Originally, job-related risk was calculated based on keywords such as “CEO,” “Accounting,” and “Manager.” Since organizations use a wide variety of job titles, this original method may not have accounted for all potential risks. To help calculate job-related risks more accurately, we developed a new data-driven approach for predicting the risks associated with each job title.

First, a score is calculated for each user with a job title based on how closely connected they are to other users in the organization. This score is reflected in their job function Risk Score factor. Then, the score data from this factor is also used to train a model that predicts a user's job function risk based on their job title. For more information on Risk Score factors, read our Virtual Risk Officer (VRO) and Risk Score Guide.

In the example below, the CEO is closely connected to every person in the organization. Because of this, the CEO receives the highest score. Both the Scholarships Administrator and Grants Administrator are directly connected to the President and CEO, but the Scholarships Administrator receives a higher score because he has one more person connected to him and the Grants Administrator has none. This also demonstrates why using keywords, such as “Administrator” isn’t as accurate as using a data-based calculation.

The model is retrained periodically, and it reconfigures Risk Scores to include new job titles. This learning model helps maintain Risk Score accuracy and ensures that all users who have a defined job title are assigned a score for their job function.