In their Q3 2016 earnings report, Bank of America reported growth in their online and digital banking adoption to 21 million mobile banking active users.
"That’s better for customers, it’s also better for our shareholders," said Paul Donofrio, Bank of America CFO. "It’s one-tenth the cost of walking into a branch.”
The key to mobile banking continuous quality success is automation.
The Bank of America example is not unique and is in line with the trend toward digital banking interactions over in-person banking. In fact, over 87% of those ages 15-35 use mobile banking according to a quarterly consumer trends survey of 3,000 U.S. banking customers, conducted by Harris Poll for Fiserv.
Consumer banking at the branch is declining and banks are reducing their staff and branches to compensate. What’s even more telling in support of this trend is that 41% of the survey participants said that they would leave their wallet at home if they could store all the information they need in their mobile wallets. What’s interesting is that 30% of participants in a USC Annenberg Center survey said they would always drive home if they forgot their phone, regardless of distance.
Historically, most users limited their mobile banking interactions to checking account balances and statements. Today, more complex tasks are accomplished, such as check deposit via scanning, making payments and money transfers. In fact, 37% of consumers have cited that they leverage their bank mobile app to deposit checks.
While the digital trend is fascinating, we must ask what the inherent implications are from an application quality perspective? In our experience, 20 to 30% of test cases now include mobile deposit user flows. If you multiply that by all smartphones and tablets that support mobile banking technology, that’s a lot of executions (whether it’s smoke or regression testing).
Fingerprint technology is helping accelerate mobile banking adoption and establishes a strong sense of trust in the security of the app. Given its prominence, ideally, 30%-40% of test cases should leverage fingerprint authentication as the login mechanism or approval for in-app transactions.
In our experience, however, most of the mobile testing for these use cases is done manually due to time and resource constraints. Manual testing can’t possibly cover a complete set of testing scenarios such as smudges, different rotations, inverted check images, rescanning of checks that were already deposited, verifying fraud protection, validation of prints that were made by multiple users and more.
It goes without saying that incomplete test coverage impacts release frequency which can breed disgruntled users who are both noisy and critical.
The bottom line is that there must be a way to automate these test cases. Test scenarios should be driven by different images and prints, with proper validation, and testing should occur across all devices the bank supports. It must be part of the regression test, at a minimum, so that bugs are quickly identified. To achieve true continuous quality, an always-on cloud of devices is necessary.