Over the past couple of months almost every aspect of our lives has changed. Many of our day-to-day activities have shifted from tactile real-world experiences to digital alternatives. Our home, work, social and commercial needs are now increasingly dependent on applications which may not have been designed to handle the sustained increases in demand thrust upon them.
We’re hearing a common theme from our customers: “Every day is like Black Friday” – massive increases in application traffic which puts significant strain on the infrastructure and downstream dependencies. However, unlike the traditional Black Friday in November which typically only impacts retailers (and is planned for months in advance), this new workload is being experienced by a wide array of services, from internal enterprise applications, SaaS apps, restaurants, delivery services, manufacturing etc. — they’re all trying to keep up.
Complete visibility is key
To accomplish this, CIOs and IT Managers need clear visibility of their end-users (both employees and customers) and the key transactions they perform against the key applications they rely upon. These may be mobile or desktop applications, 3rd party SaaS applications or internally developed and hosted applications. It is critical to have a Digital Experience Management (DEM) solution with highly detailed distributed tracing of end-user transactions from the client device through to the back-end, whether hosted on-premises, in the cloud or hybrid. End-user activity should be captured in detail, so you know exactly what your users did, how it performed, and precisely what you need to address should problems arise.
Attack the problem from multiple angles
Everything starts with the End-User—and everyone is an End-User. Are your employees able to do their jobs efficiently? Are your customers able to consume the services you provide to them in a timely manner? By monitoring every client interaction, you’ll know precisely how happy your users are and what things need to be improved.
The ultimate goal of effective DEM must be resolution. Simply knowing that you have a problem is only half the battle. What really matters is making that problem go away. To do this you need visibility of the complete end-to-end transaction from the client, through the application servers, cloud infrastructure, and any 3rd party services you rely on. You need a DEM solution which provides actionable detail on the root causes of performance problems regardless of where they lie. I like to summarize the DEM workflow as follows:
- Is there a problem? – If the end users are happy with everything they’re doing, then you’re done. Otherwise there’s more work ahead.
- Where is the problem? (aka Who needs to fix this?) – The end-user’s device? Their Wi-Fi/network? The application code? The servers? Cloud hosting? A SaaS vendor? A 3rd-party dependency?
- What is the problem? (aka What needs to be done to fix this?) – If the issue is something in my organization’s control, provide precise actionable detail to lead to a resolution. If it’s due to a 3rd party, then provide clear evidence that the issue is on their side so that they take responsibility.
- Socialize the resolution. – It’s important that we not only learn from our mistakes but are also recognized for our successes. Incident solutions should influence future architecture designs and QA methodology to ensure similar problems don’t happen in the future. Findings can be used to hold 3rd party vendors to their SLAs. If leveraged properly, reports on the reduction of incidents and downtime can promote the positive business impact your team brings to your larger organization.
Maintain visibility during Black-Friday-like volume
Many organizations struggle with maintaining visibility when they use tools that rely on sampled approaches, which only capture a small percentage of traffic to provide overall summaries. This small percentage gets significantly smaller as traffic volume increases, which creates a serious dilemma: You have the least visibility when you need it the most. A better approach would be to capture ALL transactions end-to-end, with code-level detail—even if you have billions of transactions daily. An observability-based model that captures granular detail and enables more actionable, informed remediation for every alert, so the symptom can be quickly and clearly traced to the root cause and rapidly resolved.
A new way of thinking
Companies are realizing that they must rethink their definitions and measurements around business resiliency. This crisis has made it increasingly clear that humans are critical to the success of the business. As our customers place more demand on us, and we, in turn, place more demand on our employees and partners, our business resiliency is inherently connected to the visibility we have into the digital interactions these end-users have with us. It’s no longer sufficient to measure user experience “on average” – every interaction by every user must meet their needs and expectations – for their satisfaction and for the enduring success of the business.