Failures in IT never seem to go away. Every month the news seems to be covering a new, high-profile IT issue, many of which Jay Liebowitz enumerates in his recent IT Professional article, “IT Project Failures: What Management Can Learn.” HealthCare.gov, The National Program for IT (UK), United Airlines, the Wall Street Journal, and even the New York Stock Exchange have all experienced very public, very high-profile failures. Why?
As Liebowitz notes, many organizations blame these failures on scope, cost, and time, but they can be further categorized by process-driven, context-driven, and content-driven issues.
- Process-driven: Those related to business planning, project management and control, strategic formulation, and the change-management process.
- Context-driven: Those related to the information system itself (technology, system design).
- Content-driven: Those related to the environment in which the project is being development (culture, structure, management).
One process-driven issue that is often the impetus for failure is when organizations pick the project they like the best – without taking into account the business case for that project. If you’ve been in IT long enough, you’ve seen this happen. Needless failure.
Liebowitz also points to research from Rateb Swies asserting that IT projects fail due to the following reasons (in priority order):
- High degree of customization (context-driven)
- Changes late in the design stage (process-driven)
- Underestimating the timeline (process-driven)
So the question, of course, is what can management do? An obvious solution, suggested by Liebowitz, is to better educate current and future IT project leaders. Specifically, I recommend the following:
- High degree of customization (context-driven) – Management needs to better understand the size of the project. Function points can help.
- Changes late in the design stage (process-driven) – Management needs to understand the impact of these changes. Better IT governance can help.
- Underestimating the timeline (process-driven) – Management needs to understand the probability of the estimates that they are getting. Better estimating using statistical analysis and, yes, a tool can help.
Liebowitz also suggests better coordination between business users, IT, and finance. I’ve talked about this on the blog repeatedly – the need for improved coordination between IT and the business is paramount.
We have a number of services that support this type of improvement, including Function Point Analysis, IT Governance, and the Measurement Roadmap. IT serves to support the goals of the business; if it’s not appropriately aligned with the interests of the business, then it’s not fulfilling its role and failure will be more likely.
Management has the ability and the tools available to prevent failure – so when will they start doing their part?
Read, “IT Project Failures,” here.