
Jefferson Luna-Perez was shot to death on the afternoon of May 17, 2023, in the parking lot outside Roosevelt High School in Petworth. Less than a quarter mile away, a recording device perched on a roof was listening for gunshots in the area.
That sensor is part of a gunshot-detection system called ShotSpotter that listens for gunshots around the District and notifies police. The Metropolitan Police Department has spent millions of dollars on the system since the early 2000s, but lax oversight and inadequate data tracking make it difficult to say whether the system is worth the cost.
Other major cities have cut ties with SoundThinking Inc., the California-based company that licenses the system to the District, after independent reviews cast doubts on its effectiveness, and raised issues about its potential civil rights violations.
The government oversight in D.C. has become so inattentive that, when asked, MPD could not provide up-to-date locations of the system’s sensors. In response to a Freedom of Information Act request, MPD said in November that SoundThinking stopped sharing that information with the District, though it’s unclear exactly when. “MPD contracts for a coverage area, but MPD does not have sensor placement information,” the response says.
The system failed to notify officers of the shot that killed Luna-Perez, according to an MPD spokesperson, even though it occurred within the microphone’s roughly 1,200-foot range. Luna-Perez was unconscious by the time responding officers arrived at the school. He was transferred to a nearby hospital, where he was pronounced dead.
It’s unclear whether a quicker police response facilitated by a ShotSpotter alert could have prevented Luna-Perez’s death. According to a spokesperson for MPD, officers treat ShotSpotter alerts as low-priority calls—the same classification as traffic complaints. Officers only treat the system’s alerts as serious cases when they receive a confirmation of shooting by either a call for service or another officer on the scene. The spokesperson did not answer a specific question asking to describe the benefits of the ShotSpotter system.
Jerome Filip, a spokesperson for SoundThinking, says in an email that the ShotSpotter technology “helps law enforcement quickly respond to gunfire, render aid to victims, and strengthen criminal investigations.” He says the company fills an essential role in filling the gap between gunshots that are not reported to police and allows first responders to know when and where potential shots are occurring in real time.
“The technology has led to [the] Metropolitan Police Department being directed to hundreds of gunshot-wound victims where there were no corresponding 911 calls, meaning that first responders were able to deliver medical aid to victims who may otherwise not have received the care they needed,” Filip’s statement says.
By cross-referencing MPD crime data with ShotSpotter alert data from January 2014 to January 2025, City Paper and the Investigative Reporting Workshop confirmed at least three shootings, including Luna-Perez’s killing, in the areas of the District covered by ShotSpotter that the system didn’t pick up.
They could be the tip of the iceberg.
While MPD does keep a database of the sounds the sensors pick up, it doesn’t keep track of which alerts correspond to confirmed shootings. Without syncing those two datasets, it’s hard for the department to figure out how many of the alerts are actual shootings and which are similar sounds such as fireworks. (ShotSpotter has updated its system over the years and claim to have improved its ability to distinguish between gunshots and other “impulse noises.”)
In 2005, the District first established ShotSpotter in the Seventh District, which encompasses most of Wards 7 and 8, with support from the Federal Bureau of Investigation as part of the “Building a Safer DC” initiative. The system continued to expand over the years, and MPD said in a 2018 hearing that the technology covers “17 square miles … within the First, Third, Fourth, Fifth, Sixth, and Seventh Police Districts.”
MPD has paid at least $5.16 million to SoundThinking since 2016, according to the Office of Contracting and Procurement. The agency was unable to provide the total amount the city has spent on the program dating back to 2005.
In 2019, MPD upgraded and expanded ShotSpotter’s coverage, and in 2021, ShotSpotter—now in more than 170 cities worldwide—opened its East Coast headquarters in D.C.
“ShotSpotter allows us to make the best use of our police resources,” Mayor Muriel Bowser said at the headquarters’ opening. She later stated that the District will use the technology to “enhance … efforts to combat gun violence and make communities safer.”
At that point, ShotSpotter was already the subject of nationwide controversy after a series of reports and audits of the technology cast serious doubt on its effectiveness, as well as its clustering of sensors in communities of color.
When a reporter asked Bowser at the press conference if she was concerned that ShotSpotter would contribute to overpolicing in Wards 7 and 8, she responded coolly: “No.”
But a 2021 contract between the District and SoundThinking contains a list of at least 335 sensor coordinates, and there are virtually no sensors listed in the majority-White Wards 2 and 3. Most of the sensors on the list were installed in Wards 7 and 8, where the highest proportion of the District’s Black residents reside.
