An Introduction To Data Analytics

Data Analytics is becoming more prominent as a sophisticated organisational tool for enhancing productivity and guiding human intelligence in corporate decision making. Over the last decade, these techniques have been progressively embraced across business environments to the competitive advantage of those that have taken a lead role. This brief note aims to introduce data analytics and how use within the auditing process can benefit both audit firms and their clients.

Data Analytics: What is it?
Data analytics varies depending on organisational needs but in simple terms works by analysing datasets, usually large in size, through the ability to automatically categorise and identify correlations or trends as well as other possible insights. The significance of such techniques is how they are able to examine large quantities of data almost instantaneously, which would be near impossible by manual effort. Data analytics facilitates auditors in automating manual processes, such as substantive testing procedures, alleviating time pressures to allow more room for the qualified judgement aspect of their role. This is becoming extensively more available as a tool due to developments in cloud computing, which has dramatically enhanced the rate of innovation and removed the significant investment barrier which previously existed.

Techniques in Data Analytics
Example techniques: Data Analytics in its current form represents a broad range of techniques which impact the audit process in entirely different ways. However, each technique is a way to extract anomalies or segment populations; essentially turning data into usable information. A few different techniques within audit are explored below by comparison to similar tools within more established analytical industries.

    • Risk Assessment
      Risk analytics is a tool whereby behaviour or particular characteristics can be categorised to identify the risk profiles of different items to allow an appropriate approach to be determined. Analytical techniques are used extensively in insurance companies, even more so for usage-based car insurance (UBI). The black box is a good example of UBI as it collects data by tracking driver movements including times, locations and speed to assess the likelihood of having to pay out based on expected behaviour. This ‘likelihood’ score is attributed to a risk profile and an appropriate insurance plan can be shared with the driver. Likewise, in an audit process, graphical analysis of a transaction population or account balance can be used to highlight uncharacteristically risky behaviour. An example of this could be as simple as analysing expenditure by month, which highlights a spike in manual adjustments posted by one client user immediately prior to year end. By presenting this information visually the auditor will more easily be able to identify the related risk and design appropriate audit procedures to focus on the related audit assertions and risk of material misstatement.
    • Fraud Testing
      Analytical techniques lend themselves very well to fraud testing. Typically centred around characteristic based tests, Transactions can be immediately identified from within huge populations which are posted outside of working hours, or with high-risk key words in the description field, allowing the auditor to immediately hone in on items requiring more detailed consideration, The use of third party data sets, such as companies house data to find related party transactions, adds additional sophistication to such techniques. The auditor can then use these findings to guide their next steps, extending the use of data analytics to generate journals entry testing from the population of all accounting entries posted during the year, based on the outcomes of the applied tests.
    • Business Process Analysis
      Business Process Analysis (BPA) is a way an organisation can understand an existing process and can identify improvement areas. Amazon are renowned for their use of BPA which gives them the ability to customise web pages for each of their millions of customers depending on customer behaviour and as a result enhance their revenue. Amazon do this by using customer click stream data and understand user purchases individually. Without analytics, this would be impossible to do with 152 million customers worldwide (Datafloq, 2016). Within audit, BPA can analyse the behaviour of all the methods revenue has been recognised during the year, allowing auditors to follow transactions from invoice to cash, automatically testing process flows and financial settlement. By performing such a granular analysis of 100% of transactions, an auditor can ensure that designed accounting principles and processes have been complied with. Such techniques can replace traditional substantive sample tests, with audit effort focused on unusual transactions not conforming to expectations.

 

Conclusion
Existing auditing techniques, such as ticking samples of invoices, are no longer effective. Clients are seeking efficient compliance services which add value – but business is more complex and transactional than ever before. Data Analytics offers a solution to this challenge.