Taking Finance Automation to the Next Level With AI

A company’s data is its most valuable weapon—yet very few of today’s finance teams make the best use of the data they have.

Part of the problem is that there is just so much of it. Today, even small to midsize companies collect far more information than they can ever commercialize—and as more and more devices will be connected to the Internet of Things, the amount of data transmitted will grow exponentially.

By far, most of this data will be just noise. Only a fraction of it will ever be insightful and predictive. How do you sort the nuggets of gold from this mountain of sand?

That’s where artificial intelligence (AI) comes in—specifically, machine learning.

McKinsey defines machine learning as “based on algorithms that can learn from data without relying on rules-based programming.” In the past, artificial intelligence was mostly rules-based; business analysts would develop the rules, so that if the algorithms encountered a pattern that was unusual, they would automatically trigger a specific response.

For example, if your credit card company detects a transaction for less than $5 and a few minutes later one for $200, the company’s systems might be programmed to identify that pattern as fraud. That is a business rule that a person programmed into the application.

Machine learning takes this a step further, automatically comparing the pattern to known, past instances of fraud to identify the likelihood that the credit card was hacked or stolen. While running these comparisons, the algorithms might discover that fraudulent transactions are likely to (a) be processed in separate locations, and (b) take place in stores where the credit card holder has never shopped before. The algorithms would add those criteria to the fraud evaluation, learning from the data just uncovered.

“Machine learning applications and analytics provide huge opportunities for customers to monetize their existing businesses and accelerate digital business,” according to R “Ray” Wang, principal analyst and CEO at Constellation Research.

Wang goes onto say: “Success requires a large corpus of data, strong expertise in data science, massive compute power, industry and domain expertise, and breadth of application solutions.” Sounds like a tall order.

In finance, machine learning could be a huge advantage in helping negotiate the best supplier terms and to optimize cash flow—especially during high-volume periods such as end-of-quarter. If these capabilities were incorporated into the company’s ERP systems, such insights would be presented to employees in context automatically. Further, the algorithms would continuously learn as the employees use the application, so they would have ever-improving, up-to-date recommendations that would achieve the best outcomes.

In short, the tools finance employees use to do their job could provide them with much more insight and expanded capabilities—without the aid of IT or the expense of a small army of data scientists.

This approach is the future of AI, according to Dave Schubmehl, research director of cognitive systems and content analytics for IDC.

“Within the foreseeable future, every enterprise application will be a smart application that intuitively learns from interactions with an enterprise’s data,” said Schubmehl.

Because AI is evolving rapidly, businesses face two challenges: Make sure that their business strategies can incorporate AI capabilities as they become available, and ensure that they’re building the right technical foundation for an AI future.

In an interview with OracleVoice, Des Cahill, Oracle CX evangelist, and Jack Berkowitz, Oracle’s vice president of products and data science for the company’s Adaptive Intelligence Applications, offered the following recommendations:

• Modernize your IT environment. “Cloud-based applications ensure that a company has the latest capabilities and patching, without a painfully long integration project,” Berkowitz says.
• Unlock data silos. AI technology requires a ton of data, from internal and external sources. If the flow of data in your organization is blocked by functional silos or disconnected technologies, clear away those obstacles. “As we talk about digital transformation, it all starts from a data perspective,” Cahill says. “You have to get your data straight.”
• Communicate what’s coming. Change management is 90% of the battle in implementing any digital strategy, Cahill says. With AI, which can automate tasks now done by employees, it will be particularly important to communicate the end goal (for instance, improved customer service or lowered costs through supplier discounts) as well as how employees’ jobs will change.
• Be prepared to train employees for new roles. For example, as more finance functions are automated, employees will no longer spend time tracking down data manually. Instead, they will be armed with real-time insights, which they will use as they work closely with LOBs and business leaders to identify opportunities and adjust strategies. For the employee, that means a very different skill set, new tools, and a new set of expectations.

Tapping AI capabilities in finance will take the possibilities of automation to new levels, arming employees with real-time insight and analysis that could make a big difference to the bottom line.