How to Reduce Security Manpower Without Creating Security Gaps

Reducing security manpower safely does not start by deciding how many guards to remove.

It starts by understanding:

What security functions are currently being performed, which of those genuinely require a person, which can be automated, which can be remotely managed and which still need physical response.

The most effective approach is therefore:

Analyse the task → automate where appropriate → centralise where practical → retain people where human judgement or physical intervention is required → measure the outcome.

A building may be able to reduce manpower significantly.

Another building may discover that some posts are still operationally necessary.

The objective should not be:

minimum headcount.

It should be:

the right security resources for the actual risk and operating requirement.

Why Security Manpower Should Be Analysed by Function, Not Headcount

A security contract often begins with numbers.

For example:

1 supervisor
2 officers at the main gate
1 lobby officer
1 CCTV operator
1 patrolling officer

This tells us how many people are deployed.

It does not tell us why they are there.

Before changing manpower, management should understand what each post actually does.

A gatehouse officer may be responsible for:

• recognising residents;
• checking visitors;
• answering intercom calls;
• controlling vehicle entry;
• registering contractors;
• receiving deliveries;
• monitoring cameras;
• responding to incidents.

Some of these functions may require a person.

Some may not.

That distinction is where security transformation begins.

The Wrong Question: “How Many Guards Can We Remove?”

This question starts with the desired cost outcome.

A better question is:

What security functions must still be performed after the redesign?

If a post is removed but its functions disappear with it, the building may simply be accepting a security gap.

If the same functions are instead:

automated, transferred, centralised or redesigned,

then manpower reduction can improve efficiency without necessarily reducing security effectiveness.

Start with a Security Task Analysis

For each existing security post, list the actual tasks performed during a normal shift.

Do not rely only on the contract description.

Observe what really happens.

For example:

This immediately shows that one “guard post” may actually contain several different jobs.

Separate Security Work from Non-Security Work

This is especially important for condominiums and commercial buildings.

Security officers are often asked to perform tasks such as:

• receiving deliveries;
• issuing contractor passes;
• giving directions;
• managing parking;
• handling resident complaints;
• receiving lost property;
• answering general enquiries;
• controlling loading bays.

Some may support the overall security operation.

Others are closer to:

customer service or facilities operations.

If management does not separate these functions, it becomes difficult to determine how much security manpower is truly required.

Classify Every Task

A useful approach is to put tasks into four categories.

1. Automate

Technology can perform the function reliably.

Examples:

• credential validation;
• recording;
• door monitoring;
• selected intrusion detection;
• system-health alerts.

2. Centralise

The task still needs human involvement, but the person does not necessarily need to be at that specific property.

Examples may include:

• alarm monitoring;
• CCTV verification;
• after-hours intercom;
• escalation;
• selected access assistance.

3. Retain On-Site

The task still requires a person physically present.

Examples can include:

• immediate physical response;
• managing aggressive persons;
• complex visitor situations;
• emergency assistance;
• customer-facing duties;
• certain patrol or enforcement functions.

4. Eliminate or Redesign

Some tasks may exist simply because of an outdated operating process.

For example:

manually writing every visitor’s name into a paper log

may be replaced by a more efficient visitor process.

The function is not merely transferred.

The process itself changes.

A Simple Security Manpower Optimisation Model

The methodology can be summarised as:

Identify → Classify → Automate → Centralise → Retain → Test → Review

Identify

What does each current officer actually do?

Classify

Which tasks are security-critical, operational or customer-service related?

Automate

Which repeatable tasks can technology perform?

Centralise

Which tasks need a person, but not necessarily a person at that particular site?

Retain

Which activities genuinely require physical presence?

Test

Does the redesigned model still handle normal operations and security incidents?

Review

Are there new gaps, nuisance alarms or workload issues?

This is more robust than:

“reduce 8 guards to 4.”

Which Security Functions Are Usually Good Candidates for Automation?

Technology performs best where the task is:

• repetitive;
• rule-based;
• continuous;
• measurable.

