Enterprises today are spending billions of dollars to realize productivity and automation gains from the end-user experience. The processes are critical to the bottom-line of companies, and so that’s why many companies today are embracing end-user monitoring (EUM), as part of their comprehensive IT management strategy.

What is EUM?

EUM is systematic methods and processes that assess systems performance from the user’s perspective, and incorporates two types of approaches:

• Active monitoring or End-user Experience Monitoring (EUEM), which uses scripted transactions to measure performance and availability,
• Passive monitoring or Real User Monitoring (RUM), which tracks user interactions with the system

EUM Benefits Many Parties

EUM is critical for web businesses because competition is just a click away if the end-user experience is bad or negative. For example, Dell has learned that speedy resolution of customer issues builds customer loyalty. Dell makes it a practice to quickly isolate the root cause of an issue, for example, discerning whether something is a server problem, a network issue, or a condition limited to the desktop. This is the first step of problem resolution.

End-user experience monitoring is also important for keeping internal users satisfied. While end-users within a company or organization may not complain about every issue, they often complain to co-workers about problems with an application. And eventually user complaints will make their way to upper management, including the CIO.

Lastly, when end-users are content, IT staff can focus on more strategic tasks that bring value to the business – rather than merely responding to interruptions in an attempt to minimize loss of revenues, productivity and profits. By employing EUM, IT then becomes more of a strategic partner to the business rather than just fighting fires.

How big of a problem is end-user dissatisfaction for companies? A 2004 survey of IT professionals by Forrester Research found that:

• Nearly 85% reported experiencing incidents of significant application performance degradation
• In the previous year, 51% acknowledged instances of poor application performance growing more frequent
• Incidents had at least moderate impact on employee productivity (82%), team productivity (77%) and customer service quality (79%)
• More than half delayed launches of new applications because of network performance concerns

Some areas of EUM that are extremely important include:

- usability issues

- network delays

- sluggish servers

- transaction errors.

And positive results from EUM can include:

- A better understanding of application usage patterns before an application goes live,

- Immediate verification of the impact of configuration and tuning parameter changes,

- Easier researching of a problem isolated to a specific user,

- Quicker recognition of a slowdown, outage or other problem via baselines that can be used as benchmarks,

- Through results, better visibility of areas that need more investment to improve user productivity.

Recommendation: Take the Hybrid Approach

I recommend that enterprises new to monitoring start with a hybrid approach, for example, the Monitis all-in-one on-demand monitoring service. A hybrid solution is comprised of a combination of synthetic and real-user monitoring that covers your most critical applications – plus server and network monitoring. Avoid the rather simplistic tactic of considering end-user monitoring as a new source of event-based data to combine with existing management tools you may be using. It’s more practical to frame the data already collected by existing monitoring products in the context of the end-to-end services.

Lastly, a bit of advice on what your end-user monitoring service should be able to do:

• Measure end-user experience across firewalls, load balancers, web servers, application servers, and database servers.
• Provide data in the context of an end-to-end service delivery chain.
• Provide performance data in a standardized way for active, as well as passive monitoring.
• Support round-the-clock tracking for all HTTP transactions, and from all end user locations—not just sampled data.
• Support non-web based applications, too, for instance, VoIP, SMTP, telnet, TCP.
• Enable rapid time implementation to gain maximum value and minimize administrative costs.
• Help you get to the root of problems rapidly.