One of the best practices we see at organizations with a robust audit analytics program is having a vision and a plan. When we tell this to leaders that are just getting started, we often hear, “Great, so how do I get started figuring out my vision and plan?” Here are five points to consider if you are just starting or considering where to focus next.
Depth versus breadth. Is it more important to perform analytics on a greater number of audits or go deeper on the few audits where you incorporate analytics? If your focus is to go broad, you will need more auditors involved, tools that everyone can use, and a process supporting that objective. On the other hand, if you want to dive deeper, you will likely focus on just one or a few very technically skilled resources, and tools and processes to support that auditor.
Who should perform analysis? This is influenced not only by your broad versus deep decision, but also by considering how you will consume the analysis. Are you planning to provide static data analysis information to your audience? Or are you planning to provide your audience with tools that allow them to slice and dice or otherwise perform some analysis on their own?
When will analytics be performed and consumed? Are you planning to use these tools while doing fieldwork and providing the information in reports to be read by your audience? Or will you present the findings of your data analysis? Or do you want to create a webpage where the information can be consumed 24/7 from laptops or mobile devices? The answer to these questions is based on your overall approach but will influence tool and process decisions.
Audit analytics versus business analytics. Business analytics are often used to identify risk by comparing two or more related factors. An example is revenue versus expense, and these factors are usually shown over time. Significant changes in one factor that is not also supported by corresponding factors could show a business risk. These are very often shown in charts. Audit analytics generally refers to the data analysis performed on the audit during fieldwork. These are the data analysis tests we need to perform to meet the audit objective. For example, if the audit includes examining signoff levels, we might look at every transaction to make sure each was signed off by an appropriate resource.
Audit analytics tools are designed for this kind of analysis, however, it also supports visualizations. Likewise, business analysis tools focus on visualizations, but some fieldwork testing can be performed with these tools, creating some overlap. Choosing a tool that is designed for what you most want to accomplish in this area makes it easier to perform the work and allows you to go deeper than a tool from the other category.
What does your data look like? The answer often varies based on what we are auditing. If you work in an environment where you have very large volumes of data, hundreds of millions or more of transactions, you may think you need tools and approaches that can handle this volume. But what about the other areas that you are auditing? Data tools and processes designed for very large volumes of transactions are generally not appropriate for use on smaller volumes of data. These very large data sets often require resources with specialized skills. Organizations generally need to have separate people, tools, and processes for these differing data sets. In addition to data volume, you should also consider velocity, variety, and the veracity of the data, often called the “four V’s of data.”
Whether you are just starting out or you are wondering how to expand, considering these questions can help you determine your direction.
Depth versus breadth. Is it more important to perform analytics on a greater number of audits or go deeper on the few audits where you incorporate analytics? If your focus is to go broad, you will need more auditors involved, tools that everyone can use, and a process supporting that objective. On the other hand, if you want to dive deeper, you will likely focus on just one or a few very technically skilled resources, and tools and processes to support that auditor.
Who should perform analysis? This is influenced not only by your broad versus deep decision, but also by considering how you will consume the analysis. Are you planning to provide static data analysis information to your audience? Or are you planning to provide your audience with tools that allow them to slice and dice or otherwise perform some analysis on their own?
When will analytics be performed and consumed? Are you planning to use these tools while doing fieldwork and providing the information in reports to be read by your audience? Or will you present the findings of your data analysis? Or do you want to create a webpage where the information can be consumed 24/7 from laptops or mobile devices? The answer to these questions is based on your overall approach but will influence tool and process decisions.
Audit analytics versus business analytics. Business analytics are often used to identify risk by comparing two or more related factors. An example is revenue versus expense, and these factors are usually shown over time. Significant changes in one factor that is not also supported by corresponding factors could show a business risk. These are very often shown in charts. Audit analytics generally refers to the data analysis performed on the audit during fieldwork. These are the data analysis tests we need to perform to meet the audit objective. For example, if the audit includes examining signoff levels, we might look at every transaction to make sure each was signed off by an appropriate resource.
Audit analytics tools are designed for this kind of analysis, however, it also supports visualizations. Likewise, business analysis tools focus on visualizations, but some fieldwork testing can be performed with these tools, creating some overlap. Choosing a tool that is designed for what you most want to accomplish in this area makes it easier to perform the work and allows you to go deeper than a tool from the other category.
What does your data look like? The answer often varies based on what we are auditing. If you work in an environment where you have very large volumes of data, hundreds of millions or more of transactions, you may think you need tools and approaches that can handle this volume. But what about the other areas that you are auditing? Data tools and processes designed for very large volumes of transactions are generally not appropriate for use on smaller volumes of data. These very large data sets often require resources with specialized skills. Organizations generally need to have separate people, tools, and processes for these differing data sets. In addition to data volume, you should also consider velocity, variety, and the veracity of the data, often called the “four V’s of data.”
Whether you are just starting out or you are wondering how to expand, considering these questions can help you determine your direction.