Examples include:

Staff or Resident Access

Access-control systems can validate authorised credentials without requiring a guard to recognise every person manually.

Door Monitoring

The system can identify:

Door Forced
Door Held
Controller Offline

without requiring someone to continuously watch the door.

Intrusion Detection

Sensors can monitor selected areas continuously.

CCTV Recording

Cameras can maintain visual records across many locations.

Vehicle Access

Depending on the property, vehicle credentials, ANPR or other controlled access technologies may reduce manual gate functions.

System Fault Monitoring

The platform can identify selected equipment failures instead of waiting for an officer to notice that something stopped working.

Which Functions May Be Suitable for Remote Monitoring?

Some functions still require human judgement but do not necessarily require a dedicated person at the property.

Examples can include:

• alarm verification;
• after-hours CCTV monitoring;
• intrusion events;
• Door Forced events;
• remote intercom handling;
• selected door release;
• system-health alarms;
• incident escalation.

A central monitoring operation may support several sites.

This changes the model from:

one person continuously assigned to one screen

to:

operators responding to security events across multiple locations.

Which Functions Are Harder to Replace with Technology?

Some tasks involve physical presence, discretion or human interaction.

Examples include:

Physical Intervention

Technology can identify a person entering a restricted area.

It cannot physically stop the person.

Conflict Management

An aggressive visitor or resident may require human de-escalation.

Emergency Assistance

A person may need help during:

a medical event, fire evacuation or other emergency.

Complex Exceptions

Not every visitor or contractor follows the perfect digital workflow.

A person may still be needed to manage unusual situations.

Physical Inspection

A camera may show something suspicious.

Someone may still need to inspect the location.

This is why security optimisation should not automatically mean:

zero manpower.

Can CCTV Replace Security Guards?

Not completely.

CCTV and security officers perform different functions.

CCTV can:

• observe multiple locations;
• record continuously;
• provide evidence;
• support remote verification.

A security officer can:

• physically investigate;
• communicate;
• intervene;
• assist occupants;
• exercise judgement.

The better approach is to determine:

which tasks should be performed by CCTV and remote operators, and which require an on-site officer.

Can AI Replace Security Guards?

AI can improve detection.

For example, analytics may identify:

• a person entering a restricted zone;
• loitering;
• an object left behind;
• unusual movement;
• selected vehicle activity.

But detection is not the same as understanding.

An analytic alert may still require someone to determine:

Is the event genuine?

Is the person authorised?

Is there a threat?

Does anyone need to respond?

AI can therefore help reduce the amount of routine observation required.

It does not automatically remove the need for:

verification, judgement and physical response.

Why “One Camera Replaces One Guard” Is the Wrong Calculation

A camera and a security officer are not equivalent resources.

One camera may cover only one view.

One officer may perform:

visitor handling, incident response, patrol and customer assistance.

Conversely, one remote operator may review alerts from many cameras across multiple sites.

The correct calculation is therefore:

tasks and coverage

not:

devices versus people.

Example: Condominium Security

Consider a condominium currently using:

6 security officers across its operating model.

Their work includes:

• gate access;
• visitor registration;
• vehicle entry;
• CCTV;
• patrols;
• intercom;
• resident assistance;
• incident response.

The MCST wants to reduce recurring security costs.

The wrong approach is:

“Remove two officers and see whether the remaining four can cope.”

The better approach is to examine the tasks.

Step 1: Resident Access

If residents currently rely heavily on manual recognition by guards, controlled access can shift routine authentication to technology.

Possible tools might include:

• access credentials;
• mobile credentials;
• controlled vehicle access.

The officer no longer needs to perform every routine resident entry check manually.

Step 2: Visitor Handling

A visitor process may be redesigned using:

• pre-registration;
• intercom;
• host verification;
• visitor access credentials.

This may reduce repetitive gatehouse administration.

But the building should still determine what happens when:

the visitor is unexpected,

the resident cannot be reached,

or

the visitor behaves suspiciously.

