Many organisations invest vast amounts of time and resources uprooting their systems from traditional platforms into the online space, yet it can only take a second to undo their efforts.
This is according to Michael Allen, sales director for Europe, the Middle East and Africa (EMEA) at technology performance company, Compuware.
Speaking to delegates at a seminar hosted by Mediro ICT yesterday, Allen explained how failing to monitor and manage application performance could lead to glitches, resulting in poor customer experience, negative impact on brand reputation and, ultimately, loss of revenue.
To counter this, Allen suggests organisations should constantly work on their application performance management (APM) processes to detect and diagnose problems and maintain an expected level of service.
"This can be done in a number of ways, namely, examining transactions across all infrastructures and their impact on the end-user - these include the network, bandwidth availability and identifying which Web browser best services an organisation's needs - as these directly affect response times for the end-user," he explains.
"E-commerce Web site performance research shows that online customers are getting increasingly impatient with slow performance. Thirty-three percent of end-users will abandon a site if made to wait; 79% who experience a dissatisfying visit are likely to no longer buy from that site; and 52% state that quick page uploading is important to their site loyalty."
Another key component of APM, Allen believes, is examining the business impact of performance.
To illustrate his point, Allen referenced recent outages at Amazon Web Services and Facebook. "Amazon Web Services experienced a brief outage recently, which lasted only about half-an-hour, yet appears to have cost the company about $5 million in revenue."
Facebook's outage had an enormous impact on a lot of other businesses, states Allen, because many of their Web sites are connected to Facebook through the integration of services.
APM matters now in a way it didn't a few years ago, states Allen.
Modern APM needs to monitor all mission-critical applications, whether developed on-site, licensed from a vendor, installed from open source, or any combination.
"It needs to handle not just basic client-server architecture, but any deployment model, including ones buttressed by reliance on third-party services, public or private cloud hosting, distributed infrastructures, and virtualised everything, he adds.
"Organisations should follow the application performance process from the production phase through to the end-user's experience to ensure problems don't creep in at any stage.
"This means they will be anticipating possible glitches from the early stages and will, therefore, design systems that could detect problems and send out alerts," he concludes.
Share