Data analytics is touted as being a huge help in many areas of higher education. But Alan Lees, of Kingston City Group an internal audit consortium for the higher education sector, is sceptical that it can replace the bloodhound-like investigative instincts of human internal auditors.
The exponential growth in new technologies, the increasing complexity of systems and processes, and the amount of data that is generated and stored by institutions, makes the job of the internal auditor a lot more challenging than in the good old days. In such a complex and fast-moving environment, auditors are still expected to understand how systems work, sometimes at the micro level, and to be able to spot potential errors and anomalies in an institution’s data. The manual selection of samples for testing can still work for simple processes but not necessarily where things get a bit more complex, where problems can be hidden from view.
So how does the auditor make sure they unearth potential problems in the data? ‘Traditional’ audit testing might still be a possibility but it would be uneconomic given the amount of manual work involved. It may be that data analytics are the answer, and in some cases the only answer, for the auditor to provide sufficient assurance about the correctness of data being audited.
I say “could be” because these techniques have to be used wisely, ie targeted appropriately, and the auditor must spend time reviewing the information reported, and investigating unusual trends and anomalies. Data analytics haven’t always been used as much as they should for this purpose but they are now getting a new lease of life as auditors look for new and innovative ways to provide value to their clients, and to deliver more robust assurance.
Auditor intellect and scepticism are still part of the value proposition
A recent example where Kingston City Group added value through use of these techniques was for an institution that has, apparently, paid suppliers more than once. Elsewhere, other anomalies spotted have include payments being made to employees through the purchase ledger, and to companies that have an institution’s employees also on their payroll as directors.
At the last count, some £440,000 of transactions were being investigated further to establish whether these were legitimate transactions or whether something a bit more sinister has been unearthed. Powerful stuff? Yes, particularly when you consider the payback potential from the use of data analytics simply through, as in this case, being able to:
- Interrogate and analyse large volumes of data;
- Provide assurance about the completeness of the data population;
- Objectively select samples for testing through use of statistical techniques; and
- Identify potential risk areas, control failures or fraud
So can data analytics deliver value to institutions and their auditors? Of course they can, but let’s not forget that to take advantage of them and therefore realise the value added requires an auditor to interpret the results of the interrogations, and he/she still has to investigate to confirm whether or not the results are showing something a bit iffy.
Auditor intellect and scepticism are still part of the value proposition, despite a commentator recently suggesting that auditors could be replaced by artificial intelligence. Maybe they’re right but it isn’t going to happen anytime soon, and certainly not until robots have been programmed to replicate the ‘auditor’s nose’!