That exception process may still require human involvement.

Step 3: CCTV Monitoring

Instead of expecting an on-site officer to stare continuously at numerous cameras, selected alarms or analytics can direct attention to relevant events.

Remote monitoring may take over some:

after-hours or event-driven surveillance.

The on-site officer can then focus on:

physical response and property interaction.

Step 4: Patrols

Do not automatically remove patrols simply because cameras exist.

Ask what patrols are intended to achieve.

Are they checking:

• doors;
• plant areas;
• carparks;
• physical defects;
• suspicious persons;
• unauthorised occupancy?

Some checks may be supported by technology.

Others may still require someone physically walking the property.

The patrol itself should have a clear purpose.

Step 5: Physical Response

Suppose the remote monitoring centre identifies:

an unauthorised person at the swimming-pool area after closing time.

The remote operator can verify the event.

But if someone needs to approach the person, the site still requires a response mechanism.

This might be:

• retained on-site security;
• mobile response;
• another authorised site resource,

depending on the property’s operating model.

The response method must be decided before manpower is removed.

Example: Commercial Office Building

A commercial building may currently deploy officers for:

• lobby access;
• tenant support;
• loading bay;
• after-hours access;
• CCTV;
• patrol;
• visitor management.

Some functions may be moved to:

access control, visitor management and remote monitoring.

But high-volume visitor interaction or loading-bay operations may still justify local personnel.

The outcome may be:

fewer routine static posts, but stronger event monitoring and targeted response.

That can be a better security model than simply reducing everybody proportionately.

Example: Industrial Site

An industrial facility may have a large perimeter and fewer public visitors.

Its manpower optimisation opportunity may therefore be completely different.

Technology may support:

• perimeter detection;
• CCTV analytics;
• vehicle access;
• remote after-hours monitoring.

But a large site may also require:

• rapid physical investigation;
• contractor control;
• safety coordination;
• emergency response.

A manpower model suitable for a residential property should not simply be copied into an industrial facility.

Static Guard Posts Should Be Challenged

One of the most useful questions is:

Why is this post here?

Possible answers include:

“To respond to visitors.”

“To watch the entrance.”

“To stop unauthorised access.”

“Because we have always had someone there.”

The last answer should trigger further analysis.

A static post should exist because it performs a meaningful function.

If technology or redesign performs that same function more effectively, the post may no longer be necessary.

But Visible Presence Can Have Value

It would also be wrong to assume that every static officer is inefficient.

Visible personnel may provide:

• deterrence;
• reassurance;
• immediate assistance;
• customer service;
• control during exceptional events.

At some properties, this may be important.

So the decision should consider both:

security function

and

property experience.

Security Manpower Reduction Should Not Be Based Only on Cost

Security is ultimately a risk-management function.

Reducing manpower may lower recurring expenditure.

But the redesign should also consider:

• response time;
• remaining detection capability;
• technology resilience;
• emergency handling;
• operating hours;
• occupant expectations;
• physical layout;
• threat profile.

Cost matters.

It should not be the only design criterion.

Detailed financial business-case modelling belongs in the wider security strategy process.

Operationally, the first question remains:

Does the redesigned security model still work?

Reduce Repetitive Human Monitoring First

One of the clearest opportunities is often:

continuous passive monitoring.

An officer sitting in front of CCTV may spend long periods watching normal activity.

A more efficient design can use:

• alarms;
• intrusion sensors;
• access events;
• analytics;
• remote monitoring

to direct attention when something relevant occurs.

The operator shifts from:

watching everything

to

investigating meaningful events.

Do Not Create Alarm Overload

Automation only helps if the generated events are useful.

If the new technology produces:

hundreds of nuisance alarms every day,

someone still needs to deal with them.

The building may simply replace:

inefficient manual monitoring

with

inefficient alarm clearing.

Before manpower is reduced, alarm quality should therefore be tested.

