Security Camera Statistics CCTV and Video Surveillance Data for US Businesses, Every Figure Sourced and Dated
Most security camera statistics circulating online trace back to a vendor blog quoting another vendor blog. This page collects the figures that survive a trip to the original study, names the source and the year for each, and calls out the famous numbers that fall apart when you look for them.
The best evidence on security cameras is a 2019 meta-analysis of 80 evaluations by Piza, Welsh, Farrington and Thomas, which found CCTV is associated with about a 13 percent reduction in crime overall and a 37 percent reduction in car parks, with no statistically significant effect from passive, record-only camera schemes and no significant effect across US studies as a group. In the US, only 58.3 percent of camera schemes in that review were actively monitored, compared with 88.2 percent in the UK. That single split explains most of the disagreement about whether cameras work: monitored cameras change behavior and get used in investigations, unmonitored cameras mostly produce footage nobody watches.
Security Camera Statistics at a Glance
Every row names a source and a year. Where a number is old and no newer authoritative figure exists, the row says so instead of quietly implying it is current.
| Statistic | Figure | Source and date |
|---|---|---|
| Crime reduction from CCTV, all settings | About 13 percent | Piza, Welsh, Farrington and Thomas meta-analysis of 80 evaluations, 2019 |
| Crime reduction in car parks and parking facilities | About 37 percent, the strongest effect found | Piza et al., 2019 |
| Effect of passive, record-only camera schemes | Not statistically significant | Piza et al., 2019 |
| US camera schemes that were actively monitored | 58.3 percent, versus 88.2 percent in the UK | Piza et al., 2019 |
| Crimes where CCTV footage was available | 45 percent available, useful in 29 percent | Ashby, 2017, analysis of 251,195 British Transport Police crimes |
| Robbery detection rate with useful CCTV | Rose from 8.9 percent to 55.7 percent | Ashby, 2017 |
| US property crime clearance rate | 15.9 percent | FBI Crime in the Nation, 2024 data |
| Reported shoplifting offenses, US | 1,270,097 offenses, up 8.9 percent, while overall property crime fell 8.1 percent | FBI, 2024 data |
| Retail shrink rate | 1.6 percent of sales, $112.1 billion (fiscal 2022, the last survey published) | National Retail Federation National Retail Security Survey, FY2022. NRF discontinued the survey, so any newer US shrink rate has no authoritative source. |
| Share of shrink from employee theft | 29 percent internal, 36 percent external, 27 percent process error | NRF, FY2022 |
| Revenue lost to occupational fraud | About 5 percent per year, median loss $145,000 per case | ACFE Report to the Nations, 2024 (1,921 cases, 138 countries) |
| How occupational fraud is detected | 43 percent by tip, three times the next method | ACFE, 2024 |
| Alarm calls police respond to that are false | 94 to 98 percent | US Department of Justice, COPS Office, False Burglar Alarms (Rana Sampson) |
| Share of all police calls for service that are false alarms | 10 to 25 percent, each taking about 20 minutes and usually 2 officers | DOJ COPS Office |
| Municipal false-alarm fine, Los Angeles | $176 per false alarm response | Los Angeles Municipal Code 103.206, 2025 schedule |
| Burglars who said cameras influenced target choice | Cameras were the top considered factor, around 57 percent; about 43 percent said they would pick a different target | UNC Charlotte survey of 422 incarcerated burglars, 2012. Funded by an alarm-industry foundation, and measures stated intent, not outcomes. |
| Loss in stores with self-checkout | 33 percent higher, a 0.42 point difference | ECR Retail Loss Group, 2026. Mostly European sample. |
| Construction equipment theft, US | Estimated $300 million to $1 billion per year, average theft around $30,000, recovery around 21 percent | National Equipment Register and NICB estimates |
| "Organized retail crime is nearly half of shrink, about $45 billion" | Retracted. Do not cite. | NRF removed the claim on December 1, 2023 after reporting showed the underlying arithmetic was wrong. |
| "Employee theft costs US businesses $50 billion a year" | Unverifiable. No traceable primary source. | Loosely attributed to the US Chamber of Commerce, with no published study behind it. |
| "Only one crime is solved per 1,000 CCTV cameras" | Unverifiable in its usual form. | Repeated from a 2008 London police briefing that cannot be located as published research. Ashby, 2017 is the defensible number instead. |
What the Data Actually Says About Security Cameras
Read the table end to end and a pattern shows up that neither the camera industry nor its critics like to say out loud. Cameras have a real but modest deterrent effect, concentrated in enclosed, well-defined spaces like parking structures where an offender knows the camera is watching a specific act. The effect thins out in open city centers, on public transport, and against violent crime, where it is not statistically significant at all. That is the honest ceiling on deterrence.
Where cameras clearly earn their keep is after the fact. Ashby's analysis of a quarter of a million crimes found footage was available in 45 percent of cases and useful in 29 percent, and that when it was useful, the robbery detection rate went from 8.9 percent to 55.7 percent. Set that against the FBI's 15.9 percent US property-crime clearance rate and the value proposition is clear: most property crime is never solved, and usable video is one of the few things that moves that number.
The variable that separates a system that works from one that does not is monitoring. Passive schemes showed no significant effect in the meta-analysis. US schemes, only 58.3 percent of which were actively monitored, showed no significant effect as a group. Whatever else you take from the data, take this: a camera nobody is watching is a documentation device, not a security control. We go deeper into the deterrence evidence in our guide on whether security cameras deter crime.
