What is the most efficient way to test new ideas and how long does this “incubation period” last? Banking and fintech executives in discussion at the Bank Innovation Build virtual conference last week found the answers may not be clear-cut.

“I’ve been surprised how varied the paths have really been,” said RJ Sherman, vice president of innovation at the $176.7 billion Citizens Bank. “It’s been some flavor of customer research, product development and UX design, and that experimentation mindset. You really have to be flexible in the process.”

Still, there are a few key elements of the incubation period that are common among organizations, despite varying processes.

For Toronto-based RBC, which has $1.27 trillion of assets, the incubation period is defined by the question: What are you trying to answer through this period? This, according to Rami Thabet, vice president of digital product at the bank.

“It may be technical feasibility, it may be ‘Will this fly?’ Oftentimes, what is most often ignored — that we found this process to be very helpful — is ‘What is your product business and operating model that will be required to sustain this?’” Thabet said. Sometimes, he added, the questions addressed during this incubation period will result in operating model changes, product changes or full-service changes.

Once RBC determines whether it has the prerequisites required to sustain the venture, it comes down to understanding what changes need to be made and “aligning the organization and the internal change management to create the environment for success,” he said.

For TD, which has $1.1 trillion in assets and 26 million customers, the research portion of the process is critical, especially for novel business ideas, according to Imran Khan, head of global digital experience.

The digital team has conversations with customers on a daily basis, Khan said, adding that through the pandemic the Toronto-based bank has maintained contact virtually. “You don’t want to be in a situation where you are building the whole thing and launching it, and then find out that maybe there was something you didn’t know or didn’t see,” he said. TD conducts research with a simple prototype that isn’t a big technology build or connected to the core, in order to identify blind spots with very little cost or investment, he said.

Sometimes, if you take the time to be confident in your research, Khan said, you can skip the proof of concept and accelerate the launch of a minimally viable product, then iteratively learn about the product in the marketplace.

Connecting the innovation to an enterprise solution is an important step in the incubation process, according to Vincent Bezemer, head of strategy in North America of the Amsterdam-based Backbase. Backbase provides banking software to financial institutions globally and works with Barclays, Credit Suisse, KeyBank and Lloyds Banking Group, among others.

Simply put, “if the enterprise capability isn’t there or primed to deal with it, you’ll just have a hard time implementing it,” Bezemer said. A strong intrapreneurial and startup culture can help encourage this process, he added.