What Works in American Policing Part 7—Conclusion: No Single Strategy Makes Public Safety Work

This is the final installment in a seven-part series examining the policing strategies that shape public safety in America, including what the evidence supports, what has fallen short, and what policymakers and practitioners should prioritize going forward.

This series set out to assess the main strategies behind modern American policing and treat them as real tools rather than slogans or campaign talking points. To accomplish this, we asked the same five questions about each strategy:

1. Is there solid research that it deters or helps solve crime?
2. Does it move the outcomes that matter—crime and clearance rates, use-of-force trends, and community trust—rather than just activity counts like arrests?
3. Can a traditional department pull it off given real staffing and funding limits?
4. Does it stay inside constitutional limits?
5. Above all, is it a responsible use of public money backed by data transparency and honest outcome measurement?

Judging each approach on its own, the same pattern kept surfacing: No single, master strategy can carry an agency or community’s entire public-safety load. Those with the best track records combine several at once, matching each strategy to the problem it fits best while giving the most weight to those with the strongest evidence behind them. The real question for a chief, commissioner, or sheriff is not which strategy to choose, but which combination to build and whether their agency has the people and resources to run it.

The pressure behind all of this comes from three problems laid out in the series introduction:

1. Too many serious crimes go unsolved.
2. Departments cannot hire and keep enough officers.
3. Public trust is lower than it should be.

These troubles are not separate; instead, they feed one another. When an agency is short-staffed, relationship building is the first thing cut. Without those relationships, there is no trust; without trust, there is no cooperation—the very thing needed to solve cases. And as cases pile up unsolved, the public loses even more faith in the system. Multiple strategies must work together in order to break this cycle.

Where Each Strategy Stands

The information in the following table can help agencies determine which strategies to combine for the best outcomes.

The two “Foundational” approaches are preconditions rather than options, as no combination can succeed without evidence discipline and community trust. The strategies rated “Deserves investment” should carry most of an agency’s proactive effort because the research behind them is strongest. The remaining two belong in a supporting role, to be adopted only where an agency can meet the required safeguards. In other words, combining strategies does not mean relying on all of them equally; it means weighing each according to the evidence behind it.

Putting these into practice rarely means starting from scratch. Most agencies can lean on existing evidence (e.g., CrimeSolutions and other clearinghouses) to pilot changes before scaling them. Several practical steps apply across strategies: assign patrol officers to the specific streets the data flags, pair those officers with a crime analyst to diagnose why a problem persists, add non-enforcement fixes like lighting and code enforcement, and keep the same officers in the same neighborhoods long enough to earn the trust that makes tips flow. The two surveillance-heavy approaches—intelligence-led and predictive policing—require additional safeguards before adoption, including written retention rules, independent audits, open and explainable models, and human review with the authority to override an algorithm’s recommendation. Where a single department lacks the capacity, regional partnerships and university collaborations can fill the gap.

How to Keep Them Straight

While the six strategies can blur together, they become easier to distinguish once you know what question each one answers. Two run underneath everything:

• Evidence-based policing is the method: how a department decides what to try and checks whether it worked.
• Community-oriented policing is the foundation: the trust that determines whether the public will cooperate with it.

The other four each answer a different question:

• Where is crime concentrated? (Hot spots policing)
• Why does it keep happening there? (Problem-oriented policing)
• Who is driving the most serious harm? (Intelligence-led and precision policing)
• Can we anticipate where or who comes next? (Predictive policing)

Seen that way, strategy selection begins with problem diagnosis. A commander facing a stubborn trouble spot does not ask whether to “do” hot spots or problem-oriented policing; instead, they ask what is actually driving it—a place, a condition, a handful of people, or some mix—and selects the approach that fits. The diagnosis points to the response. For example:

• A cluster of late-night robberies at one subway stop is a place problem.
• A convenience store generating the same calls every month is a condition problem.
• A feud among a few individuals behind a neighborhood’s shootings is a people problem.

Evidence and trust apply across the board, so every choice should be tested and grounded in a real relationship with the community it touches.