Indeed, at least 43 percent of ShotSpotter alerts in the District from January 2014 to January of this year originated in Ward 8, and about 32 percent originated in Ward 7. By comparison, the other six wards combined made up about 24 percent of ShotSpotter alerts, ranging from 0 percent in Ward 3 to about 9.5 percent in Ward 5.
This is consistent with findings about ShotSpotter’s use across the country, including a 2024 Wired investigation that analyzed a leaked document containing the location of more than 25,000 ShotSpotter sensors worldwide. The investigation found that the sensors included in its data were predominantly placed in low-income communities of color.
Filip’s statement says that MPD decides where to install ShotSpotter sensors, which are placed in “areas with the greatest incidence of gun violence and the greatest need for 24x7x365 coverage. ShotSpotter is deployed in 7/7 police district without bias.”
Aside from the list of sensors, SoundThinking’s 2021 contract with D.C. also outlines a list of monthly deliverables that the company is obligated to send to MPD, including an inventory of sensor locations, a list of shots reported by the system, and a monthly report of issues with the system.
MPD told City Paper that SoundThinking no longer shares this information with the department, and declined to say when it last received these contractually required reports.
Despite growing concerns with the program and lack of accountability, the D.C. Council approved nearly $1.8 million in fiscal year 2023 to integrate ShotSpotter into its new “real time crime center,” operated by the department and other local law enforcement agencies.
Ward 2 Councilmember Brooke Pinto, who chairs the Committee on the Judiciary and Public Safety, declined to say whether legislators ever completed an evaluation of the use of ShotSpotter in D.C.
“ShotSpotter is a useful tool for MPD to get alerts when shootings occur in our community to respond quickly to scenes,” her office says in an emailed statement.
Two years ago, as D.C. was facing its deadliest year in more than two decades, the department went on to solve only 52 percent of the 274 homicide cases by the end of the year, mirroring national trends in decreasing closure rates.
While the District’s homicide rate dropped significantly in 2024, and MPD’s homicide clearance rate increased to 60 percent, there’s no evidence that these improvements have anything to do with recent spending on the ShotSpotter system.
Eric Piza, co-director of Northeastern University’s Crime Prevention Lab, says that the District’s decrease in gun violence and converse increase in closure rate should not necessarily be linked to its increased ShotSpotter investment.
In a report funded by the Department of Justice, representing the largest research project on gunshot detection technology to date, Piza analyzed more than a decade’s worth of ShotSpotter data from Kansas City, Missouri, and Chicago to test the merits of the program in both cities.
Piza and his colleagues found that while ShotSpotter is a useful forensic tool—streamlining police response to gunshots and evidence collection—the program had little impact on gun violence or law enforcement’s ability to solve crimes in either city.
“What we refer to as the procedural benefits—the response and evidence collection—doesn’t seem [to] translate to crime prevention and improved investigation,” Piza tells City Paper.
An inquiry into the Chicago Police Department’s use of ShotSpotter by the city’s Inspector General came to similar conclusions, and added that the department’s “record-keeping practices are obstructing a meaningful analysis of the effectiveness of the technology.”
Chicago Mayor Brandon Johnson cited the report’s findings when he rejected an extension of ShotSpotter’s contract with the city and began removing its devices in September 2024.
“The ultimate goal is to deploy resources on the most effective strategies and tactics proven to accelerate the current downward trend in violent crime,” Johnson said. “We have to explore better options that save more lives.”
Last summer, New York City’s comptroller found that ShotSpotter actually worked against crime fighting. The audit released in June 2024 found that in eight months’ time, the system’s inability to accurately identify a gunshot sent officers to investigate 7,262 incidents that did not turn out to be confirmed shootings.
The audit also found that NYPD and SoundThinking “significantly overstate” the reduction in the amount of time it takes officers to respond to emergencies. ShotSpotter claims officers can respond five minutes faster with the help of their technology, but the audit found the average response time to ShotSpotter alerts was less than two minutes faster than to 911 reports of gunfire.
Despite the findings, NYPD signed a three-year, $21.8 million extension with SoundThinking.
Spurred by major cities’ decision to interrogate ShotSpotter’s usefulness, a group of Democratic senators wrote a letter in May 2024 to the Department of Homeland Security, urging it to investigate its use of the system and whether its over-deployment in communities of color violates the Civil Rights Act.
“Communities are better off when police are able to be notified precisely in real time,” SoundThinking CEO Ralph Clark wrote in a letter to the Boston Herald, responding to the senators’ request. “It helps them save lives, recover physical forensic evidence in the form of shell casings.”
Two years after Luna-Perez was shot outside his high school, the case is still unsolved. (His family did not respond to efforts to reach them.) Detectives have developed suspects, according to an MPD spokesperson, but have not made any arrests, and the investigation remains open and active.