Even though artificial intelligence (AI) is seen by many as a buzzword with little real application in business today, it’s a technology that is already shaping the smartphone industry and one that will have real impact on how organisations deliver services in the future.
Despite this, few CFOs are considering AI as part of their strategy, let alone making plans for how they will utilise this technology. That could be a big mistake.
Bloomberg Intelligence Analysts predict that within the $393 billion software industry, the $48 billion business analytics software segment will likely be the most disrupted by artificial intelligence. “Companies in this segment focus on collection, processing, analysing and visualisation of data,” says Bloomberg.
The office of finance has seen rapid change in recent years, with CFOs increasingly turning to powerful technologies like CPM and business intelligence to deliver better information and simplify processes. With AI set to accelerate that evolution even further, CFOs must consider their AI future and the role they will play in ensuring a positive impact on their organisations.
Learning machines require data, and lots of it
AI adoption is fuelled by a common set of benefits: making business nimbler and more responsive, eliminating manual processes, and supplying insights that support data-driven decision making. It’s the first and last on that list that makes it so compelling for finance.
That’s in part because expectations on the CFO role are rising. Boards want ever more detail and context in reporting, and they want it faster and more frequently. They also expect CFOs to be a source of business insight.
It’s the machine learning element of AI that will deliver competitive advantage and be truly valuable to business. Evolved from the study of pattern recognition and computational learning theory in AI, machine learning explores the study and construction of algorithms that can learn from and make predictions on data.
It therefore requires carefully defined rules and data to make it smart. Machines make predictions from huge quantities of data according to rules pre-defined in algorithms, without context, experience, ethics, or culture to provide context. To put it simply, they don’t learn like we do.
This data has to be defined and input in a useful way by humans if it is to achieve its promise of making business smarter and more responsive. If computers are to supply insights that support data-driven decision making, they have to be taught well.
At the moment there are three major categories of AI-driven finance solution that CFOs need to think about:
Digital assistants like Alexa or Siri which will effectively allow executives to sidestep CFOs and speak directly with their finance systems; perhaps asking the assistant for the latest financial results, any areas that fell short of forecast, with follow-on questions to understand where and why the shortfalls occurred.
Automated forecasts that improve on existing technology by selecting the most appropriate statistical model for the outcome you want to predict. This is being applied right now in AI trials that can predict the next month’s results with less than 5 per cent deviation.
Automated analytics that not only catch current deviations but are also capable of drilling down into the figures to explain the source of each delta.
As advanced technologies like these become more prevalent in daily operations and delivering information, organisations will become more reliant on them to drive the business forward.
How to career-proof against AI
In practical terms the advance of AI-accelerated automation means the office of finance will be spending less time consolidating spreadsheets and more time cultivating insights that can help the business anticipate problems and make course corrections before they become insurmountable.
As more and more transactional functions become automated, the need for strategic thinkers with cross-functional knowledge and mastery of technology will go up. That means CFOs will need to develop a different skill set that prioritises interpretive insight over mathematical acumen.
To career-proof against AI, CFOs will have to focus their efforts on finding the story that’s hidden in the data, and translating that into actionable information. This is where humans come into their own. They will need to re-train and re-focus themselves and their teams to keep abreast of the new AI innovations and how they deliver value.
That will impact recruitment as well. Tomorrow’s office of finance needs to be staffed with people who understand IT environments and how they work. As the day-to-day finance role involves more application of data-management tactics and mathematical models, they need to know how to tell the robots what to do – directing where and how to capture and analyse the data.
CFOs have to become educators
CFOs need to make bold moves now to position themselves with skills that complement the new technologies they’ll soon be investing in. We can expect machine learning to powerfully augment human expertise and experience in the near future even if that’s not a reality today. AI can provide data back-up and make suggestions to help the human decision-maker, but it’s the CFO who ultimately has to decide what to recommend.
How many CFOs are even starting to think about algorithm design and application? How will data be selected and used? This will require strong involvement from management at all levels as well as formal oversight and governance from the CFO, right down to the technicians designing the algorithms. In business, humans are the teachers of AI and CFOs will have to become educators.