The Numbers by Category
Retail theft and shrink
The last authoritative US shrink figure is the NRF's FY2022 survey: 1.6 percent of sales, $112.1 billion, split 36 percent external theft, 29 percent internal, 27 percent process error. The NRF discontinued that survey, so every "2025 shrink rate" you see online is invented. Meanwhile the direction of travel is genuinely contested: NRF's 2025 impact report shows shoplifting incidents up 18 percent among retailers surveyed, while the Council on Criminal Justice found police-reported shoplifting down 10 percent across 21 cities. Retailer-reported and police-reported are different measurements and should never be blended. Full breakdown in our retail theft statistics guide.
Employee and internal theft
The ACFE's 2024 Report to the Nations is the only large, methodologically transparent dataset here: about 5 percent of revenue lost annually to occupational fraud, a median case loss of $145,000, and 43 percent of cases caught by a tip rather than by any control. For small businesses with fewer than 100 employees, the median loss is $141,000, which is a survival-level number. See employee theft statistics for the full dataset.
False alarms and police response
The DOJ COPS Office puts the false rate on alarm calls at 94 to 98 percent, consuming 10 to 25 percent of all police calls for service at roughly 20 minutes and two officers each. Cities responded with fines and permit suspensions: Los Angeles charges $176 per false alarm, San Francisco escalates from $100 to $250, Houston allows three free per permit year. Video verification is the direct fix, and it is why verified alarms get priority response. Details and fee schedules in our false alarm statistics guide.
What we will not publish
Three of the most-quoted security camera statistics on the internet do not survive checking: the retracted NRF organized-retail-crime figure, the $50 billion employee theft number with no study behind it, and the London "one crime per 1,000 cameras" line that cannot be traced to published research. We also decline to cite market-size projections, which are paywalled vendor forecasts, not measurements. If a number is not sourced, it is not on this page.
Questions About Security Camera Statistics
Do security cameras reduce crime?
Modestly, and only in some settings. The 2019 meta-analysis of 80 evaluations found about a 13 percent reduction overall and 37 percent in car parks, but no statistically significant effect for violent crime, city centers, public transport, or passive record-only schemes. Actively monitored cameras drove the effect, which is the finding most vendors leave out.
What percentage of crimes are solved with security camera footage?
The best measurement comes from Ashby's 2017 study of 251,195 crimes: footage was available in 45 percent of cases and judged useful in 29 percent. When it was useful, robbery detection jumped from 8.9 percent to 55.7 percent. For context, the FBI puts the US property-crime clearance rate at 15.9 percent, so usable footage is one of the strongest levers available.
What percentage of retail shrink is employee theft?
About 29 percent, according to the National Retail Federation's fiscal 2022 survey, the last one it published. External theft including organized retail crime accounted for 36 percent and process or control errors for 27 percent. The NRF has since discontinued the survey, so no newer authoritative US breakdown exists.
How many burglar alarms are false?
Between 94 and 98 percent of alarm calls that police respond to are false, according to the US Department of Justice COPS Office. Those calls consume 10 to 25 percent of all police calls for service and take roughly 20 minutes and two officers each, which is why so many US cities now fine repeat offenders and prioritize video-verified alarms.
How much do businesses lose to internal fraud?
The ACFE's 2024 Report to the Nations estimates the typical organization loses about 5 percent of revenue per year to occupational fraud, with a median loss of $145,000 per case. Small businesses under 100 employees see a median loss of $141,000, and 43 percent of all cases are uncovered by a tip rather than by an audit or control.
Why do security camera statistics disagree with each other?
Because they measure different things and rarely say so. Retailer surveys count what businesses report internally, police data counts what gets reported to police, and vendor blogs recycle both without dates. Shoplifting numbers that move 18 percent up and 10 percent down in the same year are not contradictory, they are two different instruments. Always check who collected the number and in what year.
Sources
- Piza, E., Welsh, B., Farrington, D., and Thomas, A. (2019). CCTV surveillance for crime prevention: a 40-year systematic review with meta-analysis. Criminology and Public Policy. Also summarized by the National Institute of Justice CrimeSolutions.
- Ashby, M. (2017). The value of CCTV surveillance cameras as an investigative tool: an empirical analysis. European Journal on Criminal Policy and Research. Based on 251,195 British Transport Police crime records.
- Federal Bureau of Investigation, Crime in the Nation, 2024 data release (property crime clearance rate, shoplifting offense counts).
- National Retail Federation, National Retail Security Survey, fiscal 2022 (final edition), and Impact of Retail Theft and Violence, 2025.
- Council on Criminal Justice, Crime Trends in US Cities, 2025 mid-year analysis of 21 cities.
- Association of Certified Fraud Examiners, Occupational Fraud 2024: A Report to the Nations.
- US Department of Justice, Office of Community Oriented Policing Services, False Burglar Alarms, 2nd edition (Rana Sampson).
- Los Angeles Municipal Code Section 103.206; San Francisco and Houston municipal alarm ordinances, 2025 fee schedules.
- University of North Carolina at Charlotte, Understanding Decisions to Burglarize from the Offender's Perspective (2012), N = 422. Funded by the Alarm Industry Research and Educational Foundation.
- ECR Retail Loss Group, self-checkout loss research, 2026.
- National Equipment Register and National Insurance Crime Bureau, construction equipment theft estimates.
Found an error, or have a primary source for a figure we flagged as unverifiable? We would rather fix this page than defend it.
The data says monitored cameras work. Ours are always monitored.
Surveillant runs AI on every frame of your existing ONVIF or RTSP cameras, so nothing depends on a person watching a wall of monitors.