Customer expectations are naturally high when it comes to dealing with money.

The challenge for issuers is to deliver consistently excellent (fast, reliable) experiences across all channels, including mobile sites and apps, websites, call centers and in person.

However, substantial and ever-growing technological complexity is often a major impediment.

Driving routine activities out of in-person venues to to self-service digital channels like the desktop web can yield substantial cost benefits. Digital channels are also more likely to delight and less likely to annoy customers, meaning that besides cost savings, issuers benefit from higher customer retention, satisfaction and engagement.

Amazon motivational sign
In 2017, an Amazon S3 outage brought down or caused problems for thousands of websites, according to Catchpoint's Mehdi Daoudi. Bloomberg News

However, poorly performing (sluggish, unreliable) websites are a major concern, especially as these sites come under heavier traffic loads. A single external third-party element can cause problems for the entire site. In late February, an Amazon S3 outage brought down or caused problems for thousands of websites. Just last month, another S3 outage caused web slowdowns for many businesses for several hours.

The internet infrastructure is more interconnected and fragile than we like to think. A major service outage can happen anytime, anywhere, generating a substantial ripple effect. This means that even if your own data centers and servers are running optimally, that’s not a guarantee of fast, reliable web visitor experiences.

The only way to address this uncertainty is to constantly monitor web page load time, around the clock and from key user geographies (since web performance can vary significantly from one region to another, based on the multitude of performance-impacting variables such as cloud service providers, CDNs, and regional and local ISPs). This enables proactive issue detection and gives banks an opportunity to address the problem, ideally before customers are even aware it exists.

Mobile has evolved into a channel of its own, with adoption rates in banking similar to retail. According to a 2016 Verizon study, 55% of U.S. adults have used a mobile app to make a bank transaction, versus 56% who have bought something online using a mobile app. Banks and retailers need to pay special attention to their mobile websites and applications, even prioritizing mobile over desktop. Consumers usually have their mobile devices on hand, allowing them to interact with their accounts, making payments, depositing checks or viewing previous transactions, at any time.

Mobile websites and apps are prone to the same types of external third-party performance flubs as desktop-based websites; however, the constrained nature of mobile networks tends to result in an even worse impact on the user experience. As with traditional websites, the speed and reliability of mobile sites and apps must be monitored around the clock, across various user geographies and scenarios (for example, 3G and 4G networks).

The golden rule in mobile is to keep sites and apps as lightweight as possible, avoiding unnecessary images, files or scripts that can make these digital properties heavy and slow them down. It is important to design from a mobile-first perspective, a radically streamlined site with simplified transaction processes.

Personal channels such as call centers remain a key interaction point with customers, who often reach out to them in emergency or stressful situations, such as reporting a lost or stolen credit card. They also tend to report technical problems with websites and mobile apps. This makes this particular channel prime for annoyance when call center agents don’t have fast, reliable access to their internal banking and account systems.

Much like customers, call center agents are based in many locations and the speed and reliability of their business applications can vary greatly. This is because performance degrades as the distance between call center agents and the primary data center increases. Continuous testing and monitoring of applications used by call center agents is necessary, but gaining an accurate view depends on measuring performance from the closest possible vantage point to each call center operation.

There are techniques available to make this an easier endeavor, including a nodes-based approach within individual call centers, which identifies internal, location-specific problems and enables fast, accurate remediation.