What Are the Loss Prevention Industry Trends in Retail Right Now?

By awareretailgroup.com Editorial· Edited by Tom Walsh· Published 2026-10-28

A regional grocery chain reading only NRF’s 2026 headline might believe shoplifting is simply going away. These trends have current, dated data behind them: NRF’s retailer survey shows shoplifting incidents fell in 2025 while police reports from a sample of cities show it rising in the first half of 2026; more retailers report fraud, scam and return fraud increases; self-checkout carries measurably higher loss; AI theft detection is widely reported but rarely running as active computer vision; and more retailers report rising violent altercations while investing in violence-prevention training.

What NRF’s 2026 theft and violence survey found

NRF surveyed 66 retail companies representing 143 brands for its Impact of Retail Theft & Violence 2026 report, and the headline number points down: retailers reported a 12.4% decrease in shoplifting incidents and an 8.1% decline in retail merchandise theft in 2025 compared with 2024 (NRF). That is an executive survey. It reflects what loss prevention executives reported for their own companies, which can include incidents never reported to police. NRF’s report still says repeat offenders and organized retail crime activity indicate theft remains “a major concern across the nation” (NRF).

Why police data tell a different story for 2026

The Council on Criminal Justice tracks shoplifting using police reports from a sample of U.S. cities. Its mid-year 2026 update found the average reported shoplifting rate was 4% higher in the first half of 2026 than in the same period of 2025 (CCJ), after its year-end 2025 report had shown reported shoplifting down 10% for the full prior year (CCJ). The mid-2026 update, based on data through the first half of the year, calls 2026 the peak year for reported shoplifting incidents in its nine-year sample, and it says police data “almost certainly undercount total shoplifting by a significant amount” (CCJ). NRF’s same survey found 63% of retailers report fewer than half of store-related theft incidents to law enforcement (NRF). A manager quoting a theft trend should name the source and the date range, a habit covered further in is retail theft data reliable enough to cite.

Where the loss dollars actually sit

IHL produced Beyond Shrink with a vendor research partner, and it reports shares of surveyed retailers (IHL). Store theft is the headline, but IHL Group’s research puts external theft, including organized retail crime, at 25% of total retail loss (IHL), with the remaining 75% occurring in the supply chain, returns and reverse logistics, e-commerce fulfillment, and administrative or process failure. ECR Retail Loss’s 2026 benchmarking study found a related structural shift: 47% of retailers now manage loss end-to-end across stores, distribution centers and online fulfillment, and 54% of loss prevention functions now sit inside a broad or very broad remit (ECR). A store that only briefs associates on shelf-level shoplifting is training for a quarter of the problem, measured in total retail loss terms. A manager building this quarter’s training calendar should give receiving, returns and fraud scripts the same airtime as sales-floor observation.

Fraud and returns are rising while shoplifting counts fell

While shoplifting incident counts fell in NRF’s 2025 data, fraud moved the other way: retailers reported increases in phone scams (69%), loyalty fraud (51%) and gift card theft or fraud (42%) (NRF). IHL’s Beyond Shrink study found return fraud increases, reported by 68% of retailers, now nearly match the 69% reporting shoplifting increases, which it calls a sign that the returns desk is one of the least-defended doors in retail (IHL). NRF reports fraud rising and IHL reports return fraud rising, but the two measure shoplifting differently: IHL counts the revenue-weighted share of 1,421 retailers in 12 countries reporting an increase, while NRF averages the change in incidents across its 66 responding companies (IHL, NRF). A cashier script for a phone scam or a gift card request belongs next to the shoplifting talking points, as covered in how cashiers should handle gift card scam attempts.

Self-checkout: what ECR actually measured

ECR Retail Loss’s 2026 self-checkout report puts a number on the gap: losses run 33% higher in stores with self-checkout, a smaller gap than the figure reported in 2018, and the overall difference comes out to about 0.42 percentage points between stores with and without self-checkout (ECR). That gap matters more as volume grows: stores with self-checkout now run an average of 54% of transactions through it (ECR). ECR’s report cites a survey in which self-checkout hosts suggested one host per six machines was optimal, and it says the evidence shows that turning off weight scale controls increases loss (ECR). Estimates of how many non-scan events are driven by malicious intent range from 6% to 80% (ECR), which argues for training hosts on both error correction and theft cues.

