Days of Data | Insurance Research Letter - Argo Group

Days of Data | Insurance Research Letter

Advice to law enforcement agencies on managing massive amounts of data from video surveillance

By Ashley E. Bonner, WSO-CST

This article was republished with permission from Insurance Research Letter.

52,000. That’s the approximate number of hours of video that each large police department in the United States collects in a year. Put another way, that’s more than 2,100 days of footage, or nearly six years’ worth for a single year. Gathering more hours of data than time allows obviously means this footage is coming from many and multiple sources—drones, electronic stun guns, and cameras affixed to the dashboards of police vehicles, attached to the bodies of police officers, and located in all manner of public places. According to Darrel Stephens of the Major Cities Chiefs of Police Association, “This technology has deployed faster than any technology I’ve ever seen in my 45-plus years involved in policing.” Now consider the fact that there are some 18,000 state and local police departments in the nation. The mind boggles at the total amount of data from video surveillance being amassed by the country’s law enforcement agencies.

The Challenges

Not surprisingly, as Time magazine cites, these agencies “are wrestling with the challenges of storing all of this data. In New Orleans, the police department plans to pay $1.2 million for 350 cameras, with much of that total going to data storage. A 2014 Police Executive Research Forum report cited one department that reported it would cost $2 million a year for a plan that included 900 cameras — with storage again accounting for the bulk of the amount.” White House Chief Data Scientist D.J. Patil uses three questions to sum up the challenges these agencies face in dealing with such massive amounts of data: “Where do we put it? How do we access it? How do we index it?”

52,000. That’s the approximate number of hours of video that each large police department in the United States collects in a year. Put another way, that’s more than 2,100 days of footage, or nearly six years’ worth for a single year.

Storage is not merely a technical question. It’s also a legal one. States such as California, Connecticut and Georgia have created record-retention laws that require law enforcement agencies to retain data for certain lengths of time. (Brennan Centre for Justice charts, state by state, the requirements for non-evidentiary video retention.) In Georgia, my home state, counties and cities must keep possession of general footage for a minimum of 180 days, and preserve for at least 30 months video recordings that depict detention or arrest, use of force, and traffic accidents, or that are part of criminal investigations. This timeframe covers the state’s two-year statute of limitations plus an additional six months. In a self-defeating move, the State of Connecticut passed legislation governing the use of body cameras by police departments but denied the extra funding that law enforcement requested for the storage required to support the law. What good are these data without the means to store it, index it and access it?

Expert risk managers recommend that law enforcement agencies keep all footage for six months beyond the statute of limitations, for three main reasons. First, police department officials and counsel want to see as much data as possible related to incidents of possible misconduct. Even if it is bad footage it is still better than nothing at all. Second, retaining footage past the statute of limitations helps police departments not only to prove what occurred, but also what didn’t occur, which can be just as critical. If nothing happened, when a claimant makes a charge of police misconduct, it benefits police officials to be able to run video to show that no such misconduct occurred. Third, if such a claim were to occur toward the end of the statute of limitations, retaining general video for the additional 180 days ensures that evidence potentially valuable to the defense has not already been discarded.

Compounding this challenge, law enforcement agencies must preserve an unbreakable chain of custody for data from video surveillance. This chain of custody should include a record of who has handled these data. Specific care must be taken to avoid spoliation, which is the destruction of evidence necessary to pending or contemplated litigation. Law enforcement agencies must be diligent about following retention schedules and ensuring records are not destroyed prematurely. In these cases, plaintiffs’ counsel will argue that failure to comply with the retention statute constitutes spoliation.

The challenges continue. Law enforcement agencies must put in place further safeguards to prevent the negligent use of data. First and foremost, agencies should review footage routinely from each officer, not only to ensure that officers are following procedures and training mechanisms properly, but also as a defense in court. A police official on the stand should be in a position to respond affirmatively to the question: Have you ever reviewed video data and, if so, how often? To respond in the negative could give rise to negligence. This kind of safeguard can also enable agencies to show non-involvement in claims and in so doing receive dismissals from courts under motions for summary judgment.

In a self-defeating move, the State of Connecticut passed legislation governing the use of body cameras by police departments but denied the extra funding that law enforcement requested for the storage required to support the law.


Law enforcement agencies, of course, aren’t the only organizations gathering data via video. Media outlets do, too. These outlets collect vast amounts of footage yet may only air brief moments during their broadcasts, which gives rise to public outcry and even violence. As such, risk managers encourage police departments to build strong relationships with local media and use these ties to educate broadcasters on what law enforcement does and why incidents occur.

A rapidly evolving situation can be confusing to someone who is not familiar with the training options that an officer is provided with, the hierarchy of the use of force, or the nature of handling someone who is unhealthy and overcome by drug use. If the media is educated, then they may be able to share more of the story on the air, which would help departments prevent members of the community from jumping to conclusions before the full story is communicated.

The Solutions

So how can law enforcement agencies come to terms with these days of data? While research reveals that no one has the perfect solution to creating affordable video data storage, many fruitful approaches exist. Companies in the marketplace provide the storage at a cost, though they may sell or lease the cameras and other supporting equipment at or below cost. The best analogy to compare this approach with is buying a new ink jet printer but with the ink refill cartridges costing more than the printer itself.

