Does a self-checkout pay off in your supermarket? What research says about shrinkage, customers and staff costs

A self-checkout is a till where the customer scans and pays for the items themselves. Research shows that supermarkets with such tills suffer more theft, and that more items leave the shop unscanned when customers scan themselves. Whether the till itself is the cause has not been demonstrated.

For an independent supermarket, the question is therefore not whether anything disappears, but whether the till hours saved outweigh the cost of the tills and the extra shrinkage. Shrinkage is stock that disappears with no sales against it. You can measure those three numbers yourself.

In this article

1. Do customers steal more at a self-checkout?
2. Is self-scanning with a handheld scanner or phone something different?
3. Why do items leave at the self-checkout without being paid for?
4. What do customers think of the self-checkout?
5. How many staff does a self-checkout save?
6. When does a self-checkout pay off for an independent supermarket?
7. How do you measure whether the self-checkout costs you shrinkage?

Do customers steal more at a self-checkout?

Supermarkets with self-checkouts suffer shop theft more often, but it has not been proven that the till is the cause. In an English government survey of supermarkets, 86% of shops with self-checkouts were victims of shop theft, against 52% of shops without (Home Office, Commercial Victimisation Survey 2014).

The researchers themselves warn that this is not proof. Of the large supermarkets, 89% had self-checkouts, of the small ones 8%. Large shops in the survey suffered theft more often and more heavily anyway, so the size of the shop may be the real reason. The survey covered England and Wales, with 187 supermarkets in the sample.

Earlier research using figures from retail chains reached a more cautious result. Based on case studies at retailers and a survey of staff, criminologist Adrian Beck concluded that the available figures showed little effect of self-checkouts on overall shrinkage (Beck, 2011). He did recommend setting up the self-checkout area as a separately supervised zone.

A review from 2016 summarises it as conflicting evidence. According to the author, there is some evidence that self-checkouts increase theft by customers, especially by customers who would not steal in any other way. Exact figures on how much extra theft there is are lacking (Taylor, 2016).

Is self-scanning with a handheld scanner or phone something different?

Yes, and that is where the loss has been measured best. With scanning while shopping, the customer scans each item with a handheld scanner from the shop or with their own phone, and pays at the end. At one large retail chain, checks found that 3.97% of the value checked had not been scanned (Beck & Hopkins, 2017).

The researchers analysed 1.096 million checks on nearly 12 million shopping trips in one year. The unexplained shrinkage of the whole chain was 1.47%. The average of the chain and two surveys of supermarkets was 1.79%. With self-scanning, the loss was therefore 122% higher.

The real loss is probably higher. A member of staff checked on average 6 items from a basket of around 30, often from the top. Anyone deliberately not scanning puts those items at the bottom. What the figures do not show is whether customers stole deliberately or forgot something. The chain also gave no figures on what the handheld scanners saved.

Why do items leave at the self-checkout without being paid for?

Through a mixture of deliberate theft, mistakes and frustration with the machine. In a British online poll of 2,634 users, 19% admitted having taken something at the self-checkout. Of them, 57% said they gave up when an item would not scan (Taylor, 2016).

Other reasons in the same poll were that the chance of being caught is smaller (51%) and that the machine is easy to fool (47%). Fruit and vegetables were named most often as what was taken (67%), followed by bread and bakery (41%). It is a poll by a commercial website and not representative. The author uses it as a starting point, not as a measurement.

In the Netherlands, 77% of opportunistic offenders in research for the Ministry of Justice and Security felt that shop theft has become easier because of the self-checkout. Shop owners see that link too. Experts in the same research question it (Letter to Parliament on shop theft, 21 April 2026). How big shop theft is in the Netherlands is covered in a separate article on shrinkage and theft in the shop.

Whether supervision on the screen helps is not yet clear. In an experiment with a simulated self-checkout, researchers tested whether a human-looking figure on the screen led to less dishonest behaviour. That was only partly confirmed (Siebenaler et al., 2019).

What do customers think of the self-checkout?

Customers use it, but their judgement of the shop depends more on the staffed till. In two studies of supermarket customers, 313 in the United Kingdom and 474 in Australia, the quality of the staffed till had a stronger effect on satisfaction with and loyalty to the shop than the quality of the self-checkout (Sharma, Ueno & Kingshott, 2020).

For an independent supermarket, that is an important point. If your regular customers come partly for the familiar faces at the till, fewer staffed tills save you wages, but you may lose something those customers come for in the first place.

Frustration with the machine is common. In a customer poll cited by Taylor, 83% were annoyed by the message about an unexpected item in the bagging area and 73% by staff who were slow to come and help (Taylor, 2016). How much waiting time customers really save at a self-checkout in a small supermarket does not appear as a reliable figure in any of the studies for this article.

How many staff does a self-checkout save?

Fewer than the number of tills suggests. The work at the till does not disappear, it changes. In interviews with 22 checkout staff in the United States, the researchers described three changes (Moradi, Levy & Cheyre, 2025).

