Critical events impact people, assets, & operations. This gap between when an event happens & when it’s picked up by the intelligence team is called “time to detect” (TTD). Rapid TTD is the cornerstone of an effective response.
Time to Detect is foundational - you can’t act on what you don’t know.
By definition, critical events are those occurrences (e.g. fire, severe weather, terrorism, travel disruptions, etc.) that impact people, assets, and operations. The speed with which a company comes to understand these events and how they’re unfolding determines its ability to respond. In the intelligence space, this gap between when an event happens and when it’s picked up by the intelligence team is called “time to detect” (TTD).
Rapid TTD is the cornerstone of an effective response.
Once alerted to a crisis, response teams can notify employees, lock down facilities, and re-route assets in transit. But they can’t do this if the intelligence is noisy, unreliable, or – most importantly – too late in arriving.
Traditionally, tracking events and collecting intelligence has been a time-consuming and expensive process that requires monitoring news sites, social media, government feeds, and television broadcasts for information on critical events. National and international news sources, which provide thorough coverage of major events as they unfold, and local news, which reports on minor events that may directly affect local business operations, both present challenges in tracking which events may pose a real threat.
Human monitoring has grown more challenging in the era of social media and publicly available government data, which compound the sheer volume of information to review and interpret. For reasons of cost, efficiency and effectiveness, it’s impossible to add enough people to manage the data influx.
Today’s tsunami of information requires a different approach. Purpose-built artificial intelligence (AI) reshapes critical event intelligence, providing unparalleled speed, coverage, and actionability.
HOW TO ACCELERATE DETECTION
As computers learn to scan and interpret large sets of structured and unstructured data, they make it possible to identify and monitor critical events in ways that radically outpace what traditional, human-only teams can do.
Leveraging this technology requires three things:
1. Validation of Sources
When major events happen, they tend to create a storm of information. Some of this coverage is reliable, some is speculation, and some is simply inaccurate. Knowing the difference in a crisis is paramount. Ideal data sources consistently provide unbiased information and updates on critical events. While news sources are important for gathering dependable facts on major critical events and their impacts, they are unlikely to cover a crash outside of your office building, which may disrupt your people and operations. Bridging this gap requires the integration of vetted local news sources into the data stream. Likewise, social media is an unprecedented resource for collecting hyper-local news in near real time. But social media posts may be unreliable or irrelevant and need to be cross-checked against valid sources. An AI platform fed by national, local, and social media sources can perform this monitoring and validation within minutes.
2. Real-time and Predictive Sources
A growing body of sources has become available to gain real-time and predictive insight on critical events. Government alert feeds and open data portals are now widely accessible. Earthquake monitoring stations, for example, remain the fastest way to detect earthquakes short of feeling one. Seismographs around the world deliver information on earthquakes via publicly available feeds, allowing near instantaneous detection and signals of a potentially larger quake. As more information emerges about the damage, businesses can adjust their countermeasures. Public documents for planned events such as demonstrations, street fairs or parties, are also now accessible. Protests, particularly in the US, are often planned weeks ahead to secure the necessary authorizations. Information on upcoming protests allow corporate teams to formulate a plan well in advance. Likewise, social media calendar integrations, social feeds, and event pages all provide information on the time, location, and date of the gatherings. As the scope of potentially relevant information expands, the importance of rapid monitoring grows.
With the use of AI-powered monitoring, intelligence teams can predict and build plans to address the potential disruption of pending natural disasters and pre-scheduled public events.
3. Integration of Multiple Sources into a Single Lens
Even if humans were able to keep up with the myriad data sources available, they would have difficulty quickly synthesizing all of the coverage into a single event summary. This is another area that AI improves time to detect and time to react. Instead of having disparate streams of news stories, data feeds, and social media posts pertaining to a given event, a purpose-built AI platform clusters the information into a single event profile – correlating it to the employees, facilities, and assets potentially affected. Clustering and correlation are vital ingredients for reducing TTD, enabling timely, automated alerts to the people who need to be notified.
THE IMPORTANCE OF LOW TTD
Modern threats are dynamic; they evolve as they unfold. An active shooter may be in a known location, somewhere at large, or apprehended. In cases like that, the difference matters. Similarly, wildfires, hurricanes, and other extreme weather may change paths or severity and threaten different facilities. In these sorts of events, corporate teams need the most current and accurate intelligence available. The integration of validated and real-time sources (e.g., police feeds), along with a consolidated view of all relevant intelligence, makes possible a level of speed and clarity that wasn’t possible a few years ago.
Below is a sample of TTD across three sources for a West Bank shooting. Sources include Stabilitas, social media alerts, and human analysts monitoring news broadcasts. The Y-axis indicates minutes between the event occurring and detection by the three groups.
While all three picked up the event in less than an hour, Stabilitas’ AI detected it in four minutes: a 90% improvement in TTD over one-dimensional monitoring. Knowing about such an incident within five minutes means that employees could be alerted and any assets in transit rerouted. The time difference between Stabilitas and the next best alternative is the difference between reacting to an event that’s happened and reacting to an event that’s happening.
NOT ALL SOLUTIONS ARE EQUAL IN TTD
The following graph shows the average TTD for a random sample of international security events between February and March 2019. The times come from Stabilitas AI, human analysts monitoring news, and social media analysis.
Stabilitas’ AI had the fastest average TTD by a wide margin, outperforming traditional human and social media analysis. The events used in these averages were critical events with a high impact. By combining local news, public data with a wide array of international sources, Stabilitas vastly accelerates TTD.
WHAT YOU CAN DO
By leveraging the Stabilitas AI platform, you can minimize the impact of critical events on your employees, facilities, and assets. Stabilitas aggregates thousands of data sources including local news, international news, government data, and social media feeds to create the most complete picture, as quickly as possible.
Stabilitas is the trusted platform for critical event intelligence. Our AI platform integrates more than 15,000 trusted data feeds, providing unparalleled speed, coverage, and actionability for critical events. To learn more about using Stabilitas to improve TTD and critical event actionability,