Assurance analytics — reducing noise and false positives — a contemporary approach

In adopting analytics, one of the main challenges faced by assurance teams (e.g. auditors, risk assurance teams, continuous auditing teams) is the volume of exceptions generated.

The traditional sampling approach typically involved evaluation of between 5 and 50 items, and consequently the number of exceptions would not fall out of that range. However, when we use analytics across full populations, the number of exceptions produced can be quite high. The sheer volume can result in:

  • Difficulty in eliminating false positives — noise!
  • Unnecessary pain for business folk (that have to deal with the mountain of results)
  • Assurance leadership / teams losing faith in analytics

There are various approaches to overcome this, e.g. progressively categorising results into specific buckets based on key characteristics, reviewing a sample of those, and then extrapolating the results of the reviewed sample to the remainder of the population of exceptions. However, when the key characteristics can’t be easily categorised — e.g. when the exceptions are based on and include both structured and unstructured data — the traditional approach doesn’t quite fit.

An alternate method that we have been using recently involves the use of machine learning techniques — similar approach, different techniques. The result is a significant (>90%) reduction in the number of false positives, providing business stakeholders with more targeted information to consider and providing audit stakeholders with the comfort that we did not simply discard results, but used a well structured and defensible approach to create a manageable set of exceptions for follow-up.

A case study is outlined here: and explored further here:

..all using an open source solution. The tools and techniques to improve IA’s use of analytics are now readily available — let’s use them.