A False Alarm Has a Labour Cost

Every false or nuisance alarm can consume:

• operator time;
• supervisor time;
• on-site response;
• investigation;
• reporting.

A system producing constant false alarms can undermine the business case for remote monitoring.

More importantly, operators may become desensitised to alarms.

Therefore manpower optimisation depends partly on:

high-quality detection and well-tuned alarm logic.

Verification Can Reduce Unnecessary Dispatches

Suppose a perimeter sensor activates.

Without verification:

security response is dispatched every time.

With CCTV verification:

Event A

Operator sees:

maintenance contractor in authorised area.

No security dispatch required.

Event B

Operator sees:

person climbing the perimeter fence.

Immediate escalation.

Verification allows response resources to be used more intelligently.

Remote Monitoring Does Not Eliminate Response

This is one of the most important points.

Remote monitoring can:

• detect;
• verify;
• communicate;
• escalate.

But if the event requires:

physical intervention

someone needs to attend.

A manpower-reduction design must therefore answer:

Who responds?

Possible models include:

• retained on-site security;
• mobile response;
• facilities personnel for selected non-security issues;
• emergency services where appropriate.

The exact model depends on the site.

Response Time Matters

A remote security model may detect an event faster than a purely physical patrol.

But if the nearest physical response is too far away, the overall security outcome may still be weak.

Therefore the design should consider the full chain:

Detection time + verification time + escalation time + response travel time

rather than looking only at:

how quickly the alarm appeared.

Night-Time Security May Be Different from Day-Time Security

A property does not necessarily require the same security model 24 hours a day.

During the day:

• occupants are present;
• management is available;
• reception may be staffed;
• contractors are active.

At night:

• fewer people are present;
• access events are less frequent;
• intrusion becomes more significant.

A site may therefore use:

different manpower and monitoring arrangements by time period.

This can be more efficient than applying one fixed security model continuously.

Weekends and Public Holidays Can Also Be Different

Similarly, security needs can vary by:

• weekdays;
• weekends;
• holidays;
• event periods;
• construction periods.

An intelligent security model should reflect actual operations.

For example:

a commercial office may require fewer on-site resources during weekends while relying more heavily on:

remote monitoring and controlled access.

Manpower Reduction Should Be Phased

A major mistake is changing everything in one step.

A safer transformation may be:

Phase 1

Install and stabilise technology.

Phase 2

Run new technology alongside existing manpower.

Phase 3

Measure alarms, visitor volumes and response requirements.

Phase 4

Adjust procedures.

Phase 5

Reduce or redeploy selected posts.

Phase 6

Review performance.

This provides real operational data before further changes are made.

Parallel Operation is Valuable

For a transition period, the building can compare:

what the guard detected

against

what the technology detected.

This can reveal:

• missed alarms;
• false alarms;
• operational exceptions;
• unexpected workload;
• technology limitations.

It gives management greater confidence before changing the final manpower model.

Redeploy Before You Remove

Sometimes the correct answer is not:

remove the officer.

It may be:

use the officer differently.

For example:

An officer who previously spent most of the shift watching CCTV may instead perform:

• incident response;
• selected patrol;
• visitor exceptions;
• resident assistance;
• contractor control.

Technology absorbs the repetitive observation.

The person concentrates on higher-value work.

Technology Can Make Remaining Officers More Effective

Good security technology should provide better information to the person responding.

Instead of:

“Something happened at the carpark.”

the officer might receive:

Perimeter intrusion, North Carpark, Camera 14, individual moving toward Block C.

The same officer becomes more effective because the technology improves:

situational awareness.

Centralisation Can Reduce Duplication Across Multiple Sites

Consider a property group with:

10 locations.

If every site has one person dedicated primarily to monitoring alarms, the organisation is effectively maintaining:

10 separate monitoring functions.

A central operation may handle selected events across all sites.

This does not mean every site loses on-site security.

It means certain tasks are consolidated where that improves efficiency.

This model is particularly relevant for:

• property portfolios;
• retail chains;
• offices;
• industrial groups;
• multi-building developments.

