Pulling back the curtain on app visibility

IT must try to move away from firefighting incidents and strive to solve problems before they become service affecting

Corporate IT infrastructure has never been so complex. There are a host of different options open to organisations, from running on public cloud and on-premise data centres, to SaaS cloud native capabilities and serverless infrastructure. Applications are running across an ever-wider range of technologies and geographies, making it increasingly difficult to monitor and troubleshoot when something goes wrong. Unsurprisingly, companies often struggle with a messy array of different tools and technologies based in these different locations.

Application visibility is one particularly notable pain point. An estimated 80 percent of enterprises have gaps in monitoring their cloud or are totally blind to it. In these scenarios, customer experience (CX) is often hit hard. If an organisation lacks real-time visibility over application performance, the risk increases of gaps emerging between its internal view and actual user perception of how the app is performing.

This can cause damage in several ways. One potential negative impact could be on reputation; if the app goes down, users often express their frustration publicly on social media. Another possible impact, of course, is financial: downtime is estimated to be costing Fortune 1000 companies around 8 per cent of their enterprise revenue – equivalent to nearly $300,000 per hour.

Old world to the new

Amidst this infrastructure complexity and the problems that this causes, old IT processes need to change. Traditional ways of managing incidents saw different teams ringing in on a bridge call. These teams were specialised and looked solely at their own piece of the puzzle, with much of their effort focused on ‘mean time to innocence’. It was hopelessly inefficient and ineffective, with time wasted coordinating between the teams that could be spent finding the optimum route to a solution, and a lack of ownership over issues when these did arise.

This way of working is no longer fit for purpose; IT needs to be more proactive. It must try to move away from firefighting incidents, striving to solve problems before they become service affecting wherever possible.

Application performance management software (APM) is one important way to achieve this goal. Siloed views persist at many large organisations, but by turning to an APM, a business can gain an incisive set of tools to look at all user transactions as they flow through the applications. This, in turn, lets organisations identify when transactions are starting to slow down, allowing them to put the appropriate measures in place before anything gets out of hand.

Indeed, it is important to address long-term performance trends alongside unpredictable crashes or downtime in an app. It can be difficult to troubleshoot for inconsistent problems in applications, resulting in small user concerns about an app’s unreliability potentially snowballing into a fully-fledged crisis. An APM gives organisations early visibility over these problems, tackling performance, and breaking the narrative of unreliability.

Final thoughts

Today’s infrastructure is certainly a challenging environment for monitoring and managing applications. Gaps in knowledge breed uncertainty and run the risk of damage to customer experience and overall business performance. APM is a critical tool to solve these problems.

It can be truly end-to-end, producing actionable insights to respond quickly to issues even in the complexity created by Internet of Things devices. And it can help organisations get to grips with all application layers, spreading full, real-time monitoring capabilities across a variety of infrastructure like API interfaces, Java and .net applications, and even mainframe at the back end. In this way, organisations should be able to bridge the old world and the new.