About us
Contact us

Forensic Data Analytics LLC is a management consulting firm supporting projects that prevent or detect fraud through the application of Predictive Analytics. 

"I am currently graduating from the Master of Science Predictive Analytics program at Northwestern.  The Occupational Fraud Risk Scoring (OFRS™) methology using our Forensic Data Analytics Risk Scoring (FDARS) application is based on my thesis project which helps manage Organization risk based on the updated COSO standards with the Fraud Risk framework."  Kathleen Kerwin MBA CFE owner and founder Forensic Data Analytics LLC

Forensic Data Analtyics Risk Scoring is a unique methology used to build the specific Occupational Risk Scoring (OFRS™) Predictive Analytic consulting service offering to identify predominant organizational risks.

Occupational Fraud Risk Scoring manages risk and identifies potential fraudsters for Internal Audit in their substantive testing process with the results being applied organizationally to further prevent or detect those engaging in occupational fraud.

Forensic Data Analtyics Risk Scoring and Occupational Fraud Risk Scoring, as applied to fraud detection and prevention, is a new application of the risk scoring process not previously used as of the application date for both the FDARS general and OFRS trademark.

Solutions include the areas of Occupational Fraud, Accounting/Audit, Insurance, Finance and Law Enforcement.

News update:

Is Internal Audit ready for Predictive Analytics?

Forensic Data Analytics LLC is focused on supporting projects that prevent or detect fraud through the application of Predictive Analytics. This is the first in a series of articles that is a road map to help Internal Audit to strategically manage the risk identified through the substantive testing process using predictive analytics.

Given the developments made by Internal Audit towards creating a framework for the data analytics for their testing, I believe the foundation is being established to transition their attention towards predictive analytics solutions. Top professional organizations have collaborated to strategically build out the risk management process. The Committee of Sponsoring Organization of the Treadway Commission[1] (COSO) developed the internal control framework which includes processes to provide reasonable assurance[2] to achieve effectiveness and efficiency in operations, reliability in financial reporting and compliance with applicable laws and regulations within an integrated framework including the control environment, risk assessment, control activities, information and communication and monitoring of the control systems. COSO working with the American Institute of CPAs (AICPA), the Institute of Internal Audit Association (IIA), and the Association of Certified Fraud Examiners (ACFE) developed the framework for fraud risk that has been recently updated by the Fraud Risk Management Framework (Cotton, Johnigan, Givarz, 2016). Figure 1 illustrates how the original Internal Control Principles and Components and relates directly to the Fraud Risk Management Framework.

Figure 1 Components of Internal Control[3] correlated with the COSO Fraud Risk Management Framework[4]

Predictive analytics applications are managed by CRISP-DM which is a process guide. Phases two and three, Data Understanding and Data Preparation respectively, are data analytic methods that take up to 70-80% of the project; see figure 2. 

Figure 2 Steps in the CRISP-DM Process[5]

Since Internal Audit (IA) staff is already learning and building a good grasp on data analytics, they are starting to understand the basis for predictive analytics. It is among this staff that the analytic leaders will emerge. Because of this high dependence on business knowledge needed to work in IA, managers have already found out that it is easier to train their staff in IT than to bring in IT personnel to handle data analytics. This is the experience in most departments that are based on professional degrees and education in general. Many audit staff already have the education prerequisites in statistics and calculus required to learn predictive analytics. An interim step is for staff members responsible for data analytics to understand how to view data through a statistical perspective of univariate and bivariate exploration.

A predictive analytics project needs at least one data scientist and support from analytical staff, the database team and IT. Analytical leaders of tomorrow come from those who have the knowledge and interest in analytics and specifically with exceptional and energetic members furthering their education by gaining credentials through predictive analytics master degree programs. Until they make the commitment to a formal master’s program, there are numerous free tutorial programs available online which have substantial credentials that are a good starting point to help them make the decision if this is the right direction for their future growth.

