February 19th, 2014
A number of common IT projects that seem like they should add value rarely do. Here are what I consider the top IT projects that waste budget dollars.
The role of the CIO has changed more in the past five years than any other position in the business world. Success for the CIO used to be based on bits and bytes, and is now measured by business metrics. Today’s CIO needs to think of IT more strategically and focus on projects that lower cost, improve productivity, or both, ideally.
However, many IT projects seem to be a waste of time and money. It’s certainly not intentional, but a number of projects that seem like they should add value rarely do. Here are what I consider the top IT projects that waste budget dollars.
Over provisioning or adding more bandwidth
Managing the performance of applications that are highly network-dependent has always been a challenge. If applications are performing poorly, the easy thing to do is just add more bandwidth. Seems logical. However, bandwidth is rarely actually the problem, and the net result is usually a more expensive network with the same performance problems. Instead of adding bandwidth, network managers should analyze the traffic and optimize the network for the bandwidth-intensive applications.
Investing in fault management tools
On paper, it makes sense to invest in fault management. You deploy network devices, servers, security products, and other infrastructure, so of course you would want to know when devices are up and down. However, the fact is that today we build our network so redundant that the loss of any single device has little impact on the performance of applications. Also, most of the fault management tools have a big blind spot when it comes to virtual resources, as the tools were designed to monitor physical infrastructure. IT organizations should focus on performance solutions that can isolate what’s been “too wrong for too long” to solve those nagging “brown outs” that cause user frustration.
Focusing IT energy only on the “top talkers”
When I talk to IT leaders about new initiatives, it seems much of the focus is on the top 5 or 10 applications, which makes some sense conceptually as these are the apps that the majority of workers use. Instead, IT leaders should monitor all applications and correlate usage to business outcomes to determine and refine best practices. For example, a successful branch office could be heavy users of LinkedIn, Salesforce.com and Twitter. In aggregate, these might not be among the company’s top 10 applications, and the usage would fly under the radar. If organizations could monitor applications and then link consistent success to specific usage patterns, unknown best practices can be discovered and mapped across the entire user population.
Using mean time to repair (MTTR) to measure IT resolution success
ZK Research studies have revealed a few interesting data points when it comes to solving issues. First, 75% of problems are actually identified by the end user instead of the IT department. Also, 90% of the time taken to solve problems is actually spent identifying where the problem is. This is one of the reasons I’m a big fan of tools that can separate application and network visibility to laser in on where exactly a problem is. This minimizes “resolution ping pong,” where trouble tickets are bounced around IT groups, and enables IT to start fixing the problem faster. If you want to cut the MTTR, focus on identification instead of repair, as that will provide the best bang for the buck.
Managing capacity reactively
Most organizations increase the capacity of servers, storage or the network in a reactive mode. Don’t get me wrong, I know most companies try to be proactive. However, without granular visibility, “proactive” often refers to reacting to the first sign of problems, but that’s often too late. Instead, IT departments should understand how to establish baselines and monitor how applications deviate from the norm to predict when a problem is going to occur. For example, a baseline could be established to understand the “normal” performance of a business application. Over four successive months, the trend could be a slight degrade of the application’s performance month after month. No users are complaining yet, but the trend is clear, and if nothing is done, there will be user problems. Based on this, IT can make appropriate changes to the infrastructure to ensure users aren’t impacted.
The IT environment continues to get more complex as we make things more virtual, cloud-driven or mobile. It’s time for IT to rethink the way it operates and leverage the network to provide the necessary visibility to stop wasting money on the things that don’t matter and start focusing on issues that do.