Retail leaders and asset protection directors face a complex trade-off in 2025. Self-checkout kiosks and unattended payment terminals are essential for managing labor costs and increasing throughput, but they introduce substantial vulnerabilities. The National Retail Federation reports that retail shrinkage surpassed $112 billion recently, with self-checkout environments contributing disproportionately to these losses (Source: National Retail Federation).

For asset protection teams, the pain point is no longer just about catching shoplifters; it is about managing a complex ecosystem of disconnected security systems, combating organized retail crime (ORC), and reducing the operational drag of manual investigations. Traditional video systems that only record footage for post-incident review are insufficient for the speed at which modern retail fraud occurs.

This guide explores how to monitor unattended checkouts and kiosks to lower shrink by leveraging video AI, integrating point-of-sale (POS) data, and adopting forward-looking operational strategies. By using cameras with video AI to assist staff, retailers can standardize security across locations and help protect margins without compromising the customer experience.

Key terms to know

• Unattended checkout: A point-of-sale system where the customer scans and pays for items without direct cashier assistance, often supervised by a remote or floating attendant.

• Sweethearting: A form of employee fraud where a cashier or self-checkout host intentionally bypasses scanning items or applies unauthorized discounts for friends or family.

• Scan avoidance: A theft tactic where a customer intentionally fails to scan an item, often by leaving it in the cart or passing it around the scanner.

• Organized retail crime (ORC): Coordinated theft conducted by professional groups who resell stolen goods, often exploiting specific vulnerabilities like unattended kiosks across multiple locations.

• Video AI: Technology that uses artificial intelligence to analyze video feeds in real time, detecting defined behaviors (like loitering or rapid movement) and sending alerts, as opposed to passive recording.

The operational reality of unattended retail shrink

The shift toward self-service has altered the loss landscape. While labor costs decrease, the opportunity for both malicious and accidental loss increases. Research indicates that nearly one in three retail shrinkage incidents now involves some form of self-checkout manipulation.

For an asset protection director managing dozens or hundreds of locations, the core frustration is often the inability to see what is happening in real time. Traditional video systems provide evidence only after the damage is done. By the time a manager reviews footage of a barcode switch or a "push-out," the inventory is gone, and the opportunity for recovery is lost.

Common loss scenarios at kiosks

• Barcode manipulation: Sophisticated offenders swap barcodes from low-cost items onto high-value merchandise. In a documented 2024 case, an organized ring stole over $100,000 in goods from a home improvement retailer using this method (Source: Security 101).

• Mis-scans and non-scans: Whether accidental or intentional, customers frequently fail to scan every item. Without real-time alerts, these errors go unnoticed until inventory audits reveal the discrepancy.

• Employee-assisted fraud: Internal theft remains a major driver of shrink, accounting for 28.8% of losses. This includes employees overriding alerts at kiosks or allowing known associates to bypass payment (Source: National Retail Federation).

Why traditional monitoring fails asset protection teams

Many retailers struggle with high shrink rates not because they lack cameras, but because their systems are reactive and disconnected. The persona of the modern Asset Protection VP is plagued by "false alarm fatigue" and the sheer volume of data that cannot be processed manually.

The limitations of legacy video systems

• Reactive capabilities: Standard camera systems only provide value after an incident occurs. They require hours of manual footage review to find a particular event, draining resources that could be used for proactive responses.

• False alarm fatigue: Older motion detection systems generate excessive noise, alerting staff to every shadow or harmless movement. This leads to staff ignoring alerts entirely, masking genuine incidents.

• Disconnected data: When video systems are not integrated with POS or inventory data, blind spots are created. A transaction might look legitimate on video, but without seeing the digital receipt data in the same view, fraud like barcode switching is tough to detect.

Operational technology for unattended checkout monitoring

To move from reactive responses to more proactive shrink reduction, retailers are deploying a layered technology approach. This involves combining physical barriers with intelligent digital systems.

How video AI cuts kiosk shrink

Video AI augments existing camera infrastructure to support the loss prevention team. Instead of simply recording, the system can flag defined risk indicators and notify staff quickly.

1. Real-time behavior detection

AI agents can be trained to recognize behaviors that precede theft. For example, Loitering detection identifies individuals who linger in high-value areas or near unattended kiosks without initiating a transaction. This allows staff to offer customer service—a proven deterrent—before a theft occurs.

2. Monitoring unattended areas

One of the specific challenges in retail is when a kiosk or service desk is left unstaffed. Spot AI’s Unattended Checkout or Desk capability monitors these specific zones. If a customer approaches a kiosk but no staff member is present, or if a transaction is initiated at a closed terminal, the system sends a real-time alert to the floor manager. This guards against opportunistic theft at unmanned stations.

3. Combating organized retail crime (ORC)

ORC groups often scout locations before striking. By using Person Enters No-go Zones analytics, asset protection teams can receive alerts when unauthorized individuals enter restricted areas like stockrooms or cash offices. Furthermore, unified dashboards allow VPs to track suspicious patterns across multiple store locations, identifying if the same group is hitting different sites.

