Support Learning and Insight

It’s more important than ever to understand the challenges facing financial executives. Support the Financial Education & Research Foundation today.

Technology

Applying AI and Automation Technologies to Upskill Finance, Part 2


by Chen Amit

Automation software is the natural complement in the age of upskilling. With technology, the need to hire for lower end functions can be reduced while enabling organizations to put greater emphasis on higher-level work.

In our previous article, we talked about the role of RPA and machine learning in finance, but how do you actually bring AI into your organization? First, you must assess where you are as an organization. Are you on a growth or acquisitions path, or have you plateaued and are waiting to sunset? What is hiring on your team like? Do you look to employ bookkeepers or strategists/analysts?

The prime candidates for automation are businesses with a growth mindset—those with limited hiring capacity or teams that are very selective in terms of hiring. It’s not hard to bring in someone and pay them $20 an hour to do basic bookkeeping. But is that person going to be a long-term solution as the business grows? And if workload increases are imminent, will that person want to keep their career in suspended animation to take that on? And finally, will they do a good job? After all, we are talking about measurement and control of your company’s finances, which requires deep diligence and constant monitoring.

If you answer “no” to any of those questions, automation and AI can provide a viable alternative—or at the very least—augment deeply mundane finance operations tasks. Here’s how to get started: 

1. Select processes that can be automated in isolation
Not every task in finance can be automated at this time. It helps to choose tasks that are either on the periphery or deep in the trenches. Oftentimes, they require less integration with day-to-day processes or major implementations around core ERP projects. 

For example, supplier management and onboarding can be relatively non-disruptive since you can roll it out to new suppliers first before engaging the entire supply chain. Or for an insurance company, RPA could be used to perform and normalize claims processing on the backend.

2. Consider a pilot project to gain momentum 
A pilot AI project allows organizations to test a clearly defined and measurable objective that creates business value. This could be measuring workload changes, accuracy numbers, and scalable outcomes. A pilot also enables organizations to test the software with their strange quirky rules they may have in place.
The most important thing is not to focus on cost payback yet. Automation relies on economies of scale and a simple pilot does not often reveal the full broad set of benefits, such as the immediate ability to adopt best practices. It’s more important to recognize what value you’re able to capture, then use that data to communicate its potential to executive leadership. 

3. Deploying winning solutions
Once executive teams are on board, proper training will be required in order for the technology to remain cost-effective and efficient. As AI is set to transform different jobs across a range of departments, this training allows the workforce to understand how their jobs will change over time. 

Transitioning out of some tasks will be easier to convey than others. For example, manual menial tasks that no one enjoys doing are more likely to gain acceptance, particularly for those who have other responsibilities. These are usually highly transactional tasks that deal with issue resolution and areas where staff needs to engage in a lot of repetitive work.

Finally, as controller or CFO, AI is going to lead you to delegate some of your own efforts—strategic execution, analysis, fund management. And, honestly, that might not be a bad thing as you up level your own career.

Chen Amit is the CEO and Co-Founder of Tipalti.