Self-checkout host training and staffing

Only one study in ECR’s review linked self-checkout training to measurable loss improvement: one retailer reported shrink improving by 23 basis points after enhanced training of self-checkout hosts (ECR). That is thin evidence, and ECR itself says the sector needs more rigorous evaluation, shared openly (ECR). In IHL’s analyst coverage of Lidl putting shoppers on screen at self-checkout, Greg Buzek writes that IHL’s own study found low staffing to be “the single most controllable operational lever in the entire study” (IHL). Check self-checkout staffing before leaning on new technology; see also how to build a grocery loss prevention training program.

AI adoption claims outrun AI deployment

IHL’s Beyond Shrink survey reports 53% of retailers use AI theft detection and alerts, 51% use video analytics or computer vision, and 48% use self-checkout monitoring (IHL). IHL’s separate AI Reality Check study found only 9% of retailers run an active computer vision deployment, even though 40% already have cameras mounted in their stores (IHL). One figure counts retailers that report using AI theft detection or video analytics; the other counts retailers with computer vision actively running in stores. Buying an analytics package does not mean anyone in the store is trained to act on its alerts. Treat an AI flag as a prompt for a trained associate to check, and budget the training alongside the software.

Retailers are investing in violence-prevention training as altercations rise

IHL’s Beyond Shrink study found 72% of retailers saw violent altercations in their stores increase over the past 12 months, against 9.3% reporting a decrease (IHL). NRF’s 2026 survey found retailers investing in management training (72%), employee training (65%) and risk-intelligence technologies (61%) to prevent workplace violence (NRF). California makes it law: SB 553 amended Labor Code section 6401.7 and added section 6401.9, which requires employers to provide training to employees about workplace violence prevention with “training material that is appropriate to the employees’ educational level and language” (Cal/OSHA), with subdivisions (b) through (g) of section 6401.9 operative since July 1, 2024. A California store can document that match to educational level and language as part of its SB 553 file. A store anywhere else can use the same standard as a floor for its own refresher, covered in what California SB 553 requires a manager to train.

Why organized retail crime’s national scale is hard to measure

NRF’s theft and violence report ties rises in phone scams, gift card fraud, and cargo and supply chain theft to organized retail crime activity that keeps evolving (NRF). The FBI’s National Incident-Based Reporting System collects theft data reported to state and local law enforcement, including shoplifting, but it “does not capture ORC specifically,” according to a Congressional Research Service report (CRS). CRS names a primary barrier: there is “a lack of a consistent, widely accepted definition” of organized retail crime for agencies to measure against (CRS). A training program cannot anchor a huddle to a reliable national ORC figure, because federal crime data do not separate ORC from other theft. It can anchor to indicators: a pattern of repeat visitors, matched items across incidents, covered in how to train associates on organized retail crime.

NRF’s last shrink benchmark is from FY2022

NRF’s National Retail Security Survey put the average shrink rate at 1.6% of sales in FY2022, up from 1.4% in FY2021, equal to $112.1 billion in losses as a share of 2022 retail sales (NRF). In 2024, Retail Dive reported that NRF would not release a shrink report that year and that the latest available shrink statistics were in the prior year’s National Retail Security Survey (Retail Dive). A store citing “1.6% shrink” or “$112.1 billion” in a 2026 huddle is citing FY2022 figures. For a current, store-level number, run the chain’s own shrink math with the shrink calculator instead of reusing 2022’s, and check the definition against what retail shrink is.

Which loss prevention trends are backed by current data, and what do they mean for store associates?

Not every number above carries the same weight. Some are counts from police reports, some are executive survey shares of retailers reporting an increase or a tool in use, and two rows compare different measures that are not directly comparable. The table below labels each trend as a record-based count, a survey share, or different measures, and attaches the training implication a store briefing should carry.

What to do this week

• Pull the store’s own shrink and incident numbers before quoting a national trend in a huddle, and name the data source and date range out loud.
• Add a phone-scam and gift-card script refresher to this month’s training huddle, not just a shoplifting reminder.
• Check self-checkout host staffing against roughly one host per six machines, and confirm weight-scale controls are switched on, not disabled for speed.
• Route any camera or AI theft alert to a trained associate who reviews the footage or transaction and reports what they found.
• Confirm California stores have an SB 553 training file on hand, with material matched to employees’ educational level and language.