Why do IT projects fail?
According to a 2013 survey by Innotas, half of business had an IT project fail in that year [1] Other research by Forrester Consulting [2] found 39% of the 474 executives surveyed believed their internal IT organizations have the ability to regularly deliver projects on time and on budget. Similarly the Standish group report that only 30% of projects meet their goals.
Examples of project failures include £34 million in IT costs written off by the department of Work and Pensions, £148 written off by the Co-Op Bank on a cancelled system and the cancellation of a system by the USAF cost $1.1 billion [3].
Half of the executives in Forrester survey claimed their department where overburdened. Other issues such as a lack clear executive direction, Development talent and a lack of stakeholder consensus where cited as problems by around a third of respondents. Innotas found that IT departments had an increasing burden to deliver even when described as overworked and understaffed. Having no way to prioritize projects was also mentioned as an issues.
There are various causes for these failures. According to Henry Mance [3] writing in the FT, They include, poor forecasting of costs, poor communication between people and organisations, projects being overambitious in what they set out to achieve with a lot of upfront planning and an unwillingness to admit to when things are going wrong. The "black swan" scenario is also not anticipated, that is, something entirely unexpected and with a big impact may occur.
What seems to get blamed the most, is the shifting of the goal posts. That is changing a project's specifications after it has started. 56% of those executives surveyed by Forrester blamed changing specifications for causing problems. Other factors reported include, 34%, believing they lacked clear executive direction, and another 34% point to a lack of the right development talent. A similar number, 32%, cite a lack of stakeholder consensus.
The attitude of business executives is described as regarding IT as a cost centre and not a "value-add". This according to an article by Joe McKendrick [2] reflects Organisational Dysfunction not problems with the technology. McKendrick suggests that this alignment between business and IT will improve as a younger generation increasing enter positions of responsibility, members of the so called Generation X and Generations Y who were raised on computers and more inclined to be supportive of the use of this technology. There may be a disproportionate emphasis on the failures and over-runs instead of the long term benefits, making IT be regarded as cost rather than a benefit. The Failure of a project may be a matter of perspective.
Solutions to project failure include the use of Cloud Computing to simplify projects by removing considerations for local hardware. The Forrester Consulting report recommends Integrative thinking. This is considering the wider impact value and user experience added by applications to the business. This requires both an understanding of technology and the business.
Be Agile
The adoption of Agile methodologies [4][5]is the biggest solution put forward to make IT projects more resilient. The other more traditional approaches are known as waterfall methods, these are top-down cascades where projects are broken down in to distinct stages. Typically in a waterfall project, stages progress from analysis, design to coding, testing and implementation. Waterfall frameworks include SSADM and PRINCE 2. These methods seek to provide governance, accountability and documentation to any project. Without these there is no plan to following in order to guarantee meeting specification. These things are not forsaken in agile. Compromises over documentation and testing lead to a technical debt, which if not addressed sooner rather than later lead to problems over time. Forecasting, prioritising and quality are all found in the agile frameworks.
In the waterfall approach it is assumed requirements will remain fixed, however this is rarely the case for IT, which often operates in a complex and chaotic environments. The Agile approach by contrast expects change. This works by breaking projects into smaller chucks called iterations or sprints which allow for feedback loops, allowing for ad-hoc changes. Value gets delivered continuously and earlier. With Agile there is less design upfront and an emphasis in people over processes to empower greater collaboration and communication.
The Agile Manifesto [6] set out the goals and principles behind agile software development. Written by representative of agile in 2001 who together formed the Agile alliance. The Manifest sets out the common principles. Agile methods for software development Include, Dynamic Systems Development Method (DSDM), Extreme Programing and SCRUM. These approaches to IT projects have the origins in Japans post-war industrial renaissance. Manufacturing improvement programmes such as Kaizen, Lean and Kanban have influenced frameworks for IT development. The Japanese systems tend to include improvement cycles, empirical process from which lesson learned from experience are put into practice.
Is Agile Better
Agile versus waterfall has been researched. The Standish group compared the two against a database of IT projects undertaken between 2002 and 2010. Classifying them as either successful, Challenged or Failed.
| Successfully | Challenged | Failed | |
| Waterfall | 14% | 57% | 29% |
| Agile | 24% | 49% | 9% |
Adopting Agile has shown to make projects more resilient and successful. It is important however to bear in mind that Agile is not a set of rituals to be followed in order making you buzzword compliant. This, almost by definition is not agile. Instead it can involve a difficult shift in organisational culture. It is attitude that is the key being agile.
Next:
The 7 Heresies of Agile
Read More
1. Why Are So Many IT Projects Failing? Sharon Florentine, December 2013 - Last retrieved 05/12/2015
2. Only 39 percent of IT projects successful? That's a good start Joe McKendrick October 24, 2013 - Retrieved 05/12/2015
3. Why big IT projects crash By Henry Mance, Sept 18 2013 - Retrieved 05/12/2015
5. Open Source and 'Agile' methods gaining political support
6. Manifesto for Agile Software Development
March 2016
| Duncan Stonebridge.
Back
