Credit unions may be slower to adopt technologies like artificial intelligence (AI) and machine learning (ML), but they have the benefit of learning from the successes and pitfalls of the financial institutions that have gone before them.

Credit unions must focus on implementing technology quickly and efficiently, according to a recent panel discussion with credit union executives and consultants hosted by AI solutions provider Neuton.AI.

Although they face barriers such as cost and resource limitations, credit unions can be technology innovators if they prepare appropriately, said Todd Lindemann, senior vice president of payments at $6.9 billion Redwood Credit Union. During his tenure at the Santa Rosa, Calif.-based Redwood, he worked to bring a new digital banking system to the credit union and currently leads monthly updates for online and mobile banking.

Redwood established a member feedback channel to seek out challenges that its members may be facing with its banking system. The credit union found that 10% of its members were hitting limits while trying to withdraw cash. Redwood employed an AI solution to assess all ATM transactions and predict which customers are likely to need higher limits, and adjust if necessary.

“We are going to have ATM limits actually going up to $10,000, and that’s pretty unheard of in the industry,” Lindemann said. The credit union is looking to expand the limit flexibility beyond ATMs to cards and point-of-service transactions as well, he added.

Redwood Credit Union also added a self-service feature to its mobile app that allows users to view their ATM limit and increase it, if necessary. The higher limit stays in place for three days, then reverts to the default. “That’s just an example of how we used data and AI to build out an innovative process for our members,” Lindemann said.

Bringing innovation to production

While credit unions may not be first in banking technology implementation, that can also be an advantage, panelists noted.

“Credit unions typically don’t have the risk appetite or the financial war chest needed to be cutting edge, and they’re typically going to fall into those fast follower or early majority phases of innovation adoption,” said Jay Lauer, senior innovation strategist at credit union service organization PSCU.

This allows credit unions to study what other financial institutions have done with new technologies and approaches in order to determine what’s right for them, Lauer noted. The key is to move swiftly when adopting new tech solutions.

“We need to grease the skids a little bit, since we may not be the fastest to adopt new technologies and new processes,” Lauer said. “We need to figure out how we can go faster and how we can bring innovation to production more quickly and more effectively.”

Embrace data for everyday decisions

Cost is a substantial barrier to credit unions obtaining and implementing technology such as AI, said Blair Newman, chief technology officer of Neuton.AI. Once technology is purchased or produced, it still must be applied and its planned value realized, he said.

Seventy percent of digital transformation efforts at organizations, including credit unions, fail to deliver on their promised value, according to research from Boston Consulting Group.

“I see in my conversations with credit union leaders across the nation on an ongoing basis that [not delivering on perceived value] is so true,” said Naveen Jain, founder and president of credit union business consulting firm CULytics and former vice president of data analytics at $14.1 billion First Tech Federal Credit Union.

Credit unions need a “credible, comprehensive vision” of what to achieve with data projects, Jain said, and only then can they move forward from the perspectives of data management and governance, business intelligence, AI and ML to achieve that vision.

The average credit union has between 60 and 100 data systems and must choose which of those to use to assess its operations, said Anne Legg, business strategist with credit union consultancy Thrive Strategic Services. Using data, credit unions must craft a clear roadmap of business challenges to solve, she added.

In addition, credit unions must have consistent data governance and reporting, Redwood’s Lindemann noted. “Define your data and get a good dictionary,” he said, so that when implementing AI and ML, the organization can “embrace” data and use it in everyday decision-making.