Signature Alerts Help Deter Offenders to Minimize Losses

LPRC kickoff meeting
The Loss Prevention Research Council (LPRC) held a 2020 kickoff meeting at Bloomingdale’s New York City flagship store on January 15 following the National Retail Federation Big Show. More than 70 LPRC retail and solution provider members and guests were given updates on new developments at the LPRC and brainstormed 2020 initiatives. The meeting was hosted by Peter Chie, operating vice president for asset protection and risk management at Bloomingdale’s, and Read Hayes, PhD, managing director of the LPRC. Jordan Burchell and Kevin Tran from the LPRC also presented to the attendees.

Everybody likes a heads-up. The sooner we know we have a problem, the better. And with tech sensors continually improving, we’re more capable than ever at deterring or disrupting a bad actor to minimize the damage they can do because we more rapidly know we have an issue.

Since the dawn of crime and loss prevention, law enforcement and LP practitioners have used sketches, Polaroids, teletype messages, and phone or radio alerts to let them know a potential problem person was on the way, arriving, or on the property. And today, as in yesteryear, the alert is just that—an alert. It’s a heads-up and no more.

Values and Priorities
LP practitioners exist to enable their organizations by suppressing crime and loss problems, thereby allowing profitable commerce in a crime-saturated environment called “the world.” Our overall priority is simple: we want to do our level best to safeguard vulnerable people in confined spaces. Our employees, customers, and vendors need to feel and in fact be as safe as is reasonable while they’re with us. We also strive to protect other critical assets to stay in business, but people are primary.

- Sponsor -

Safeguarding vulnerable people is made more likely when we have good alerts. We can be more focused and, most importantly, proactive when we have a heads-up. If we don’t deter an event, all bets are off—the first domino has been activated. We need to know before a criminal launches their efforts, so we can try to stop them.

It’s Not Easy
Barriers to enhanced life safety include physics and technology limitations, budgets, and what some label the “privacy illusion.” All of us, every day, generate biological and digital exhaust and signatures. Digital signs are newer than bio signatures, but not very new. Radios, teletypes, and early mobile telephones have given off “interceptable” signals for decades, but like now, users made value-to-risk decisions. Perceived value for giving up total privacy can include convenience, time-saving, an entertaining experience, health monitoring, or life safety. Think social media posts and signing up for our favorite products’ advertising.

Risks today are like risks of any public action (even presumably private action), over millennia: exposure and exploitation. Mobile phones, social media posts, emailing, landline phone usage, wearable device emissions, ticket purchases, credit card usage, public mail including catalogs, and public movement and speech—all expose our utterances, preferences, and behavior to others. And unfortunately, stored physical or digital data has and will always have unplanned access vulnerability.

Capability Can Lead to Greater Human Safety
As mentioned, all humans generate bio and digital signatures. Part of our protective mission is to enhance and evaluate offender signature detection, definition, and response capability to create safer and more secure places. Capabilities include objective and evidence-based early threat detection, accurate threat validation and definition, focused alert use, alert to action speed, threat change detection and definition agility, and unity of effort whereby private and public protectors can work together to better head off and/or handle problems.

Key Signatures to Consider
Violence, theft, and fraud remain a growing concern across retail organizations, and the Loss Prevention Research Council’s Loss Prevention Innovation Working Group (LPIWG) is working the Signatures Innovation Initiative to formalize emerging signal recognition, identification, and action-support research and development (R&D).

Signatures might help us better protect and serve vulnerable people, places, and assets including but not limited to:

  • Protect employees and visitors via smarter place-access control to screen out violent offenders
  • Detect the presence of crime tools like magnets, foil, and weapons
  • Identify a suspect, crime plan or attempt, or pattern for action
  • Identify and track assets, protective devices, their security, and their status
  • Reconcile/validate checkout, delivery, and return transactions with items and people
  • Provide informed, focused sales attention to a good customer
  • Perceive the possible mood or intent of a shopper or visitor

To detect these and other active and passive signatures we use fixed and mobile sensors, which includes everything from canines, cameras, and online monitoring.

Separating Issues
I personally support privacy rights and limiting data access, but life safety takes priority, especially when citizens voluntarily opt into public behavior. It also seems important to separate issues to facilitate reasonable discussion, R&D, and real-world processes.

Issue 1: using prioritized sensor-generated alerts proactively safeguards people. Digital signatures come from wearables, sensors, RFID, and mobile devices, or social media posts. Bio signatures include body and face features and movement gaits.

Issue 2: alerts provide heads-up initial information to place managers, as do conventional photos, but do not dictate or conduct further action. A human must verify or confirm that the signature alert indicates a threat is inbound or present by validating whether the indicated person, for example, is most likely the suggested threat as if referring to a photo in their pocket to make sure it’s the right person. Signature accuracy may vary, but again the human takes any possible action based on their interpretation and further separates information and investigation as with “wanted photos,” fingerprints, or DNA. Limited or biased algorithms may initially or persistently exist due to “training” protocols and technical or data access limitations, but as with other scenarios, the human must decide and take any further action; alerts themselves cannot dictate or take further action. Additionally, it is in the organization’s best interest to develop and improve alert accuracy to maximize people safety. Negatively biased processes will not accomplish this and instead will create issues and reduce desired outcomes. Research-informed place-manager training is important. Managers need to understand what the alerts are letting them know, what their response options are if they validate an alert, and how to execute responses, and of course action ramifications.

Issue 3: collected signature data protection is important. Organizations strive to reasonably protect collected data via encryption, intrusion prevention, detection, neutralization, limited retention periods, and other methods. Data privacy is vitally important, part of the overall process, but is segmented from alert needs, accuracy, and usage.

Much more alert R&D and discussion is to come for all of us, but it seems important to publicly help us all shape the landscape together to safeguard very vulnerable people in our places and spaces by heading off dangerous offenders.

Stay Updated

Get critical information for loss prevention professionals, security and retail management delivered right to your inbox.