10.1177/1098611104269531

Dorriety, J. (2005). Compstat for smaller departments. Law & Order, 53(6), Retrieved from

Forrester, D., & O’Connell, P. (2009). Think differently: update your stats to unlock outcomes. Public Manager, 38(4), Retrieved from

Lai, E. (2006). Nypd launches third phase of data warehouse project. Computerworld, 40(36), Retrieved from

Magers, J.S. (2004). Compstat: a new paradigm for policing or a repudiation of community policing?. Journal of Contemporary Criminal Justice, 20(1), Retrieved from doi: 10.1177/1043986203262312

Samaha, J. (2008). Criminal law 6th edition. Belmont, CA: Thomson Learning.

Schmalleger, F. (2009). Criminology today: an integrative introduction. Columbus, Ohio: Pearson.

Worrall, J.L. (2007). Criminal procedure: from first contact to appeal. Boston, MA: Pearson.

Leave a Reply

Your email address will not be published. Required fields are marked *


You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>