Home › Our Approach
Methodology
We show the arithmetic.
A public safety recommendation is only as good as the derivation behind it. This is how ours are built: the data we collect, the models we run, and the standards we hold the results against.
The single most common weakness in public safety studies is an unexamined denominator. A report recommends twelve additional officers; the recommendation rests on an assumption about how many hours a year an officer is genuinely available for patrol; that assumption is never printed, and so it is never tested.
We take the opposite approach. Availability is calculated line by line from the agency’s own payroll and timekeeping records. We begin with the base work year and deduct every category of leave, all on-duty training, court and witness time, and administrative duty. The resulting figure is published in the report. A finance director who disagrees with one of those lines can substitute their own number and see exactly how the recommendation moves.
Phase one
Assessment and baseline establishment
Before any modeling begins, we build a factual picture of the current state from primary records rather than from what the organization believes about itself.
Records, not impressions
Computer-aided dispatch exports covering multiple fiscal years, records management extracts, payroll and timekeeping data, court appearance logs, training records, position control plans, adopted budgets and capital improvement programs, collective bargaining agreements, annual financial reports, comprehensive plans, and building department permit records. On a comprehensive engagement this routinely means analyzing hundreds of thousands of individual dispatch records across four or more years.
Structured stakeholder engagement
Customized data collection instruments, structured interviews and workshops with command staff, line personnel, civilian staff, human resources, finance, and labor representatives. Where a project affects the public directly (a referendum, a change in service model, a new facility), surveys and focus groups capture community priorities, so the recommendation is informed by more than the organization’s internal view.
The analytical core
Named methods, applied to your data.
These are the frameworks that produce our findings. They are established, published methods drawn from the professional standards bodies of each discipline rather than proprietary black boxes, combined with analytical structures we have refined across hundreds of engagements.
Net availability derivation
The foundation layer, calculated separately for each role because the deductions differ. A patrol officer, a detective, and a dispatcher each start from a standard work year and lose materially different amounts to leave, on-duty training, court time, and administrative duty. Every downstream staffing figure divides by these numbers, so we derive them explicitly from agency records and publish the line items.
Workload-based patrol staffing
A bottom-up computation that measures committed call-handling time, administrative and support time, and a reserved proactive capacity, then returns the officers required per shift. It relies on neither peer ratios nor population heuristics nor the department’s own historical authorized strength. We run it under multiple sensitivity cases so the client sees a defensible floor and ceiling rather than a single figure.
Response time and demand analysis
Response performance is broken into call processing, dispatch, travel, time on scene, and total call duration, then reported at the median and the 90th percentile rather than the mean. Averages conceal the tail that residents experience. Demand is partitioned by hour of day, day of week, priority level, and geographic district, producing the temporal and spatial picture that deployment decisions require.
Investigative workload by case type
An advance on the caseload-ratio approach that has long dominated investigative staffing. Rather than benchmarking cases per detective, we decompose each case category into its component tasks, among them evidence handling, digital forensics, subpoena work, surveillance, victim and witness interviews, and prosecutor consultation. We then assign hours to each and weight them by how often each task is actually performed. The result is a workload figure a supervisor can recognize and argue with.
Fire and rescue operational analysis
Evaluated against the national standards that govern the discipline: minimum staffing and response benchmarks, effective response force assembly, station location and apparatus deployment efficiency, and unit hour utilization. We also model the fiscal and operational consequences of alternative shift configurations on overtime cost, fatigue, and retention.
Communications center staffing
911 and dispatch centers are staffed against their own discipline’s standards, using target console utilization, call processing time, and peak-window modeling, then measured against accepted service-level and abandoned-call thresholds. Turnover is treated as a staffing variable in its own right, because a center that replaces every position it loses can still be losing ground.
Compensation and benefits benchmarking
Peer sets are built on stated criteria: population and demographic comparability, geographic and operational similarity, labor market proximity, and regional compensation positioning. They are often stratified into tiers so a client can see itself against direct peers, against the larger metropolitan markets competing for the same candidates, and against federal employers. Salaries are standardized to a common annual-hours basis before comparison, commonly 2,080 hours for forty-hour classifications and 2,756 or 2,912 for fire schedules, then evaluated at every career stage rather than at entry alone. Base pay is compared in isolation first, before incentives and benefits are layered in, so that the cause of any gap is visible. Total compensation and take-home pay are both reported, since employee pension contributions can reverse a favorable ranking. Full detail is on the compensation and benefits page.
Fiscal impact and risk modeling
Cost projections are built on fully loaded personnel cost, which combines base pay with retirement, insurance, payroll taxes, workers’ compensation, equipment, training, and backfill, so that recommendations reflect true fiscal impact. Multi-year projections are presented under conservative, moderate, and aggressive scenarios, and on major engagements we run probabilistic simulation across the variables that actually drive the outcome, reporting results at multiple probability thresholds instead of a single point estimate that implies precision no forecast possesses.
Growth projection and forward planning
Forward staffing is not a uniform growth rate applied to a headcount. We divide the service area into planning zones with independent growth trajectories, weight the development pipeline by how far each project has actually progressed through permitting, and classify every position in the organization by what it scales with: workload, population, fixed coverage requirements, span of control, or nothing at all. The ten-year projection is then auditable position by position.
Space needs programming
Facility requirements are developed room by room against published space standards for each function, validated against the operational and staffing model, and projected forward on the same growth horizon. Programming addresses security zoning, evidence and records custody, public access separation, and code and statutory compliance, then produces an estimate of probable cost built on current regional construction data.
Triangulation
Where it matters, we run more than one model.
Independent methods applied to the same question rarely land on the same number. That is useful information, and we report it rather than hiding it.
Bracket, don’t assert
Two accepted staffing models applied to the same department will produce a conservative floor and a defensible upper bound. Presenting both, and explaining why they differ, is more honest and more useful than choosing one and calling it the answer.
Cross-validate
Findings from workload analysis are checked against utilization measures, peer position, and the department’s own performance trend. A conclusion that only one method supports is flagged as such rather than presented as settled.
State the limits
Every report names its own data limitations, benchmarking limitations, and projection limitations, and assesses whether any of them is load-bearing on the principal finding. Studies that claim no limitations invite the discovery of one.
What you receive
Deliverables built for two audiences at once.
A study has to work for the chief who will operate from it and for the council that will fund it. We write for both, and mark clearly which sections serve which reader.
- ReportA comprehensive written report with an executive summary written in plain language for elected officials and administrators.
- MatrixA consolidated recommendations matrix listing every recommendation with its lead time, fiscal impact, and priority, each traceable to the section that justifies it.
- ModelA multi-year fiscal impact model under conservative, moderate, and aggressive scenarios, with the assumptions exposed and adjustable.
- BenchmarksPeer comparison tables covering staffing, compensation, benefits, and service performance, with the peer selection criteria stated.
- ProjectionA multi-year staffing projection built position by position rather than as an aggregate growth rate.
- ProgramWhere facilities are in scope: a room-by-room space program, options analysis, and estimate of probable cost.
- RoadmapA sequenced implementation plan with responsible parties, critical path, and contingency strategies for the risks identified.
- PresentationFormal presentation of findings to command staff, management, and the governing body, plus supplementary briefings as needed.
A focused single-agency assessment is typically delivered within 120 days of notice to proceed. A comprehensive multi-agency assessment runs closer to 180 days. The schedule is agreed at the outset and reported against throughout.
Start the conversation
Tell us what your agency is facing.
Whether you are preparing a budget request, responding to a growth pressure, or standing at the beginning of a facility project, a short conversation is usually enough to tell you whether a study is the right next step, and what it would involve.