POLICE INFORMATION SYSTEMS ARE DOMINATED BY ADMINISTRATIVE NEEDS AND FUNCTIONS AND OFTEN FAIL TO MEET THE DATA NEEDS OF DECISIONMAKERS, POLICE PLANNERS, AND RESEARCHERS. OFTEN THE SYSTEMS FAIL TO TAKE INTO ACCOUNT THE NEED FOR FLEXIBILITY, TIMELINESS, INTEGRATION, STANDARDIZATION, AND USER EDUCATION. POLICE PLANNING AND CRIME ANALYSIS DEMAND A FLEXIBLE DATA STRUCTURE THAT WILL PERMIT A VARIETY OF INQUIRIES BASED ON TIME AND LOCATION CRITERIA. SUCH INQUIRIES ARE PART OF STUDIES OF CHANGES OVER TIME OF CRIMES KNOWN TO POLICE (E.G., ARREST REVIEW; VICTIM ANALYSIS; OFFENDER TRACKING; RELATING CRIME TO DAY OF WEEK, TO MONTH OF YEAR, AND TO LAND USE; AND EVALUATING OFFENSES AND ARRESTS IN RELATION TO THE SOCIOECONOMIC CHARACTERISTICS OF THE PLACES WHERE THEY OCCUR). THE PROPOSED STRATEGY, ILLUSTRATED IN ITS APPLICATION TO CRIME FILES IN BUFFALO, REPRESENTS A PRELIMINARY STEP IN PROVIDING THE USERS OF CRIME DATA WITH THE CAPABILITY TO PERFORM MORE COMPLEX TIME-SPACE RESEARCH. THE RELEVANCE OF THE PROPOSED STRATEGY TO THE FURTHER DEVELOPMENT OF A COMPREHENSIVE SYSTEM INTEGRATING CENSUS AND OTHER LOCAL DATA WITH CRIME DATA IS NOTED. DEVELOPMENT EFFORTS UNDERWAY AT GEOGRAPHIC INFORMATION SYSTEMS LABORATORY, STATE UNIVERSITY OF NEW YORK AT BUFFALO, ARE DESCRIBED. DIAGRAMS ILLUSTRATING CRIME FILE ORGANIZATION AND OTHER ASPECTS OF THE PROPOSED STRATEGY ARE INCLUDED.