• At the self-checkouts, several customers ask for help at the same time.
• The work is mainly about solving problems and keeping watch.
• The ordinary till becomes harder, because the easy customers go to the self-checkout.

So do not count all till hours as savings. Someone has to be at the self-checkouts to check age for alcohol, solve faults and carry out spot checks. Those hours count as costs. The staff in the study also more often found themselves in a supervisory role towards customers. They had to make extra effort to keep their dealings with customers good.

When does a self-checkout pay off for an independent supermarket?

When the staff costs saved are greater than the annual cost of the tills plus the extra shrinkage. You work that out with three numbers: the till hours you really save, what an hour of till work costs you, and how much turnover goes through the self-checkout.

The statutory minimum wage for those aged 21 and over has been € 14.99 per hour since 1 July 2026 (Rijksoverheid, minimum wage amounts 2026). The collective agreement for the food retail sector (cao Levensmiddelenbedrijf) expired on 1 April 2026 and is no longer generally binding. The hourly wage for age 21 in group A was € 14.51 in that agreement (Staatscourant 2025, 14579), so below the minimum wage. A new agreement may be under negotiation. Pension with the sector pension fund BPFL remains compulsory.

What an hour really costs you is higher than the hourly wage. Holiday pay, employer contributions, pension and paid time off come on top. How you calculate that and set it against your turnover per day is covered in a separate article on staff costs against turnover. The worked example below assumes € 22 per hour. That is an assumption, not an amount from the collective agreement.

In the example, the shop saves 20 till hours per week, after deducting the hours for supervision. That is € 22,880 per year. The tills cost € 10,000 per year in depreciation, maintenance and software. That leaves € 12,880. If € 900,000 of turnover excluding VAT goes through the self-checkout, the extra shrinkage on it may not be higher than about 1.4%.

If the difference were as large as with the handheld scanners in the study by Beck and Hopkins, 2.5 percentage points, the extra shrinkage would cost € 22,500. Then the shop loses money. With a fixed self-checkout under supervision the difference may be smaller, but the studies for this article contain no comparable measurement of that. So your own measurement is decisive.

How do you measure whether the self-checkout costs you shrinkage?

By measuring shrinkage per product group before and after the introduction, over the same period of the year. Start at least a few months before the self-checkouts are in place, so that you have a starting point. How you calculate shrinkage from stock, purchases and sales is covered in the article on shrinkage and theft in the shop.

1. Look especially at fruit and vegetables and bread, the groups that customers weigh or select themselves. Those groups were named most often in the poll as what was taken.
2. Carry out spot checks at the self-checkout and record for each check how many items you looked at and what had not been scanned. Do not only look at the top of the basket.
3. Keep track of how much turnover goes through the self-checkout. Without that number you cannot calculate the extra shrinkage as a percentage.
4. Look in the till reports at aborted transactions and corrections by staff at the self-checkout.

Bear in mind that turnover itself also changes. If your prices rise, turnover rises without you selling more, and the same shrinkage in euros looks smaller as a percentage. How you separate price and volume is covered in the article on turnover, volume and margin in the food shop.

Sources

• Home Office (2015). Crime against businesses: findings from the 2014 Commercial Victimisation Survey. London: Home Office. gov.uk
• Beck, A. (2011). Self-scan checkouts and retail loss: understanding the risk and minimising the threat. Security Journal, 24(3), 199-215. doi.org
• Taylor, E. (2016). Supermarket self-checkouts and retail theft: the curious case of the SWIPERS. Criminology & Criminal Justice, 16(5), 552-567. doi.org
• Beck, A. & Hopkins, M. (2017). Scan and rob! Convenience shopping, crime opportunity and corporate social responsibility in a mobile world. Security Journal, 30(4), 1080-1096. doi.org
• Minister of Justice and Security, Letter to the House of Representatives on the research into shop theft, 21 April 2026. Parliamentary Papers II 2025/26, 29911, no. 502. officielebekendmakingen.nl
• Siebenaler, S., Szymkowiak, A., Robertson, P., Johnson, G.I., Law, J. & Fee, K. (2019). Honesty, social presence and self-service in retail. Interacting with Computers, 31(2), 154-166. doi.org
• Sharma, P., Ueno, A. & Kingshott, R.P.J. (2020). Self-service technology in supermarkets: do frontline staff still matter? Journal of Retailing and Consumer Services, 59, 102356. doi.org
• Moradi, P., Levy, K.E.C. & Cheyre, C. (2025). Pseudo-automation: how labor-offsetting technologies reconfigure roles and relationships in frontline retail work. Proceedings of the ACM on Human-Computer Interaction, 9(2), 1-21. doi.org
• Rijksoverheid, Minimum wage amounts 2026, accessed 29 October 2026. rijksoverheid.nl
• Minister of Social Affairs and Employment, Decision declaring the collective agreement for the food retail sector generally binding, 22 September 2025. Staatscourant 2025, 14579. officielebekendmakingen.nl