Data We Collect
Our products and solutions are fueled by the following types of data:
- Law Enforcement Citations
- Criminal Arrests
- Judicial Filings
- Judicial Dispositions
- Correctional Confinement Listings
- Sex Offender Registries
- Driver License Suspensions
- Crash Reports
The data DDS collects often differs in depth and substance from that typically sought by aggregators. For example, we place an emphasis on obtaining the underlying judicial case filings so that we can alert interested customers (in non-employment contexts) of recent conduct that could signify an unacceptable organizational risk.
How We Access Data
Unlike some organizations that choose cost savings over geographic coverage, Data Driven Safety is not satisfied with purchasing second-hand, inexpensive and easily procured data sets. Instead, we procure data directly from the governmental source that created the information, often on a daily basis. This translates into an agency-by-agency collection approach that often includes seeking data from agencies that may only process a handful of relevant records each month. The result is more accurate, robust and reliable data to fuel our products.
Below are examples of the types of agencies that produce the data we consume:
- Local Police Departments
- County Sheriff’s Offices
- State Law Enforcement Agencies
- Trial Courts
- Probation Departments
- Prosecutorial Authorities
- State Correctional Facilities
- State Motor Vehicle Agencies
Many agencies are pleased to make public records available to us. They will attest to our impeccable industry reputation. Some of the thousands of clerks, public information officers and IT wizards with whom we interact have even taken the time to keep in touch with us on a more personal basis.
When the data is not readily provided, our public record attorneys and data consumption teams work diligently to allay individual agency objections to the disclosure of public information. We strongly believe that one of the best barometers for the health of a democracy is the level of transparency it maintains with the citizenry it was created to serve.
How We Manage Data
Since our founding in 2009, we have developed and continually refined a purpose-built data aggregation engine. This proprietary collection system is based on adaptive consumption technology to allow the efficient collection of public record information. It gives us the ability to access a myriad of data services in an automated manner, ranging from file pick-ups via ftp sites, to subscription search pages, to web-generated file queries, to emailed reports, to physical media extraction, to OCR translation, to image retrieval and beyond.
After DDS has identified source data (and, in many cases, secured the creation of a custom periodic report), we create the appropriate consumption script and schedule its automated run sequence. All source information is pulled into our databases in a standardized format. Inputs are mapped to known outputs, resulting in database records that, save for differences in incoming data coverage, appear to be from one source. Furthermore, the databases include fields to capture virtually any data fields encountered, allowing us to store more data per record and avoid the issues surrounding “unstructured” data sets. To make the data more meaningful to our customers, we use sophisticated parsing methods to take “messy” data and place it in the appropriate database fields.
To ensure that we accurately and completely consume data in a correct manner, our data collection system recognizes changes in source data presentation and, in many cases, adapts automatically. If it cannot adapt, the system flags the source for manual adjustment. The entire system operates continuously, running individual source collection on variable schedules. It is supported by an automated quality analysis system that is constantly seeking deviations between expected values, counts and field densities.
Our completely configurable matching algorithms allow us to dial up or down the matching sensitivity by source and provides the ability to filter potential matches against factors including: plan specifics, employee type, source type, geographic location, offense type, offense level, outcome type and matching sensitivity. To further ensure that the data we provide is current and correct, all alerts for our employment services rely on at least one level of manual confirmation.
How We Secure Data
Data privacy and security are of paramount importance to us. We have implemented industry-accepted (and best practice) sets of policies, systems and processes to securely house and carefully safeguard all information we collect.
The heart of our approach is to maintain secure systems that limit access to those Data Driven Safety employees with a documented need-to-know.
This translates at a tactical level to a system in which we manage our own servers rather than relying on third party solutions. In addition, we employ multi-factor authentication, full disc encryption, and other best practices. Regimented data asset classification and inventory management tools are coupled with regular personnel training and review of privileged user access rights. All of these are supported by a program that includes routine security audits (including periodic third party PEN testing).