Do Not Centralise Everything

Some decisions require site knowledge.

For example:

Is that contractor normally allowed into this loading bay?

Why is this resident requesting unusual access?

Is there a special event occurring tonight?

Remote operators need:

good information, procedures and access to site contacts.

Centralisation works best when local and remote operations are connected.

Define Site Escalation Clearly

For every monitored property, the remote team should know:

Who is the first contact?

Who is the backup contact?

Who may authorise access?

When should on-site response be dispatched?

What constitutes an emergency?

What happens if nobody answers?

If escalation depends on improvisation, the managed model will be inconsistent.

Use Escalation by Risk

Not every alarm requires the same response.

For example:

This is illustrative.

Each site’s procedures should be designed around its risk and operating environment.

System Failure Must Be Included in the Manpower Plan

Suppose manpower is reduced because remote monitoring takes over selected functions.

What happens if:

Internet connectivity fails?

CCTV is unavailable?

access control fails?

monitoring platform is offline?

The building needs a degraded mode.

For example:

temporary additional patrols, local monitoring or manual procedures

may be required until the technology is restored.

Manpower optimisation without failure planning creates dependency risk.

Technology Availability Becomes More Important as Manpower Reduces

If a building relies heavily on people, a single camera failure may have limited operational impact.

If manpower is reduced because CCTV now performs an important monitoring role, that same camera may become critical.

Therefore:

the more the operating model depends on technology, the more important maintenance and resilience become.

This is why manpower reduction and technology maintenance cannot be considered separately.

Security Equipment Should Have Operational Priority

Not every fault should receive the same response.

If a camera supporting the only entrance verification view fails, its operational impact may be high.

If an auxiliary overview camera fails, the impact may be lower.

As manpower is optimised, management should identify:

which technologies have become critical to the new operating model.

Those assets may require:

faster maintenance response, redundancy or temporary mitigation.

Do Not Forget Cybersecurity

Greater use of connected security technology creates stronger dependence on:

networks, servers and remote connectivity.

A manpower transformation therefore should not create a security model that is:

physically strong

but

digitally fragile.

Cybersecurity and infrastructure resilience need to support the new operating model.

The detailed engineering belongs in PATH’s separate Technical & Engineering Guides.

What Should an MCST Review Before Reducing Security Manpower?

For an MCST, useful questions include:

1. What does each current security officer actually do?
2. Which functions are security-critical?
3. Which are primarily concierge or property-management tasks?
4. Which functions can technology perform?
5. Which functions can be monitored remotely?
6. Which functions still require someone physically on-site?
7. What happens when a resident needs immediate assistance?
8. How will visitor exceptions be handled?
9. Who responds to a verified security event?
10. What happens if the technology fails?
11. What happens if Internet connectivity fails?
12. Is the technology maintained to the required standard?
13. Will residents experience a major service reduction?
14. Can the transition be implemented gradually?
15. How will management know whether the new model is working?

What Should a Managing Agent Review?

A Managing Agent should focus heavily on practical operations.

Ask:

• Who handles visitors?
• Who handles contractors?
• Who manages delivery exceptions?
• Who receives alarms?
• Who contacts residents or tenants?
• Who dispatches a responder?
• Who handles security-system faults?
• Who coordinates with maintenance?
• Who reports incidents?
• Who reviews recurring problems?

A design that looks efficient on a spreadsheet may fail if these everyday responsibilities are not assigned.

What Should a Security Manager Review?

Security managers may look deeper at:

• threat coverage;
• alarm priorities;
• verification capability;
• response time;
• contingency arrangements;
• system availability;
• incident trends;
• degraded modes;
• operator workload;
• escalation procedures.

The security question is:

Does the revised operating model still manage the identified risks?

How Should Manpower Reduction Be Measured?

Do not judge success only by:

number of officers removed.

A better assessment looks at several outcomes.

Security

• incidents;
• missed events;
• security breaches;
• response quality.