What Blending Looks Like

Most serious problems involve more than one of those things at once, which is why the strongest agencies blend responses. A single violent corner is often a place, condition, and people problem, and addressing only one of those allows the others to persist.

Hot spots analysis might flag the block as one of the city’s worst (the where). A problem-oriented perspective might note a vacant building, an unlit lot, and a bar that overserves at closing—all conditions that code enforcement, better lighting, and licensing pressure can change (the why). Intelligence and focused deterrence identify the few people behind most of the violence, pairing a direct warning with an offer of help (the who). Neighborhood residents are willing to share what they see with trusted officers, and the department tracks calls and violent incidents both before and after implementation to confirm the efforts are working (or to adjust if not). No single strategy could have solved the problem alone; together, all gaps are covered.

The opposite is also true: strategies that succeed together fail when used alone. Place-based enforcement without community trust devolves into the indiscriminate crackdowns known as zero-tolerance policing, breeding resentment and potential lawsuits. Problem-solving without an analyst is guesswork. A predictive tool employed without the discipline to check it just automates yesterday’s bias. Understanding how and when to blend strategies is a skill every department must build.

The Capacity Problem

Running any of these strategies, let alone several at once, takes more than willpower—it takes people who can read crime data, partners who can tell whether something is working, and a culture willing to act on the answer. It also takes officers. The same staffing shortages described at the start of this series limit how much proactive work any agency can absorb. Most of the country’s roughly 18,000 police departments are small, run by a few officers with no analyst and no budget to hire one. Without the staff to implement it, a strategy remains just an idea on paper.

The encouraging part is that capacity can be shared. Regional partnerships, pooled analysts, and university collaborations give small agencies access to analytical capability they could not produce on their own, and several of these strategies work in small and mid-sized jurisdictions as well as in large cities. A mid-sized department in Kansas drove crime down citywide with no overtime or new hires by building focused, data-guided patrol into regular shifts. This can scale to agencies of any size with the willingness to let data rather than routine determine where officers focus.

Accountability Is Not Optional

Every enforcement- and surveillance-heavy strategy shares one requirement: guardrails. Precision, intelligence-led, and predictive policing all work by zeroing in on specific people and places, which is why transparency, written rules for who gets targeted, independent audits, honest reporting, and clear constitutional limits are not optional. They are what keeps a focused strategy from sliding into mass stops or a quiet watchlist.

Openness also pays trust back. The same records that let a department test and sharpen its own work offer some of the most convincing evidence to put in front of a doubtful public. An agency that can show what it did and what changed argues from strength, and because cases get solved when people are willing to come forward, that transparency contributes directly to higher clearance rates.

Translating This Into Practice

The takeaway for lawmakers is that analysts, data systems, training, and partnerships outlast the short-lived task force, and any support should come with evaluation and transparency attached. Be skeptical of any product—especially a proprietary algorithm—that asks for backing before it has shown independently that it works.

The takeaway for police leaders is that strategies must be combined deliberately and their results measured. Send resources where the evidence points, build or borrow the analytic capacity these strategies assume, and reward officers for solving problems and earning trust rather than for raw activity counts. Build accountability into anything surveillance-adjacent before deployment rather than after failure.

The discipline is the same for everyone. Aim resources where the harm actually occurs, test whether the response works, and track the results openly. Applied consistently, those commitments can turn separate strategies into a coherent public-safety plan.

A Final Word

Maintaining public safety is one of the most important jobs a government has and one of the largest expenditures in nearly any community’s budget. What the field needs now is the patience to fund what the evidence backs, the humility to keep testing what only looks good so far, and the honesty to call the unproven what it is: the standard a core public responsibility deserves.

The strategies outlined in this series are complements rather than competitors, but they are not equals either. The evidence justifies investment in some, further testing for others, and guardrails for the rest. Alone, most are not enough. They work best in combination, with evidence-based targeting, enforcement that maintains constitutional boundaries, and a foundation of public trust. Strategies thoughtfully combined by agencies with the capacity to run them and evaluate the results offer the most credible path to safety and legitimacy for communities and law enforcement alike.

All posts in this series are available here: What Works in American Policing: A Strategy-by-Strategy Assessment