A more flexible option is the cloud, with data residing in a center somewhere on the planet. For law enforcement agencies that opt for the cloud, plan for the long term because this method becomes part of your permanent infrastructure. Cloud technologies also use some form of a wireless method to upload data in real time, which increases costs. Inflating costs further, law enforcement agencies should ideally have two forms of storage: a primary system and a secondary off-site system, such as a cloud-based service, to ensure that if a glitch occurs in one storage system, it is backed up elsewhere. Further, it is strongly recommended that storage occurs on United States soil/governance in order to prevent international political interference.

Faced with all these demands, police departments are increasingly turning to private, high-volume storage businesses such as Veripatrol, a cloud-based service owned by body-cam maker VieVu, and, a similar service owned by competitor Taser International. Sydney Siegmeth, a Taser spokesperson, says that a video is uploaded to every 1.6 seconds, equaling 2.1 petabytes (one petabyte equals a million gigabytes). For these companies, storage is becoming big business. In the third quarter of 2015, Taser saw $36.9 million in storage sales, up from just $5.9 million in the first quarter of 2014.

Examples of Costs

Pricing for Taser’s cloud storage service ranges from $15 to $79 per month, per officer. Taser’s Officer Safety Plan, which automatically replaces old cameras every 2.5 years, costs $99 per month, per officer. VieVu sells its Veripatrol cloud service as a bundle priced at $55 per month, per officer. After purchasing an LE3 camera for $199, the VieVu Solution includes the Veripatrol secure file-management software and 60GB of storage, which can be increased for 12.5 cents per gigabyte per month. An onsite storage software bundle sells for $25 per month, per officer. VieVu pitches its video service as more compliant with policies of the FBI’s Criminal Justice Information Service, because it’s based on the Windows Azure Government cloud, which uses more advanced security and requires government audits.

But how do police departments manage such expenses? According to the International Business Times, while body cameras “may range in price from $300 to $800 per officer, [they] can cost hundreds of thousands more in monthly video storage costs spread over time. In San Diego, for instance, the police department recently purchased cameras for about $500 each, but will spend $1,495 per year, per unit, for video storage costs.” Compounding the problem, most departments don’t have in-house expertise or personnel to manage the operation. Grants are one solution to the cost challenge. Some jurisdictions supply grants that help law enforcement agencies solve their data management and storage requirements. For instance, the Office of the Governor in Texas makes available grants of up to $10 million. While grants don’t completely solve the problem, departments should search and try to take advantage of what is out there. Grant-writing services are also available in the event that a department does not have an in-house writer or someone who is able to take on the challenge.

The Plan

So what approach should your organization follow? To determine your best solution, put together a plan that identifies your agency’s particular needs. While each plan will be unique, all should take into account the following three considerations:

  1. Determine what kind of disaster recovery plan your storage contractor follows and whether it is tested properly. The test should be a true test and not just some scripts run to see if it looks to work on paper or in simulation. At a minimum, your contractor should carry out a full documented test every year that involves basically pulling the plug on one data center, then seeing if you can still access your information. A loss of data could be devastating on many levels, so confidence in the testing process is critical. It is recommended that when a test is conducted, results are available so you and your supplier can carry out a review of what went wrong and right.
  2. Know the retention policies of your local, state and federal court systems. As pointed out earlier in this article, these systems differ from city to city and state to state. The leading edge of this new type of evidence will change, as will the limitations of where data from video surveillance may be used. So make sure your department’s choices can meet these changes. Start by having your end vendor supply you with updates every six months. This field is changing rapidly. You will want your vendor to certify that it is adhering to the latest changes and what may be impacted by newer changes to your current plan.
  3. Find out whether there is a need for local storage within your jurisdiction and for how long. If the method is platform agnostic, you should be able to retain chain of evidence from the local storage to the cloud. A platform agnostic product runs equally well across more than one platform. Your storage type should be platform-agnostic, which will save you money later on. As more storage businesses fill the market place, you can move your data to lower-priced storage based on cost.
Law enforcement agencies should ideally have two forms of storage: a primary system and a secondary off-site system, such as a cloud-based service.

These three considerations are the most important elements of your plan. Yet that blueprint should reflect three additional salient points:

  1. It should account for how your organization would man-age video evidence related to acts or violations outside its jurisdiction. These cases can drag on for years or even decades. Your organization must be able to maintain data that can be used in these kinds of cases.
  2. It should ensure your storage platform meets the standard for chain of custody and access tracking set down by the FBI’s Criminal Justice Information Services Division.
  3. It should specify that your data are stored within the United States, as any storage outside the country can be influenced by needs and demands of foreign jurisdictions rather than your own.

As you go about preparing your plan, ask questions of independent sources. For instance, go onto Internet forums. You can do this anonymously. The biggest mistake you can make is to take at face value what you are told by salespeople. Confirm what you believe you are getting from a company’s tech support. Use search engines to help you break the technological information down into easier concepts to understand. Make sure to engage your IT staff, especially since they will work with the system. Also, take advantage of your local business partners for major and even minor tasks involved.

I also would be happy to work with you. Should you have any questions or need additional information, please feel free to contact me.

The Takeaway 

Above all else, as you set about to tackle the challenges that come with managing massive amounts of data from video surveillance, bring to the task the values of the finest law enforcement agencies: protect and serve the public; put in place solid policies and procedures, reviewing, and improving them annually with documentation; train your officers based upon these standard operating procedures; and most of all, do the right thing regardless of policy or procedure, and nurture that kind of culture within your department. The days of data have arrived for all law enforcement agencies in the United States. And the volume of data is measured in literal days – years even. Managing this massive amount of data from video surveillance will surely be one of your organization’s supreme tests.