Operations

• alarm volume;
• false alarms;
• visitor issues;
• workload;
• response time.

Technology

• availability;
• critical faults;
• recurring failures.

Service

• occupant complaints;
• resident or tenant experience;
• unresolved issues.

Cost

• manpower;
• technology;
• monitoring;
• maintenance;
• response.

A cheaper model that performs poorly is not optimisation.

Watch for These Warning Signs After Reduction

Management should pay attention if:

• alarms remain unresolved;
• remote operators are overwhelmed;
• residents complain about access;
• officers cannot leave their posts to respond;
• excessive false alarms occur;
• response times increase;
• technology faults remain outstanding;
• staff begin bypassing security procedures;
• incidents increase in areas previously covered by personnel.

These may indicate the revised model needs adjustment.

The Goal Is Security Productivity

A useful way to think about the objective is:

security productivity.

That means improving the security outcome produced by each:

• person;
• camera;
• sensor;
• system;
• operating dollar.

A security officer should spend less time performing tasks that technology can do better and more time performing tasks where human presence matters.

Technology should generate useful information rather than unnecessary alarms.

Remote monitoring should concentrate specialised operators where their attention creates value.

The whole operation becomes more productive.

A Practical Example: From 8 Security Officers to a Hybrid Model

Consider a hypothetical building using eight officers across its security operation.

The current duties include:

• access checking;
• visitor management;
• CCTV monitoring;
• patrols;
• loading bay;
• intercom;
• incident response.

A transformation does not begin by saying:

“Our target is four officers.”

Instead:

Step 1 — Assess

Identify all tasks and periods of activity.

Step 2 — Introduce technology

Improve:

• access control;
• visitor handling;
• intrusion detection;
• CCTV;
• intercom.

Step 3 — Centralise selected functions

Move:

• after-hours alarm monitoring;
• selected CCTV verification;
• technical alarms

to remote operations.

Step 4 — Redesign remaining posts

Retain personnel for:

• physical response;
• high-volume periods;
• resident/tenant interaction;
• loading operations;
• emergencies.

Step 5 — Measure

Review:

• incident volumes;
• false alarms;
• response times;
• technology uptime;
• user feedback.

Only then should management determine the final manpower requirement.

The final answer might be:

six officers.

It might be:

four.

It might remain:

eight, but with different roles.

The correct number is the one supported by the operating requirement.

Common Mistakes When Reducing Security Manpower

Starting with a headcount target

This risks designing around cost rather than security function.

Assuming technology is automatically cheaper

Technology introduces:

capital cost, licences, connectivity, maintenance and lifecycle replacement.

Buying technology before redesigning the operating process

The building may simply end up with:

new systems + same old workflow.

Removing physical response capability

Remote monitoring can verify an incident but cannot physically intervene.

Creating excessive alarms

An overloaded remote operation can perform worse than the original model.

Ignoring customer-service functions performed by guards

This is especially common in condominiums.

Removing posts before technology is stable

This makes the first real incident the commissioning test.

No contingency for network or power failure

The new model becomes too dependent on technology.

Keeping inefficient posts because “we have always done it this way”

Historical practice is not the same as current security need.

Measuring only cost savings

Security effectiveness and service levels must also be considered.

A Better Decision Framework

For every current security task, ask:

This creates a rational basis for the final operating model.

Conclusion

Reducing security manpower safely is not an exercise in replacing people with cameras.

It is an exercise in redesigning the security operation.

The starting point should be:

What security work needs to be performed?

Then decide whether each function should be:

automated

centralised

retained on-site

or

redesigned entirely.

Technology can remove repetitive work.

Remote monitoring can centralise selected security functions.

On-site personnel can focus on:

physical response, judgement and human interaction.

The strongest model is therefore not necessarily the one with the fewest people.

It is the one where:

every person and every technology component has a clear purpose.

For building owners, MCSTs, Managing Agents and security managers, that is the real opportunity:

reduce unnecessary manpower without reducing the building’s ability to detect, verify and respond to security events.