To say that opinions about self-checkout (SCO) are mixed is an understatement. While retailers love the labor cost savings and claim the self-service kiosks are more convenient for shoppers, they are also concerned about the loss associated with SCO. Some consumers complain about the lack of customer service and the long lines that can form when shoppers must clumsily scan their own items, and others enjoy not having to interact with or wait for an employee to check them out. The debate is so hot that multiple Reddit threads are dedicated to the topic.
One thread about a Gizmodo article titled “The Self-Checkout Nightmare May Finally Be Ending” provides an interesting glimpse into the psyche of some shoppers:
“I’ll be fine with self-checkout when they give me the employee discount for using it,” wrote one person. “Why should I do work for the company while they cut labor and still raise prices? Should I swing around the back and see if they need help unloading trucks too?”
“My only self-checkout nightmare is if they get rid of it and I have to stand in a 45-minute line to buy s**t because grandma is counting pennies or Methany is arguing about her 7-year-old coupon so she can save 14 cents on Altoids,” wrote another.
Continue to peruse the site, and you’ll find another thread asking how many people steal at self‑checkouts and yet another full of retail employees discussing how they knowingly allow people to not pay at SCO kiosks. The general consensus is that they “do not get paid enough” to deal with potentially irate customers or to even care about the losses their employers face.
“People have strong feelings about self-checkout, no matter what ‘side’ they fall on,” said Johnny Custer, CFI, senior director of retail risk solutions at ThinkLP. “It seems like everyone is impacted—both the retailers and the public at large. You can talk to almost anyone that does any brick-and-mortar shopping and you can be rest assured that they will have an opinion, and will be more than willing to share it, along with some stories of their misadventures while ringing themselves out. This is why I call [retailers’ use of SCO] the ‘Great Self-Checkout Experiment.’ Today, it continues to remain controversial. With brick-and-mortar constantly battling the ease of online shopping, can brick‑and‑mortar afford to continue supporting this controversy? Only time will tell.”
Because of all of these factors, loss prevention professionals have mixed feelings about the use of self-checkout in stores as well. LP Magazine recently polled our LinkedIn followers, asking whether self-checkout has created more advantages or disadvantages in their stores. Of the 205 who voted, 53 percent said it had created more disadvantages, 7 percent said more advantages, 39 percent said it had created both advantages and disadvantages, and 1 percent said neither. A follow-up poll asked followers whether they feel that retailers will be increasing, decreasing, or keeping the same amount of self-checkout kiosks in the near future; 61 percent said they would be decreasing the number of kiosks, 23 percent said they would be increasing kiosks, 14 percent said they would be keeping the same number of kiosks, and 2 percent said they were not sure.
So, how did we get here? Let’s dive in.
The Evolution of SCO
The first self-checkout kiosk was installed in a Kroger store in Atlanta in July 1986, after David R. Humble, president of electronics company CheckRobot, grew tired of waiting in long checkout lines. However, many hurdles had to be crossed before the technology gained popularity among retailers.
“The concept [of self-checkout] didn’t gain widespread adoption until later in the ’90s when larger retailers started to implement their own versions of self‑checkout systems,” Custer said. “One of the first major retailers to adopt self‑checkout on a large scale was Walmart in the early 2000s. Since then, self-checkout has become increasingly common in retail stores around the world.”
According to Custer, the concept of SCO has remained relatively the same in both theory and execution since its inception, enhanced by advances in technology assisting with setup and monitoring.
“Rather than only being given the abilities to ‘self-scan’ and ‘self-pay,’ now customers have many sales capture activities that they can conduct—scanning items, hand keying, departmental code entry, scanner guns, product lookup entry, weighing items, and even entry of scannable coupons may be allowed,” Custer added. “However, sales-reducing activities (SRA) are locked down even more today than they were a few years ago. A front-end supervisor is often required to enter a code or flick a key to produce an SRA, such as a discount, price override, voided item, etc. This requires the supervisor to be always in the immediate area, responding to flashing register lights and other calls for help.”
