A customer buys a few bread rolls, two tomatoes, and a banana. Instead of searching through a long product list, the cashier immediately receives a suggestion generated from a camera image. A few checkout stations away, a beverage passes through the scanner, but the barcode is not read. The system detects a discrepancy between what the camera sees and what has been registered on the receipt.

Not long ago, solutions like these were associated mainly with self-checkout systems. Today, more and more retailers want to use them at traditional cashier-operated checkouts as well. The reason is simple. Technologies originally designed to improve the customer shopping experience are proving equally effective at supporting cashiers, reducing losses, and speeding up service.

From Two Separate Worlds to One Ecosystem

For many years, traditional checkouts and self-checkouts evolved almost independently of one another.

At traditional checkout lanes, the cashier played the central role. They identified products, scanned items, and assisted customers. Technology primarily served as a supporting tool.

In self-checkout areas, some of these responsibilities were transferred to customers. This is where solutions based on cameras, image analysis, and artificial intelligence first began to appear. Their purpose was to streamline shopping, reduce errors, and improve control over the sales process.

At first, it seemed these technologies were designed exclusively for self-checkout environments. In practice, however, retailers quickly realized that the challenges on both sides of the checkout counter are remarkably similar.

Regardless of the checkout type, retailers want to:

  • Reduce losses
  • Speed up customer service
  • Minimize mistakes
  • Improve employee comfort
  • Increase checkout area efficiency

This is why solutions developed for one part of the store are increasingly finding applications in another.

The Less Discussed Problem: Losses at Traditional Checkouts

When retail losses are discussed, the conversation usually focuses on shoplifting or irregularities at self-checkout stations.

I have previously written about how the introduction of self-checkouts is not the root cause of retail shrinkage.

Far less attention is paid to situations that can occur at traditional checkout lanes. Yet errors leading to revenue loss can occur there as well.

Some result from haste, distraction, or fatigue. Others stem from product scanning issues.

Imagine the morning rush in a grocery store. A queue forms at the checkout, and the cashier is processing customers at a rapid pace. Among the products is a multipack of beverages. The item passes through the checkout area, but the barcode is not successfully scanned. In a busy environment, this may go unnoticed, and the product never appears on the receipt.

In practice, there are also situations where a product is passed across the scanner with the barcode facing away from the reader, or the barcode is partially covered by a hand. The outcome is the same—the product physically leaves the store but is never recorded in the sales system.

Similar situations can involve cosmetics, confectionery, beverages, or other small products. A single incident may seem insignificant, but across thousands of daily transactions, even a small percentage of such events can generate noticeable financial losses.

An additional challenge is that these situations are not always easy to detect through standard audits and controls.

As a result, more retailers are looking for tools that can better monitor the checkout process without disrupting employees’ daily work.

Self-Checkout Technologies Move Beyond Self-Service Areas

One of the key development directions has been camera-based image analysis supported by artificial intelligence algorithms.

In self-checkout environments, such solutions have been used for years. Their purpose is to compare what the camera sees with what has been registered by the point-of-sale system. If a discrepancy appears, the system can flag the transaction for additional review.

In practice, this creates an additional layer of verification based not only on barcode scanner data but also on the actual visual image of the product at the checkout.

The effectiveness of these solutions has led retailers to recognize their potential at cashier-operated checkout lanes as well. After all, a camera does not distinguish whether a product is being scanned by a customer at a self-checkout or by an employee at a traditional register. If technology can detect discrepancies in one environment, it can support the process in another.

Industry analyses show that computer vision systems are now being used not only for loss prevention but also for queue management, customer behavior analysis, and workforce optimization.

Faster Service Without Searching for Products on a Screen

Loss prevention is only one side of the story.

Another area increasingly attracting retailers’ attention is accelerating the customer checkout process.

Every cashier is familiar with situations where customers present products without barcodes.

These most commonly include:

  • Fruits
  • Vegetables
  • Bakery products
  • Weighed items

For example, a customer buys several types of bread, an avocado, onions, and tomatoes. In a traditional workflow, the cashier must manually search for the appropriate items in the POS system. In stores with a large assortment of fresh products, finding the correct item may take several extra seconds.

At first glance, this may not seem like a major issue. However, during peak hours, when a dozen people are waiting in line, every additional second impacts queue length and the overall shopping experience.

This is where image analysis once again becomes valuable.

Using a camera, the system can recognize the product placed in front of the lens and automatically suggest the most likely matches to the cashier. Instead of navigating an extensive product list, the employee receives an immediate recommendation.

This does not replace human decision-making. The final selection still belongs to the cashier. The technology simply helps identify the correct item faster and reduces the number of clicks required to complete the transaction.

The benefits are noticeable for both employees and customers. Cashiers can focus on customer interaction rather than searching product databases. Customers spend less time waiting in line. Retailers increase checkout throughput and can serve more customers within the same period.

This is a clear example of how a solution originally developed for one process can deliver measurable benefits in an entirely different area of store operations.

Why Are Retailers Asking About These Solutions Now?

Just a few years ago, image-analysis technologies were viewed mainly as interesting add-ons for self-checkout systems. Today, they are increasingly becoming part of retailers’ operational strategies.

There are several reasons for this.

Growing Pressure to Reduce Losses

Retail chains are paying closer attention to every factor affecting profitability. Even a small percentage of incorrectly registered products can translate into significant losses across an entire network.

As a result, interest is growing in solutions that help identify potential irregularities without increasing the number of audits or requiring additional staff.

The Need for Faster Customer Service

Today’s customers expect efficient shopping experiences regardless of whether they use self-checkout or a cashier-operated lane.

Any technology capable of reducing transaction times by even a few seconds can have a meaningful impact during peak shopping periods. Across an entire day, this translates into shorter queues and higher checkout throughput.

Increasing Acceptance of Artificial Intelligence

Not long ago, many businesses approached AI with considerable caution. Today, retailers are primarily interested in practical applications that solve real operational challenges.

If technology can help reduce losses or speed up customer service, it is no longer viewed as an experiment but as a tool that supports everyday store operations.

The Technologies Have Already Proven Themselves

Perhaps the most important reason is even simpler. Image-analysis solutions have already demonstrated their value in self-checkout environments.

Retailers are no longer asking about future technologies. Increasingly, they are asking how proven solutions from one area of the store can deliver similar benefits elsewhere.

Why Are the Boundaries Between Checkout Types Starting to Blur?

For many years, retail technology evolved separately for different checkout formats.

Today, it is becoming increasingly clear that this distinction is losing relevance.

Whether customers use self-checkout or are assisted by a cashier, the objectives remain the same:

  • Faster service
  • Accurate product identification
  • Reduced losses
  • A smoother shopping experience
  • Greater operational efficiency

That is why image-analysis technologies are increasingly viewed not as solutions for a specific checkout type, but as tools that support the entire sales process.

In the coming years, the boundaries between different store zones are likely to continue fading. Solutions developed for one area will increasingly be applied wherever they can deliver tangible business value.

The evolution of checkout technology demonstrates that innovation rarely remains confined to a single place. Solutions that initially helped self-service customers are now increasingly supporting cashiers and store managers as well.

On one hand, they help reduce losses caused by incorrectly registered products. On the other, they accelerate the identification of barcode-free items and improve daily workflows for store employees.

This illustrates that the future of retail is not about choosing between traditional and self-checkout systems. Increasingly, it is about applying the same technologies wherever they can create the greatest value – regardless of who is using them.