- Getting AI Right in Finance Isn’t Just About Investing the Most Time or Money
“The use of AI in finance departments is still nascent with the majority only having begun in the last two years,” said Jacob Joseph-David, director, research in the Gartner Finance practice. “The majority also fail to quickly realize the anticipated returns from such projects.”
Given the infancy of AI in finance, CFOs lack a clear definition of, and strategy for, success. To support CFOs, Gartner identified four critical actions for finance AI success (see Figure 1).
“The departments taking these four actions are finding twice the number of AI use cases on average compared to those who aren’t taking them,” said Joseph-David. “This translates to more significant business outcomes, such as new product lines, as well as finance department outcomes, such as greater accuracy and shorter process times.”
Figure 1: Four Actions for Finance AI Success
Source: Gartner (June 2022)
Generally, there are three options for securing talent with AI skills and expertise: hire new talent, upskill current talent or borrow talent from the IT department. Organizations that focus their talent strategies on hiring outside AI-skilled staff are significantly more likely to become leading AI finance organizations. Yet around half of finance organizations see upskilling as their primary talent strategy.
AI-specific staff bring invaluable experience in working with the nuances of AI, which allows the organization to overcome inertia in working with AI applications and shortens the technical learning curve. Conversely, while upskilling finance staff may be less expensive, doing so runs the risk of slowing progress and introducing greater potential for error. Additionally, new AI-specific staff provide the opportunity to move beyond traditional processes and mindsets by bringing with them new ideas to support AI deployment.
Invest in Software with Embedded AI for Quick Wins
Purchasing software with embedded AI capabilities allows organizations to more easily experiment with AI and apply it to more finance use cases; they can more easily build pilots for unique business problems. By contrast, building in-house AI solutions for all finance processes creates far more work and reduces finance’s bandwidth to explore new pilots or use cases.
Experiment Early and Broadly with Pilots
Top finance AI organizations are taking a fail fast experimental approach to AI deployment initially rather than making a few big bets. With more early pilots comes more uses of AI, and deployment is faster as the organization can zero in on the most successful pilots.
Typically, the most successful organizations are still exploring the same use cases as the less successful organizations with the three most common being accounting processes, back-office processing, and cash flow forecasting. The one exception is customer payment forecasting, which is a use case explored by approximately half of leading organizations but very few of the less successful organizations.
Choose an Analytical AI Implementation Leader
CFOs must select the appropriate person to head AI deployment in order to realize AI benefits. For example, this could mean the head of financial planning and analysis (FP&A), or the head of finance analytics, leading AI implementation rather than a controller.
Heads of FP&A and finance analytics are successful in leading AI due to their strong analytical and data backgrounds. They rely less on understanding traditional finance processes and more on understanding the complexities of AI in a business setting.
