Understanding the Meta, Comcast, and LinkedIn Outages

Customer Stories

Gaining End-to-End Application Visibility at a Top 10 Financial Institution

By Sean R. Coombs
| | 10 min read


Gaining End-to-End Application Visibility at a Top 10 Financial Institution

Modern consumers have significantly different expectations for their financial institutions than they did five years ago. As customer experience has shifted away from personal interactions at brick-and-mortar locations to digital interactions distributed across web services and mobile applications, consumer loyalty and their continued patronage has become increasingly ephemeral. As such, the customer digital experience has become a make-or-break proposition for even the largest financial service companies.

At a recent ThousandEyes Connect event in New York City, we heard from a VP of Advanced Technical Services at a leading enterprise financial institution about the importance of the digital customer experience for their business—and what they are doing to ensure their IT operations are aligned with this growing business need. In this blog, we’ll share a summary of their presentation.

Challenges to Delivering Digital Experience

As the financial industry becomes more and more dependent upon a positive digital experience at both the consumer and commercial levels, any disruptions in service can affect customer loyalty and hurt the potential to attract new customers. While the criticality of resolving customer-facing issues is well understood internally, the organization faced a number of obstacles that made it difficult to achieve its goal of providing a seamless digital experience.

One of the logistical challenges this organization faced with respect to gauging application performance was getting a proper problem definition in order to recreate and trap the issue. When tech support was called in to triage an issue impacting consumer facing applications such as their e-banking portal, they had to rely on secondhand information provided by bridge-line participants to determine the appropriate troubleshooting approach. Oftentimes, this secondhand information was incorrect or incomplete, which hampered recovery efforts. Needless to say, the team spent a significant amount of time troubleshooting application issues, which in turn prolonged the MTTU, MTTI and MTTR metrics that they were accountable for.

On the technical side, a lack of application awareness was a major challenge they faced with respect to gauging performance. Over the years, the team had no insight or knowledge of what applications were running over the network, and because they were not aware of them, they had no idea what “normal” looked like with regards to availability and performance. Without advanced knowledge of an application’s behavior profile on the enterprise network, every triage instance soon became a fire drill with regard to data collection, analysis by attempting to reproduce the customer issue. In an ideal world, they would have a good understanding of steady-state performance and availability for all of their applications to benchmark against while having a true representation of the customer experience from their customer base.

Gaining End-to-End Application Visibility

In order to become proactive with regard to the awareness of the applications that live within the enterprise network but are accessed by an external customer base, the organization implemented a host of monitoring and management tools. By gaining a multi-dimensional view of the end-to-end application experience, they can better simulate a user experience that is truly representative of its customer base.

In addition, the team wanted to utilize a complementary toolset in order to baseline application performance, understand application behavioral characteristics and provide end-to-end application visibility. They chose to stitch together three perspectives of application performance using best-of-breed providers: ThousandEyes for the external or Internet view (Figure 1), Netscout for packet capture and wire-data analysis within their enterprise DMZ zone and AppDynamics for the internal view. Using these three tools, they can gain a good understanding of KPIs for performance and, subsequently, receive alerts when performance suffers.

Monitoring Application Performance
Figure 1: Monitoring Application Performance

In addition, by pre-configuring the critical applications within this complementary toolset, network operations personnel can be highly “aware” of the applications on the network, and they can review meaningful data and analyses that were previously unavailable.

Benefits of End-to-End Application Visibility

Having true end-to-end application visibility benefited the bank in several ways. They can abandon, to some degree, the traditional top-down or bottom-up approach of triage because they can see the performance of all layers in one pane of glass. Now, they can clearly separate out the network layers and reduce the overall time it takes to define a problem and reach a solution. Before using ThousandEyes, initial triage took 1.5 to 2 hours, on average. Now, the initial triage takes about 20 minutes.

"Before using ThousandEyes, initial triage took 1.5 to 2 hours, on average. Now, the initial triage takes about 20 minutes."

Furthermore, because different tech teams are now able to view and understand the same data, rather than using disparate technology, everyone can agree the data is integral and complete. While this may not totally eliminate the “It’s the Network” battle cry, the phrase is uttered less emphatically around the organization. In addition, the data visualization and dashboards within ThousandEyes are user-friendly and can be understood by a cross-section of support groups, including many managers at the bank.

Lastly, the alerting aspect of the platform enables the bank’s team to truly understand the important questions that come up during production-impacting issues: “Who is affected,” “What is affected,” “Where is the problem located” and “When did the issue begin and is it ongoing?” The team is also using reports to provide insight into positive and negative trends with respect to application performance. These reports (Figure 2) are then utilized by support personnel in order to provide management with a “State of the Union” perspective for critical applications and services.

Application Performance Reports
Figure 2. Application Performance Reports

Going Beyond Monitoring Applications

The team at this financial institution recognizes the importance of ThousandEyes to its overall success. To better understand other critical processes at the enterprise, the team has deployed ThousandEyes in other ways, such as:

  • Monitoring third-party DNS dependencies to track SLAs. Over the years, the bank contracted out its DNS to external hosted third-party vendors. Using ThousandEyes, they are able to monitor the availability and performance of those third-party services in order to hold those vendors to the agreed-upon SLAs.
  • Monitoring the integrity and security of external-facing entities. Due to a large number of mergers and acquisitions over the years, the bank had accumulated a good amount of IPv4 space that was not being used. Using ThousandEyes, they are able to monitor those entities to make sure they are not hijacked, intentionally or otherwise.

It Started with the Network

Using ThousandEyes, the application visibility this financial institution was able to build has paid dividends. Some of the outcomes include improved maintenance, activity processes, enhanced component monitoring, and identification of gaps in monitoring in networks outside of the bank. In addition, using the new methodology has enabled them to identify several issues before bridge lines were opened in response to either customer complaint or Line of Business synthetic application performance monitoring.

In the future, the team is looking ahead to continue to develop proficiency in this next-generation triage and troubleshooting approach to the point Advanced Technical Services becomes the de facto center of excellence in this regard for the bank. They also hope to spread the adoption and increase the user base to all interested application support personnel so that they can more quickly identify issues with either application code or issues with components.

Digital customer experience doesn’t stop at the edge of your corporate network. Learn how we fill in performance monitoring blind spots in our eBook, “There’s a Cloud-Sized Hole in Your Monitoring Stack.”

Subscribe to the ThousandEyes Blog

Stay connected with blog updates and outage reports delivered while they're still fresh.

Upgrade your browser to view our website properly.

Please download the latest version of Chrome, Firefox or Microsoft Edge.

More detail