In an increasingly data-driven world, a new and fruitful partnership between the accountancy world and that of data analytics is taking hold. The task of uncovering previously hidden insights within what would once have been impenetrably large datasets no longer falls solely to data scientists and statisticians. Data analytics is being seen more and more as a core capability within business and within the profession.
There shouldn’t be any doubt as to the step change data analytics represents. Rick Payne, manager of ICAEW’s Finance Direction programme, points to the wide range of industries making use of its predictive powers. A smart grid that detects and predicts spare capacity in power generation will enable operators to redirect it towards other customers or utilities, potentially reducing our overall dependence on fossil fuels. Sensors embedded in the manufacturing process will enable companies to better assess their processes and improve efficiency.
Use of data by a business no longer needs to be reflective and retrospective, says Dave Coplin, CEO of technology consultancy The Envisioners and author of Business Reimagined. It can also be the prompt that causes traditional businesses to re-engineer themselves and become more effective business partners to customers. “Rolls-Royce used analytics to predict when their engines would need servicing, enabling them to change their whole business model to a vendor of flight time and making them much more cost-effective. Lift manufacturer Thyssenkrup uses analytics to anticipate the lifetime of components within its products and repair them in quiet times, which minimises disruption to customers,” he says. Accountants who can engage with data analytics will be able to have a positive impact on their organisations, says ICAEW member Andy Wills, director of data insight at Experian. “Being able to understand data and perform analytics are now critical skills to understanding the business,” he says.
Transaction-heavy organisations such as retailers or credit card providers collect millions upon millions of data points. “It’s only by getting right into the data that you begin to see what’s driving customer activities and determining more profitable outcomes. That in turn means professionals with these skills can define more profitable activities and generate growth in their businesses.” The increased take up of data analytics is much in evidence within audit firms, and in the corporate finance arena firms are investing significantly – recognising that the ability to analyse large datasets accurately can provide potential buyers with insights into potential savings and synergies in the due diligence process.
There is, says David Petrie, head of the Corporate Finance Faculty at ICAEW, a crucial role for chartered accountants here in interpreting the results of sophisticated analysis. “This is a role that needs to be played by people with a formal business training. It is important that chartered accountants remain in the foreground in this respect.”
Time to up-skill
Becky Shields, a partner at Moore Kingston Smith, says her firm started building a data analytics team three years ago as a means to improve audit quality and efficiency. The breadth of the accountancy/data analytics skillset required became clear early on as issues around data compatibility hit. Initially, the firm worked with external consultants and a small project team.
The value of having that dual perspective became apparent right away. “If you don’t have someone who understands the problems they are trying to solve, the risk is that you will end up with a solution that is either over-engineered or not fit for purpose, or with an interface that will not work for clients,” she says. At the beginning of the firm’s data analytics journey, they found that one employee had taken it upon himself to learn programming languages Python and R on his own initiative. Now Moore Kingston Smith has an analytics rotation within its training scheme. And while not everyone will have an aptitude for analytics, Shields says, the marriage of these data skills with the broader business foundation that the ACA provides creates much-needed solidity.
Programming languages such as SQL and Python are available via the Institute’s own learning and development programme, she notes. Education providers including Kaplan and BPP also offer analytics courses – and the ACA itself is evolving. “What you’re getting with these courses is context,” Shields says. Besides these external offerings come two crucial measures from ICAEW, a Data Analytics Community and research conducted this summer into members’ appetite for an ICAEW-branded Data Analytics Certification.
“Members have always been involved in data analysis. Now there are increasing amounts of structured and unstructured data available and the ready availability of analytics and modelling tools, it needs an alternative, more sophisticated approach,” says Jonathan Levy, ICAEW director of product development. “We think there is huge opportunity for chartered accountants in the area of advanced analytics and the new Data Analytics Community will support members by providing content and structured learning to help them become proficient in this area.”
The research found that with increasing take up of data analytics, members are turning to their own organisations’ software providers such as SQL, SAP and Alteryx or online learning platforms like Udemy. Respondents were supportive of an ICAEW-curated programme of learning. Crucially, they felt that certification would formalise an otherwise ad hoc approach to gaining data analytics skills and provide an attractive alternative to members having to effectively educate themselves in this area. An ICAEW-branded offering would bring validity and a business foundation, helping to put chartered accountants in a position where they can bridge any communication and understanding gap between data scientists and more mainstream finance professionals.
Looking at the core skills within data analytics provides insights into how accountants can interact and bring to bear business acumen, professional scepticism and ethical awareness among other skills and qualities. The spectrum of challenges that accountants engaging in this area will face, says Coplin, includes how data is stored and used, its quality, and the purpose for which it was originally captured. However, the sheer volume of data needed in analytics is becoming less of an issue, he says, as the tools to harness it become more accessible. “Increasingly, the more data you have, the easier it is to find the signal within the noise.”
Of more importance are the ethical and strategic questions around data use. Quite apart from the data protection regulations that forbid the application of data for anything other than the use for which it was originally collected, there is the question of unconscious bias. At its origin, a business dataset will have a human decision to collect the data. “Human decisions are riddled with bias, so understanding bias in data, being able to spot flaws, is increasingly important,” Coplin says.
The National Audit Office (NAO)
The NAO has been keen to embrace the possibilities data analytics holds to increase efficiency. It has been working on how it applies data analytics to audit for several years and now actively uses analytics within its public audit work.
“Our analytics work started around our Value for Money reports,” says Rachel Kirkham, head of data analytics research. “When I started to learn programming I could see the value. There is a lot of data manipulation in audit work. “The original data analytics team was quite small and comprised of auditors and data scientists, but the thinking was to teach the auditors to program so that when we build analytics for financial audit they can apply their audit knowledge.”
“We have also been looking at the wider training. Our graduate intake is given access to a platform and set a syllabus. It gives everyone the opportunity to develop their data analytics skills. You can bring in a data scientist [to an audit analytics project] but they may struggle with audit.” Many entrants, she explains, show an affinity for the data analytics work, but it’s not compulsory, so much as an important primer and foundation.
“For those who don’t want to pursue the analytics element, it’s more about applying the principles. Technology can sometimes feel distant. Actually it’s coming closer all the time.” The NAO’s data analytics project team has worked on automating audit tasks and building specific tools and routines that will, for instance, help visualise revenue by location, provide a profile of revenue over time and an understanding of how that revenue relates to receivables over time. Research extends to work on machine learning aimed at enabling auditors to gain more granular and sensitive readings of large datasets.
So for instance, the team has also been working on general ledger analytics that incorporates machine learning to identify outliers within account areas. This uses clustering algorithms to identify unusual flows of transactions as opposed to using more traditional methods of analysis. Natural language processing will also take its place in the advanced tech methodologies, particularly within Value for Money studies, as the NAO seeks to collate and analyse unstructured data.