“I can think back to Safeway in the ’90s when they were trying different kinds of SCO systems using a scan-and-go system, but it sort of faded away partly because of issues around the technology,” said Adrian Beck, emeritus professor at the University of Leicester and academic adviser for the ECR Retail Loss Group. “It’s only really in the last ten years I’d say that the use of SCO in grocery has really grown quite quickly for a number of reasons: One is the reliability of the tech has improved, and second is that grocers really saw the labor savings, so it’s a massive opportunity to significantly reduce their labor cost—that’s what’s driven a huge amount of expansion. More recently, what we’re seeing are a range of variants of SCO, so there isn’t just one type anymore. I have a list of at least twelve different types of SCO systems that exist, so it’s now becoming much more fragmented as a technology.”
CONTROLTEK President Tom Meehan echoed that the evolution of self‑checkout systems has been significant since they were first introduced.
“Originally designed to reduce labor costs and improve customer throughput, these systems were initially basic and not very user-friendly,” he said. “However, technological advancements over the years have greatly enhanced their functionality. Modern self-checkouts feature intuitive touch screens, advanced scanning technology, and reliable weight sensors, making them more accessible and easier to use across various retail settings beyond grocery stores, such as pharmacies and fast food outlets.
“The integration of mobile payment options like Apple Pay and QR codes has further improved these systems, catering to a consumer preference for contactless transactions—a trend that became particularly prominent during the COVID-19 pandemic. Additionally, recent implementations of artificial intelligence and machine learning have addressed challenges such as error reduction and theft prevention while also personalizing the shopping experience with targeted suggestions based on customer purchases.”
But while the technology itself has improved, retailers’ awareness of its association with loss has grown. Beck explains it in three phases:
At first, there was no sense that SCO generated any loss, Beck said, so the technology was shrink-neutral. “This was partly a model pushed by those who wanted to sell SCO to retailers,” he explained. “They pedaled the line that it was neutral in terms of loss, so for quite a while, few retailers thought it had an impact.”
The second phase saw a growing realization that SCO was associated with loss. In the third phase, which we are living through now, Beck says there is a very clear recognition that there is, indeed, potential loss associated with SCO, and many companies are working on strategies to manage that loss.
“How retailers deal with the third phase varies, and some companies are far more mature than others in how they manage SCO, but probably for the last four or five years, a significant portion of the industry is more aware of the risks with SCO,” Beck added.
The State of SCO Today
It could be said that the use of self‑checkout in retail is hitting a bit of a tipping point these days. Yes, the technology has advanced to be easier for customers to use and better at detecting fraud. But now that retailers are realizing just how much of an impact it has on loss, customers are growing frustrated with the lack of customer service, and SCO supervisors are facing the brunt of those frustrations with verbal and even sometimes physical attacks, something has got to give. And as a result, some retailers—including Walmart, Safeway, and Five Below—are scaling back on their use of the technology.
“During the past year, Walmart chose to convert the self-checkout lanes to staffed checkouts at select locations,” Walmart Global Communications Senior Director Charles Crowson told LPM. “The conversions were based on several factors, including feedback from associates and customers, shopping patterns, and existing business needs. We believe the changes will improve the in-store experience and give our associates the chance to provide more personalized service. As we do with many of our processes, we’ll continue evaluating our stores and listening to those who shop and work with us when considering future decisions.”
According to a Global Study on Self‑Checkout in Retail commissioned by the ECR Retail Loss Group and conducted in 2022 by Beck, SCO systems accounted for as much as 23 percent of retailers’ unknown store losses, with malicious losses representing 48 percent. In addition, two-thirds of survey respondents believed the challenge of SCO-related losses was becoming more of a problem in their business.
