IT failure, as a whole, is something that will never go away — just like in all professional industries. That’s true no matter how many teams adopt agile strategies or devote their time to pure-intentioned DevOps. The only thing that’s changed, essentially, is the scope of failure. Nathan Sykes investigates.
Past IT project failures might have resulted in serious financial shortcomings, long development cycles, and even vaporware content that never made it past production. Such failures can — and still do — happen. Major IT company IBM succumbed to failure with their non-completed $110 million upgrade for the State of Pennsylvania’s unemployment compensation system. In fact, they were hit with a lawsuit because of that mess up. Oops.
Today’s environment of failure is different in that can be attributed to the many systems and processes implemented by teams. Agile, DevOps strategies, continuous testing and delivery and the fail-fast movement have all made the nature of failure much less impactful. They are essentially all methodologies or strategies designed to help minimize failure and mitigate the impact of any failures that do occur. That reduction only applies, of course, when said methodologies are appropriately adopted.
But that doesn’t mean IT failures have evaporated completely. They still happen, and sometimes when they do, the results are absolutely devastating. In general, IT projects fail anywhere from 20 to 50 percent of the time.
You still want to try to avoid failures as much as possible, though, which requires an understanding of how to prevent them.
- Choose Your Methodology
There’s Waterfall, Agile, Critical Path Method, Hybrid, Scrum and several more. You have your work cut out for you when it comes to choosing the appropriate method. Just take your time, and make sure you understand each approach before deploying it.
- Develop a Risk Management Plan
Understand the journey your team is going to take to complete the project. Where should they be during a certain stage of development? Are they lagging behind? Are they moving faster than expected? Use this information to discern potential problems and deal with them, as soon as possible.
- Taper Expectations
Collaborate with your team, stakeholders and even customers to find the appropriate boundaries. Then, do what you can to encourage your team to push beyond those boundaries, though you shouldn’t necessarily demand it. Allow them to build enthusiasm and enjoyment of their work, which will push them well beyond their plateau.
73 percent of executives admit the projects they undertake are usually “doomed right from the start.” An additional 27 percent always feel this is true with every opportunity they greenlight.
- Implement Continuous Improvement
The general idea is to constantly analyze present work, identify potential issues, solve them or improve whatever needs to be fixed and continue moving forward. Over time, you plan for nearly every contingency, so you have a solution for every bottleneck that may appear. It helps improve the success of a project simply because you’re reducing the risk of failure and eliminating potential obstacles.
- Identify and Focus on Stakeholder Requirements
Whatever the case, you want to ensure that your final product — and thus all the work you do leading up to completion — meets any and all stakeholder requirements. This goal dictates focus on said requirements throughout the entire scope of the project.
Identify those points early by meeting with your stakeholders and then passing relevant information on to your workforce. More importantly, establish a channel of continuous communication between your product development chain and your stakeholders. That way, you can double check any changes or updates.
A whopping 57 per cent of projects fail due to a breakdown in communications. Don’t be a part of that statistic. Follow these guidelines, and you’ll find yourself much closer to success than failure.
Bio: Nathan Sykes is a business technology writer from Pittsburgh, PA. Check out Nathan’s blog to read his latest articles.