4. Cutting investigation time

Legacy systems require hours of scrubbing through footage to find an incident. Modern Video AI platforms utilize natural language search. An investigator can simply search for "person in red shirt near self-checkout at 2 PM," and the system retrieves the relevant clips in seconds. This can cut investigation time significantly, allowing teams to resolve more cases and improve recovery rates (Source: Spot AI).

Integrating video with POS for broader visibility

The most effective strategy for monitoring unattended checkouts involves bridging the gap between physical observation and digital transaction data. Integrating video AI with Point-of-Sale (POS) systems helps teams review transactions alongside video for faster, more accurate investigations.

Transaction-video correlation

When video analytics are integrated with POS data, transactions are time-stamped and linked to the corresponding video clip. If a transaction is voided, refunded, or heavily discounted, the system flags it as an exception. An investigator can then review the video clip associated with that specific transaction swiftly to verify if the void was legitimate or a case of sweethearting.

Identifying high-risk patterns

Integrated systems allow for batch analysis. Asset protection directors can run reports to see which specific kiosks, shifts, or employees have the highest rates of voids or "item not found" errors. This data-driven approach moves loss prevention from random spot checks to targeted coaching and intervention.

Operational best practices for unattended retail

Technology is most effective when paired with strong operational protocols. Lowering shrink requires a culture of operational excellence where staff are trained and empowered.

1. Training as a deterrent

Research from the Food Industry Association indicates that customer service is one of the most effective theft deterrents. Staff stationed near unattended checkouts should be trained to engage customers, make eye contact, and offer assistance. This psychological barrier helps deter opportunistic theft (Source: FMI).

2. Strategic store layout

Design the self-checkout area to minimize blind spots. Ensure that the attendant station has a clear line of sight to all kiosks. Place high-shrink items in areas with high visibility or require them to be processed at staffed lanes.

3. Standardizing response protocols

When a video AI system triggers an alert (e.g., "Loitering at Kiosk 3"), there must be a standardized operating procedure (SOP) for how staff respond. The response should be non-confrontational, such as asking, "Did you need help finding anything?" This approach balances security with customer experience, avoiding the negative impact of aggressive loss prevention.

Comparing video analytics solutions

For asset protection leaders evaluating technology, it is crucial to select a platform that scales easily and integrates with existing infrastructure.

Spot AI enables retailers to use existing cameras with added video AI features, helping large deployments control costs.

Measuring ROI on unattended checkout monitoring

To justify the investment in video AI and monitoring technology, asset protection directors must demonstrate clear Return on Investment (ROI).

• Direct shrink savings: Retailers implementing comprehensive video analytics and integration strategies report cutting shrinkage by 15-30% within the first year. For a retailer with $50 million in revenue and 2% shrink, cutting it by 25% saves $250,000 annually (Source: Security 101).

• Labor efficiency: Automating monitoring allows retailers to lower guard spend. Instead of paying for a dedicated guard to watch a monitor ($20k+/month), AI agents support continuous monitoring at a lower cost ($1-2k/month) (Source: Spot AI).

• Investigation speed: Reducing investigation time by 50-70% saves thousands of dollars in administrative labor annually, allowing skilled investigators to focus on complex ORC cases rather than reviewing tape (Source: Spot AI).

The Path Forward for Asset Protection

Monitoring unattended checkouts and kiosks to reduce shrink benefits from a shift from reactive monitoring to more insight-driven workflows. By leveraging Video AI, retailers can uncover hidden operational risks, help deter theft, and streamline investigations.

Spot AI helps asset protection teams standardize security across shifts and sites, turning existing cameras into a coordinated monitoring system. This approach supports the bottom line and a safer, more efficient shopping environment.

See Spot AI in action—request a demo to explore how video AI can help cut shrink and deliver real-time visibility across your retail locations.

Frequently asked questions

What are the best practices for monitoring unattended checkouts?

Best practices include integrating video monitoring with POS transaction data to correlate scans with video evidence, using AI analytics to detect non-scanning behaviors in real time, and positioning attendants to maintain clear lines of sight. Training staff on de-escalation and customer service is also critical for deterrence.

How can technology help cut shrinkage at self-checkouts?

Technology helps cut shrink by automating detection. Computer vision can verify that the item scanned matches the physical product, smart scales detect weight discrepancies, and video AI alerts staff to suspicious loitering or missed scans in real time, allowing for intervention before the customer leaves.

Do AI cameras help cut shrink at kiosks?

Yes, AI cameras significantly cut shrink by detecting behaviors that traditional cameras miss, such as "scan avoidance" gestures or barcode switching. They also minimize false alarms, ensuring staff only respond to genuine security events, which improves the overall effectiveness of the monitoring strategy.

What compliance requirements exist for unattended payment systems?

Unattended payment systems must comply with PCI DSS standards to protect cardholder data. This includes encrypting data transmission, not storing sensitive card information, and ensuring physical terminals are tamper-resistant. Video systems monitoring these areas must also adhere to privacy laws regarding data retention and employee monitoring.

About the author

Sud Bhatija is COO and Co-founder at Spot AI, where he scales operations and GTM strategy to deliver video AI that helps operations, safety, and security teams boost productivity and mitigate incidents across industries.