“There’s a lot of talk about ORC, but I often compare it to if you have an ORC gang come into the store and steal $3,000. It will make the news, but if you have 3,000 customers stealing a dollar at the SCO, it adds up to the same amount of loss—even though it doesn’t get the headlines, it’s costing retailers just as much if not more,” Beck said. “You’re vastly increasing the pool of people with opportunities not to pay. It’s a big issue in retail, and they’re now waking up to the fact where it’s a tipping point, and they have to reflect on how they want to manage their SCO.”
Custer believes that now that we’ve hit this tipping point, more retailers will start scaling back on SCO.
“I would guess that very few retailers have implemented SCOs because they believe customers like ringing themselves up,” Custer said. “Most likely, this is a cost-saving decision, allowing a single associate to cover five or more registers at once. I think many customers hate the additional ‘touches’ of ringing and bagging their own items, especially when we, as a nation, are paying significantly more for significantly less. As a result, I just don’t believe any retailer is fully committed to making SCOs the future of retail POS.
“We see constant changes in the permissions/abilities of customers while ringing, ongoing changes to the count of SCOs at any given location, and continuously evolving LP best practices like receipt checking at the SCO exit point. It all just feels that, as an industry, we aren’t fully committed and are awaiting a more AI‑based long-term solution.”
In a world where the customer is (usually) right, consumer preference will likely lead retailers’ future SCO strategies.
“You’re bringing in something customers absolutely hate, and we saw that in the mid-period of SCO’s adoption, where it was driving people insane,” Beck said. “Retailers are working in an extremely competitive world and need to rebalance how they can ensure customers have a reasonably good experience. We did a study on SCO supervisors a year ago, and one thing that came out of that was a realization of the level of violence and verbal abuse they face, especially depending on the number of machines they must look after. There’s a reevaluation for retailers to protect their staff, and make sure customers can get through that space comfortably and accurately.”
This reevaluation doesn’t necessarily mean completely doing away with SCO, though—just reimagining its use.
“Some retailers got into SCO when they shouldn’t have—when you’ve got stores where you’ve only ever employed one to two staff and therefore make really low guardianship spaces in the first place and then add the opportunity of SCO,” Beck explained. “For bigger retailers, what we’re seeing is some are reimagining what their SCO space needs to look like and beginning to think through what a sensible SCO strategy is, dependent on the type of store in which it’s placed. When you tailor SCO to the risk of a particular store and manage those losses, it becomes a viable proposition because there are still labor savings to be had. But it’s a rebalancing now of looking at the return on investment and whether it’s as marvelous as SCO vendors promised years ago. Still, I think it’s here to stay—retailers have invested too much money and don’t want to go back to the labor models of the past, but they have to invest far more in control. The days are gone where one person is looking after ten-plus SCO machines.”
Custer reiterated that he believes SCO is on its way out unless drastic changes are made.
“I do not think SCO is long for the world,” he said. “Of course, I’ve been saying that for the past five years, so I could very well be wrong. Based on retailers’ continuously changing SCO policies and approaches, I believe that many may revert back to human cashiers before too long. Or, if sensor-fusion technology proves itself worthy, perhaps we will see more of those cashier-less stores we’ve been hearing about for years.”
Holding onto Hope
While the industry is sitting at this tipping point, what retailers choose now will determine whether SCO continues to be an effective cost savings tool.
The biggest change that needs to be made, it seems, is around guardianship.
“When you look at the early days of SCO, the staff who were considered the poorest performers were put in the SCO area, and the reality now is that, for some retailers, as much as 80 percent of all transactions are going through the
SCO—that’s a phenomenal amount of profit for the least able person to be looking after it,” Beck said. “You need the very best people to work in those spaces.”
The other piece of this puzzle is increasing the number of people who work in SCO areas. Beck recommends accounting for the volume and velocity of transactions and the type of alerting being used in the self-checkout when determining how many supervisors to have working at once.
“Having a store associate present and aware at self-checkout stations during the checkout process deters what would appear as a free pass to leave the store without scanning items or not paying for the entire basket,” said Senior Manager of Zebra Technologies’ Global Solution Center Andrew Doorty. “Noticeable security cameras at the self‑checkout station display live videos of the shoppers on the POS, which inform anyone thinking of cheating that they are being monitored. Also, exit greeters auditing the basket to the receipt are a major deterrent.”
