22 March 2012

Setting up IT projects for failure

Posted by Ron Pila

Image courtesy of jurvetson
In our earlier post (Just what is causing IT project failures?) we referred to a report prepared by the Victorian Ombudsman in conjunction with the Victorian Auditor-General (Own motion investigation into ICT-enabled projects released November 2011) (Ombudsman's Report) which looked into the factors that contributed to the failure of public sector ICT projects.

Those factors can be divided into:
  • things done at the outset that result in a project being set up for failure; and
  • 'mistakes' made during the course of the project.

In this post, we consider the factors that set up projects for failure. In a subsequent post, we'll look at the 'mistakes' made during projects that were identified in the Ombudsman's Report.

Many IT projects are perceived to be failures because they substantially exceed their budgets or run significantly late. Often, at least in part, this is due to the manner in which the budgets and timelines were set in the first place. Several of the key themes and factors identified in the Ombudsman's Report illustrate how this occurs. These include: 

Planning

The Ombudsman's Report commented on the fact that insufficient time and effort was devoted to planning of projects and this had a material impact on their success. This was particularly evident in the manner in which business cases were prepared and used.

Business cases are often prepared as a 'tick the box' exercise to obtain funding rather than a medium to allow management to make a fully informed decision. They should be seen as a project road map, not a sales pitch to secure funding.

In the projects reviewed in the Ombudsman's Report, many of the business cases that had been prepared were not sufficiently comprehensive. Also, they covered only the Agency's preferred way forward and did not present alternative options. 

Optimism bias

The Ombudsman's Report refers to 'optimism bias', which is a tendency of individuals to underestimate cost, time and complexity and overestimate benefits.

An example of optimism bias at work is the Myki project, which was scheduled to take two years to deliver despite the fact that no similar project in the world was conducted in under four to five years. It is no surprise that it is running significantly overdue. As project costs are linked to project duration, this would also account for part of the budget blow-out.

If we accept that optimism bias is human nature, it will not be easy to overcome. However, it is important to recognise this tendency and to curtail its impact by rigorous planning. 

Funding processes

With limited funds and lots of potentially competing projects, project sponsors can be tempted to 'gild the lily' when applying for funding. This could be in the form of understating the project costs or timelines or overstating the resulting benefits. This will set a level of expectation that the project is simply unable to meet, leading it to be labelled a failure. 

Public announcements

In the public sector, it is common for Ministers to make public announcements about projects, including how much they will cost and when they will be complete. This is often done on limited information and prior to obtaining commitments from vendors about the actual cost and time frames involved. These announcements take on a life of their own as project participants are reluctant to admit that what has been announced cannot be achieved. 

Procurement

The Ombudsman's Report comments on the potential for the use of innovative procurement techniques to get projects off on a better footing. In particular, reference is made to the interactive vendor engagement process which allows the customer and vendor(s) to have a detailed dialogue about the project scope before finalising their contractual commitments. This allows vendors to provide more accurate pricing with fewer assumptions, providing greater certainty to both parties.

Setting up projects on a more sound basis will go a long way towards ensuring expectations are met and that those projects are seen to be successful.

0 comments:

Post a Comment