지난 몇 년 동안 내부 감사 기능에서 데이터 분석 활동의 사용이 폭발적으로 증가했습니다. 데이터 수집 및 저장에서 볼 수 있는 발전과 데이터 분석을 위한 정교한 기술의 개발로 인해 데이터 분석 역량을 구축하고자 하는 내부 감사 기능이 증가하고 있습니다. 그러나 데이터 분석 역량을 성공적으로 구현하는 것은 사용하는 기술 및 수집하는 데이터 그 이상의 의미가 있습니다. 성공적인 데이터 분석 구현을 위해서는 내부 감사 기술 개발을 포함한 내부 감사 기능, 내부 감사 운영 모델 조정, 감사 방식의 스타일 및 접근 방식에 걸쳐 전반적인 변화가 필요합니다.
이 기사는 실제 경험에 기반하여 성공적인 내부 감사 데이터 분석 전략을 개발하고 구현하는 방법을 설명합니다. 성공적인 데이터 분석 구현을 위한 10가지 팁을 제공합니다.
It is important that you ensure people are held accountable for pushing your use of data analytics forward. If you are to achieve wide-ranging use of analytics it is important to integrate reward and recognition of this into your performance management system. One organization I worked with ensured everyone, right through the internal audit function, had a clear objective around the use of data analytics in audit work and the function as a whole had targets for the percentage of audit work that deployed simple analytics and more complex audit work that was reviewed and signed off by the QA function. Whilst this is relatively crude it emphasized the point that this mattered and everyone should be looking for ways in which they could make the use of data analytics endemic in the function’s activities. It had considerable success and now data analytics is integrated into all that the function does and is delivering considerable results in terms of audit findings, continuous business monitoring, and the deployment of audit developed analytic capabilities into the first and second line for them to use as part of their day-to-day control management activity.
There are many ways in which you can celebrate your achievements and every organization will have its own culture and approach to sharing success. At one organization I worked for we prepared a six-monthly set of case studies of audit successes. We presented at our six-monthly audit leadership event a handful of these, that had been voted as our proudest moments by the whole team, to the chief executive, chairman, and members of the audit committee as 3–5-minute vignettes of success. A great evening for the team, but also a chance for the audit committee and executives to see a collective view of the value we were adding and the energy with which we brought to our work. These case studies were not all data analytics-driven, but what we did see over time was an increasing number of them becoming analytics-led as momentum and support for the data analytics initiative took hold.
It is important that you truly make using data analytics an ‘opt out’ not ‘opt in’ part of your audit methodology. Auditors should, at key stages of the audit process (tollgates), have to show why data analytics cannot be used rather than how they can. This opt-out approach will focus the mind of the internal audit team on looking for data analytics opportunities at all stages and increase the proportion of audits using analytics.
It is very easy, audit by audit, to throw together a spreadsheet to analyse some data. However, this won’t be reusable in the future. Encourage your auditors when developing analytics, whether that be through Excel, in TeamMate, or using other software such as PowerBi and Python, to build with reuse in mind. This will take a bit longer, as will require setting up in a way that someone can pick it up in the future, but the benefits will pay off for the function longer term. Consider giving time at the end of each audit for the team to clean up the analytics and store it in a well-organized data analytics library. Central data analytics teams can really help here if you have the scale to do this. They can take the lead in ensuring this systematization is carried out when analytics are used and also develop pre-configured testing scripts and ensure the libraries set up are well maintained and useful.
This is all about how you set up your operating model to deliver your data analytics. Some larger functions have dedicated well-resourced teams to deliver analytics, but increasingly functions are taking a hybrid approach where a small group of specialists work alongside front-line auditors who have been trained and encouraged to use analytics. This blend allows capability built to be widespread and penetration of analytics work to be deeper into more audit activity than if the work is done in a separate central unit.
There is no silver bullet for the successful development and implementation of a data analytics strategy, but hopefully, the series of tips outlined will be a useful catalyst to your work as you consider the conditions that need to be right for you to achieve momentum around your data analytics work. Good luck!