Receipt checking must be executed properly to avoid lawsuits, though.
“Just like anything else LP-related, consistency is key,” Custer said. “You can’t initiate a policy like receipt checking as customers leave the SCO area and only check certain individuals. If this is going to be a policy, then everyone needs to be checked. In addition, LP teams need to be especially vigilant with their companies’ required steps before making an external apprehension. Things can get very convoluted at the SCO, often taking away the clarity on whether a crime was committed or if intent was present. Following company protocol 100 percent should help to minimize the potential for litigation.”
Investing in better technology can also help reduce loss at the self-checkout.
“Well-trained and engaged SCO attendants—and adequate staffing presence of attendants, a ratio of one attendant for four SCO registers—along with the usage of AI is essential,” said Mike Lamb LPC, owner of MLamb Consulting Services.
Meehan believes the most notable improvements in SCO have come from the integration of artificial intelligence.
“AI has had a major impact on reducing theft and fraud at self-checkout stations through various technologies,” Meehan said. “Computer vision systems now help identify deviant behavior using video surveillance, while advanced algorithms provide early detection of deceptive actions.”
RFID, too, can be a useful tool for self-checkout.
“RFID at self-checkouts can eliminate mis- scans and many common scams because each tag has a unique serialized number that links to an item and its appropriate UPC,” Meehan explained. “Additionally, RFID allows you to track what enters and exits the self-checkout area, providing valuable insights into where and when shrinkage is occurring.”
To truly make an impact on loss at the SCO, RFID could go even further than the self-checkout area.
“Better monitoring systems for a shopper’s trip will preload what is expected to be scanned at the self‑checkout lane,” Doorty said. “The audit of a shopper’s basket needs to be streamlined. Utilizing passive RFID technology, especially for higher value items, will help retail operations identify items that are moving past the POS. Exit greeters will have RFID-enabled mobile computers that can easily scan a basket and identify items not purchased. Also utilizing computer vision at the exit will provide a method for speedy audits.”
Comprehensive data collection while testing different technologies is essential in determining whether they are worthy investments, according to Custer.
“Long-term data and full visibility into what is actually happening at the SCOs is probably most important,” Custer elaborated. “Only through that data will you be able to fully monitor the ongoing ROI of SCO use, and only through the data will you be able to make educated decisions on how to move forward.”
Ultimately, it’s a combination of all these considerations that are necessary for an effective SCO program.
“Like many other shrinkage challenges, self-checkout requires a comprehensive approach to reducing or mitigating losses,” Meehan advised. “This involves strong customer service, appropriate training and education for staff, effective signage, the placement of physical security devices such as cameras, and regular maintenance of the self-checkout devices themselves. One of the challenges with self-checkout is that it has been controversial from the start, which makes mitigating shrinkage even more cumbersome due to opposition from those who are against self-checkout systems. Additionally, continually testing new technologies and seeking feedback from peers in the industry can provide valuable insights and innovative solutions. To truly address shrinkage at self-checkouts, I believe it is essential to implement all of these measures.”
Thieves will always find a way to steal, Doorty said, but we as an industry can at least do everything in our power to thwart their efforts.
“People who are going to steal are going to probe retailers’ processes and look for easy ways to steal—when one avenue to steal is closed down, they will find another avenue,” he said. “I believe a better staff-managed self-checkout environment and the latest technology will help prevent theft.”
The advantages of SCO, especially with the labor shortage retailers are facing today, is undeniable. Only time will tell whether the use of self‑checkout will continue to grow—or die away.
“I believe there will always be the balance of labor savings versus the cost of shrink, and I doubt we’ll see a day where SCO no longer exists,” Mike Lamb concluded. “Arguably, there will just be more of a critical view of how and where the technology is deployed. In the era of the customer voting with their wallets, they will dictate how we see it evolve going forward.”