Overseeing an e-commerce platform for one of the world’s largest technology companies is no simple task. At the heart of it is ensuring that every enterprise we support can provide its customers with the best digital experience possible, wherever they are in the world.
In the ecommerce space, I’ve seen a mindset shift in business — from concerns around monitoring availability to using observability to ensure excellent performance and doing everything you can to prevent your sites and services from being slow. People don’t like slow. No one wants to waste their time, and if a customer runs into a problem as a web page loads or during the checkout process, there are many opportunities to spend money elsewhere.
One thing that makes the task so complicated for the enterprises we support is the sheer number of components that must work together flawlessly to provide an exceptional experience. Yet understanding all the various factors that impact performance — backbones and networks, CPUs and RAM, APIs and CDN providers, not to mention all the payment systems your APIs need to connect to around the world to work — can become overwhelming.
One glitch anywhere within the technology framework can undermine the performance of the entire ecommerce system. Troubleshooting the situation can quickly become a nightmare, mainly since most of those components are beyond your direct control and outside traditional lines of visibility. With the wrong monitoring tool, staff would have to spend hours or even days reacting to a situation to get things back on course when a problem popped up.
Today, thanks to end user observability, like that which we rely on Catchpoint for, and other tools such as Dynatrace for application performance monitoring (APM) that we bring together in a single dashboard, we’re able to diagnose and resolve performance issues in minutes. As a result, we can ensure the highest level of site performance for our entire global enterprise base and strategically use our staff and resources.
Pushing Performance to a New Level
Monitoring systems and performance at a granular level is critical in today’s high-stakes business environment. Customers demand convenience, flexibility and speed. When they visit a site or service, they aren’t concerned about the source of a problem; they need things to work and work quickly.
In the past at SAP, we sometimes found ourselves in reactive mode. In many cases, a problem became apparent only after dozens or even hundreds of customers encountered an issue. By then, the damage was done. In the cases where we conducted monitoring, it was mainly a hit-or-miss proposition.
Powered by a robust partnership with our observability vendor, that is no longer the case. If we detect a baseline deviation, we can take a proactive stance and ensure that all the various components are in sync and functioning correctly before any problematic element wreaks too much damage. Continual observability and the option to harness a vast range of different synthetic test types means we can detect issues immediately. We can also understand whether an undesirable issue was a temporary glitch or a recurring event.
Reliable predictive analytics also help us with seemingly simple areas, like SSL certificate expiry. For example, we can detect when a business partner has a site certificate expiring in 30, 60 or 90 days. We can inform the company in advance so they can act and avoid any possible costly interruption.
All this is integrated into our single dashboard to display the likelihood of different problems through an availability risk score. These predictive analytics might say there’s an 85% probability that an issue is due to a CDN, database or server problem, for instance. With this information, support engineers can get a head start on knowing exactly where to look. As a result, the mean time to resolve issues has dropped from hours to minutes or even seconds.
Unlocking the Keys to Success
The unfortunate truth is that cloud providers and other vendors’ tools for monitoring aren’t enough. They deliver only a partial snapshot of what’s taking place, along with limited tools, means and resources for fixing problems.
At SAP, we have found that there are five best practices for harnessing the power of observability:
- Know what to measure. It may sound basic, but you can’t count the things that matter unless you know what to monitor. Adding to the challenge, you may need to watch different things in different regions or markets. The technology you use must scale, but it must also be capable of finding issues deep within product and vendor integrations.
- Focus on comprehensive visibility. Look for a solution that delivers a comprehensive view of your IT framework, whether it’s an ecommerce site, streaming media delivery service or something else. The days of monitoring CPUs are gone. With hundreds of components dynamically shaping a web page or app — and the user experience — there’s a need for a complete view from the end user perspective.
- Establish a single dashboard. Today, organizations require broad and profound insight that extends across an enterprise, out to the internet and many service providers. A dashboard must accommodate multiple vendors and solutions to deliver end-to-end insights.
- Develop strong vendor relationships. Remarkably, we had the existing networking and application monitoring tools in place several years ago, but we were underutilizing them. It wasn’t until we opened a dialog with the vendors that we realized the full extent of what’s possible with these solutions. Allow time to learn how to maximize features and capabilities — and ensure that you keep up with products and trends. For instance, we recently learned about the super-interesting WebPageTest, which Catchpoint recently acquired, that lets you run speed tests from around the globe using real-world browsers at consumer connection speeds, which we can leverage for our QA processes.
- Think like a customer. Conditions change. Components come and go. While an enormous technical fabric supports today’s digital experiences, all roads must lead to value for the customer. That’s what creates an exceptional customer experience. With a single view and the right observability to hand, creating a unique experience isn’t just possible; it’s highly probable. Your brand and sales can only benefit.
Visibility is a Best Practice
Finger-pointing and excuses aren’t acceptable in today’s business environment. Whether a company operates in B2B or B2C marketplaces, it’s critical to ensure that you’re working at peak performance.
The right type of observability, used wisely, can deliver a comprehensive view — and the information needed — to anticipate problems and take immediate steps when they occur. The result is a business that boosts performance, trims costs and enables IT teams to work more strategically. That’s a recipe for success.
Martin Norato Auer is VP of CX Observability and Automation Foundations at SAP. He is an experienced leader and observability enthusiast. Driven by a customer-first mindset, he takes pride in providing the best observability solution to his customers. As a program manager, his goals include a stable large-scale installation and continuous delivery of new features. In addition to his primary job functions, SAP and customer executives recognize Auer for his extraordinary commitment to delivering a best-in-class observability solution.