The current use of data analytics while aligning with business strategy is the foundation for building the groundwork for the next phase of greater analytic maturity which is developing a predictive analytics capability. Figure 3 shows that in 2015, 55% of Internal Audit departments were aligning with corporate strategy.

Figure 3 IA departments aligning with business strategy[6]

The use of data analytics while aligning to business strategy demonstrates that the right mindset and foundation is in place for seeking higher level analytics capability.

One of the most efficient and effective means of making the transition to predictive analytics is through an iterative approach. Changing the mindset of leadership and audit staff is a paradigm shift with no requirements on departmental budgets.

The key driver of a strategic predictive analytics program is executive management becoming data evangelists that inspire their managers and teams to change the way of doing business from applying experience and intuition to adding fact based data decision making to their current way of working and thinking. Rather than using a traditional Waterfall project management approach, the Agile scrum process is more efficient and effective. The scrum leaders are executive management that identify the strategy and requirements for the solutions and applications needed to provide them with the insight they need to prescribe back into the organization. It is certainly possible to hire a data scientist to build predictive analytic solutions without executive management being directly involved in the process. This is a tactical not a strategic application. However, the real power in using predictive analytics comes from establishing a link to the organization strategy. The owner of the organization strategy is executive management whose deep involvement is a critical success factor.

Change management, whether formal or informal, in the initial phases in the transition from a data to a predictive analytics focused organization is founded in a change of perspective in management and staff. Informally there is no budget cost, only a change in mindset. More likely than not, organizations and/or departments that are not profit based have not found enough reason to incorporate predictive analytics since the funds needed are not justified with projected revenue gains. Developing iterative approaches to assimilating predictive analytic ways of working into their organizations is not cost intensive as much as motivation intensive. When a consulting firm builds a project and throws it over the wall for an organization to handle, management and staff are required to get up to speed very quickly to be able to run the application. It is not uncommon for these applications not to exactly meet requirements since translation has taken place to some degree. Preparing in advance gives an organization control of the process, run their transition and develop their own applications. Building skills within helps an organization manages the conversion including having staff members being ready for the skills needed and executive management driving the efforts directly to their specifications.

An iterative approach anticipates a perception lag and time for an organization to mature into the changes needed for the conversion process. One of the most remarkable aspects of the Agile way of working is a “fail fast to succeed sooner” approach. With the traditional Waterfall project management approach, a consecutive series of tasks are planned with those within the process often not completely meeting eye to eye on the overall requirements until the final stages which is too late. The scrum project management team works in small highly qualified teams that meet eye to eye immediately and quickly find out where the misunderstandings exist then iron them out before advancing further. There is an emphasis on mutual respect and giving themselves and others the permission to fail so they can move together to build the best solution. Since the predictive analytics team is inherently small, executive management working directly to build and confirm that requirements are being met is the core of building an effective strategic solution.

Kathleen Kerwin MBA, CFE and MSPA

[1] Committee of Sponsoring Organization of the Treadway Commission (COSO) developed the Internal Control framework in 1992, http://www.coso.org/ 

[2] Internal Control - Integrated Framework, http://www.coso.org/documents/Internal%20Control-Integrated%20Framework.pdf

[3] PricewaterhouseCoopers LLP, Internal Control - Integrated Framework, 2013 for COSO, http://www.coso.org/documents/Internal%20Control-Integrated%20Framework.pdf

[4] Cotton, David, Johnigan, Sandra, Givarz, Leslye (2016), Fraud Risk Management Guide, a joint effort by the ACFE and COSO released Sept 2016, Retrieved from http://www.acfe.com/press-release.aspx?id=4294994630

[5] CRISP-DM 1.0, SPSS, 1999, Retrieved from https://www.the-modeling-agency.com/crisp-dm.pdf

[6] Abdolmohammadi, Mohammad, D’Onza, Giuseppe, Sarens, Gerrit (2015), Benchmarking Internal Audit Maturity, IIA Research Foundation, Retrieved from https://institutes.theiia.org/sites/oman/resources/Documents/2016-July-CBOK-Benchmarking-IA-Maturity.pdf