The fictional crime-solver Sherlock Holmes once referred in a conversation to “the curious incident of the dog in the night-time.” A Scotland Yard detective replied, “The dog did nothing in the night-time.” Holmes retorted, “That was the curious incident.”
In the field of analytics, the equivalent of the dog that didn’t bark is the relatively low level of adoption of advanced analytics in finance and accounting functions. Despite being a quantitative field by nature, finance has trailed other functions like marketing, supply chain, operations, and even human resources in employing advanced analytics to make key decisions.
Some finance professionals may have experimented with an occasional regression model in a spreadsheet format, but for the finance function to make advanced analytics a core capability — on the same level as external reporting or the closing process — is quite rare.
Finance groups, of course, have long used descriptive analytics (also called business intelligence) to do their work, including reports, dashboards and scorecards, and online queries. But descriptive analytics don’t tell the user anything about underlying patterns in the numbers, and they only describe the past.
More advanced approaches involve predictive analytics, which use statistical models of past data to make predictions about the future, and prescriptive analytics, which use data and analytics to recommend decisions and actions for workers.
It is rare to find extensive use of predictive and prescriptive analytics in finance. This lamentable situation is beginning to change, however. We interviewed 10 organizations in which finance functions were already working with advanced analytics. CFOs in these companies are becoming champions of analytics, and a variety of finance and accounting-based analytical applications are being implemented.
Given the amount of data available to finance functions, and the level of insight that can be achieved, it seems inevitable that finance activity with analytics will grow further over time. And because of the rise of new analytical and cognitive technologies (machine learning, sensor data analytics, and robotic process automation, for example), finance will have to build capabilities rapidly just to keep up.
There are two possible roles for finance organizations with respect to advanced analytics. One involves “sticking to their knitting” by building an advanced analytics competency to address finance problems and objectives. The other involves an even more ambitious role for finance: taking the lead for analytics within a company and becoming the primary provider of analytical insights for non-finance functions like sales and marketing, human resources, and operations.
While those functions may already have some advanced analytics capability, more companies are beginning to see the value of analytics that transcend functional boundaries.
For example, at Toyota Financial Services (TFS), finance traditionally focused on measuring financial performance. But in the last few years the company has built a comprehensive analytical capability by leveraging people, tools, and data, Amit Shroff, a TFS finance executive, told us. Today the function plays a broader role in measuring and enhancing product profitability, sales effectiveness and customer loyalty.
Finance partners with the business to derive insights from volumes of loan and lease contract-level data to improve profitability by geography, product, and channel. Additionally, finance-developed analytical tools combined with sales’ local market knowledge enable consultative relationships with dealers.
For example, multi-dimensional correlation analysis of TFS insurance products sold and the corresponding positive impact generated for the dealership (e.g., service visits, parts sales) allow Toyota and Lexus dealers to receive critical insights into customer behavior and loyalty. Thus, the analytical capability contributes to sustainable growth for TFS and the overall Toyota ecosystem.
Advanced Analytics for the Finance Function
One reasonable approach for analytically oriented finance leaders, of course, is to focus on advanced analytics that relate specifically to finance. There is no shortage of possible applications here.
A finance organization might, for example, focus on understanding the drivers of financial performance, both financial and nonfinancial. It might assess whether capital investments are well spent (typically using a “design of experiments” approach, with test and control cases), or whether employees are likely to be participating in fraudulent activity. These types of activities can add significant value to the traditional activities performed by the finance function.
One company that is aggressively pursuing this approach is Intel. A small number of finance professionals began to advocate for greater use of analytics two years ago. They presented to the senior finance team the idea of building a generalized competency in advanced analytics, and the team was very supportive of the idea.
One early step was to compare Intel’s finance analytics capabilities to those of leading firms in the area, and Intel found that some high-tech firms in Silicon Valley (which have strong analytical orientations in general) had more capabilities than its finance team had.
Intel’s finance group started several new initiatives in the forecasting area, including statistical forecasts of revenue and inventory levels, and prediction of impairments in Intel Capital’s investments. Intel has also embarked upon a broad effort to educate finance professionals and managers about advanced analytics topics, and is planning certification programs for them as well.
Finance as the Organizational Analytics Leader
Another role for finance that we have observed in some companies involves assuming leadership not only for financial analytics problems, but also for advanced analytics initiatives involving other business functions or units — and sometimes across the entire company.
In some cases, the justification for this preeminent finance role is that financial investment and returns play a role in the initiatives. That might mean, for example, analytics projects to determine whether marketing investments really pay off, to assess what kinds of new hires provide the greatest economic benefit to the organization, or to identify ways to optimize inventory levels to reduce carrying costs.
Since finance is an organization that is experienced at organization-wide collaboration, focusing on analytics-oriented services is a logical extension of the finance role. Collaboration is, of course, required with the IT organization, but in roughly 40% to 70% of U.S. firms IT reports to the CFO, so this collaboration is likely easier to engender.
Several finance organizations have taken on this leadership role relative to advanced analytics. At a large automobile manufacturer, for example, a new “Global Data, Insight, and Analytics” organization was created in 2015 and reports to the CFO.
Within finance, the office is addressing broad capabilities like visual analytics, global reporting tools, and the optimization of risk and credit. Outside of finance, the new group is focusing on such analytics projects as connected vehicle data analysis, data-driven pricing, volume and revenue optimization, and minimizing recall and warranty costs.
On the data side, the group is focused on topics like data governance and infrastructure, as well as cybersecurity (which is increasingly becoming more analytical itself).
Deloitte LLP, with which both authors of this article have a relationship, also has a finance organization that leads analytics for the U.S. firm for non-client purposes. Frank Friedman, CFO of Deloitte LLP, established an analytics group in the finance function several years ago, but it works with advanced analytics throughout the organization.
In terms of financial analytics initiatives, the group has focused on optimizing receivables and reducing risk. Outside of finance, it has addressed such problems as employee attrition, the structure of profitable client engagements, and partner compensation.
Friedman comments that since finance is one of two common languages of the organization (clients being the other), and since the finance function has an objective and credible image, it’s the logical home for advanced analytics.
Attributes of Finance Organizations that Pursue Analytics
Finance leaders whose organizations have been successful at advanced analytics have several attributes in common.
These firms all have a finance leader with a passion for analytics. He or she sets the vision and drives his or her organization down that defined path. Successful finance leaders spend substantial time communicating the value of analytics; they have had to play “evangelist” and persuade non-finance functions to accept their analysis and conclusions. Several mentioned the importance of “telling stories with data.”
These leaders are also not afraid to experiment. Our interviews suggested that they didn’t spend a lot of time deliberating about which projects to take on, but rather experimented with projects they believed would add value.
Finally, they are also democratic; they want their finance and even company-wide employees to use analytics on a daily basis on all their tasks and decisions.
We anticipate that at some point, nearly all finance functions will embrace advanced analytics and employ them to improve financial decisions and processes. A good percentage will likely drive analytics for their entire organizations.
The only remaining mysteries may be why this analytical focus didn’t take place any earlier than it did, and what technological and business drivers were most influential in the finance analytics journey.
Tom Davenport is the president’s distinguished professor of information technology and management at Babson College and a fellow of the MIT Center for Digital Business. Adrian Tay is the managing director and U.S. finance analytics leader for Deloitte Consulting.