From rivets to reconciliations: how RPA saves financial firms money

March 13, 2017



Like the auto industry before it, financial services is turning to digital labor to cut costs.

In the 1990s, US car companies faced issues that should sound familiar to financial institutions. Margins were shrinking and foreign manufacturers were pressuring US automakers to make a better car for less money. So manufacturers examined  how they were currently going about making cars and trucks. Faced with high labor costs, they began to shift some production offshore and double-down on automation. They also looked deeply into their processes and found they were often doing work that was simply unnecessary. They began asking themselves, “Is there a better way to do this?”

Financial services companies didn’t embrace this sort of self-examination for a variety of reasons—until now. Today, margins are shrinking and financial institutions are facing competitive threats, but unlike the automakers, the threat is home-grown. FinTech firms are finding ways to offer services that are very similar to what banks and insurers sell—at a fraction of the cost. And consumers, who have seen this story play out in other industries, ask themselves, “Why do I need a money manager when a robot can just follow the stock market much more cheaply?” To find a way to adapt to these changes, financial services companies should look internally and learn how to do more with less.

A secret sauce to cutting costs?

Many financial institutions already have consolidated operations, dropped business lines, moved some functions offshore, and set up shared services centers. So where can they find the next round of savings? Again, we can look at manufacturing. Automakers found many repetitive tasks in their processes that could be automated. Financial institutions have more repetitive tasks of their own than you might think, and while the work is more sophisticated than you’d see on an assembly line, digital labor has evolved considerably. For example, to fulfill a simple reporting request, some companies have people manually reconciling data across multiple systems front to back. Until now, automating this kind of work would have been impractical because it required a lot of IT heavy lifting from already constrained budgets.

Enter robotic and intelligent process automation (RPA/IPA).

Not your father’s robot

RPA refers to small, easy-to-program software tools that can automate a range of digital activities. These robots execute pre-programmed rules on mostly structured and some unstructured data. Some robots can even learn from prior decisions and data patterns to make decisions by themselves. In our 2016 survey of RPA use in the financial services industry, 57% of respondents told us they’d done one or more proof-of-concept projects but hadn’t done a broad implementation of the technology. In 2017, this is likely to change.

Get the most out of your RPA

RPA can help you spend less time on tasks that don’t add value, but you have some homework to do if you want to get the most out of it:

  • Understand your process. If you blindly apply RPA to a bad process, you won’t receive the kinds of benefits that you should. By redesigning a process, even with only minor changes, you can make RPA far more effective.
  • Center yourself. As your RPA gains momentum, you may find that your teams are using it in completely different ways. This is fine for now, but once you move to scale, you should make sure you bring the teams together to share what they’ve learned. Then you can apply RPA in a way that makes sense for all units while still maintaining the nimble, agile benefits of the technology.
  • Understand what RPA can do and what it can’t. While there are many ways RPA can help, it’s not a replacement for an IT transformation project. Additionally, RPA is still software, and, like any other program, it requires good documentation and effective controls.

Identify your needs

If US automakers didn’t focus in on their process, they wouldn’t have solved the quality problems with their cars and trucks. At the same time, if they didn’t find new ways to automate the process, they would have been sunk by high labor costs. Surprisingly, financial institutions’ issues are extremely similar. In 2016, the top firms focused on processes to eliminate needless work and find areas to automate. In 2017, we expect more firms will discover the value of applying digital labor.

For more information on this topic, PwC has an ongoing series on RPA and digital labor. You can find out some of the basics with our Q&A on Digital Labor. To understand more about how you can bring digital labor to your firm, you can read our paper From theory to practice: Onboarding digital labor in financial services. Lastly, once you have digital labor implemented, Payback time: Improving ROI from digital labor in financial services discusses how to get the the most value out of RPA.

Read more about other issues facing the financial services industry in our report, Top financial services issues of 2017: Thriving in uncertain times.



Chris Curran

Principal and Chief Technologist, PwC US Tel: +1 (214) 754 5055 Email

Anand Rao

Global Artificial Intelligence Lead, PwC US Tel: +1 (617) 530 4691 Email