Computer Weekly's 2003 Programme Management Survey highlighted the fact that average overruns on budget and schedule are down to 18% and 23% respectively.
This is welcome news, but the survey also found that only 16% of projects hit all their targets. One common problem is that many IT projects fail to win the support of end-users before new software or hardware is rolled out.
Experience in the IT market provides plenty of examples of this kind of problem.
Take the life insurance company that wanted to increase sales. It initiated a project to provide its salespeople with laptops. It provided bespoke software to gather and analyse information to select the right product and generate policy documents, all in the clients' homes.
The salespeople liked the technology and the substantial time saved through this. However, management noticed no significant increase in sales and discovered that many salespeople worked to a specific earnings level, and once it was achieved, did not pursue extra sales.
This is a classic example of a single-minded focus on the IT goal - introduce new laptops and tailored software - rather than the business benefits of a project - improving productivity.
Technology enables business improvement, it is not the end-goal. The system delivered by IT at the life insurer worked perfectly, but the project failed to increase sales volumes, which was the true goal.
Happily, all was not lost; with minor changes, the system became an effective mechanism for enforcing UK insurance regulations, thus reducing the risk of fines for non-compliance by the salespeople.
Consider also the manufacturer forcing SAP on its finance function. Backed by the board and IT director, a crack team of implementers was flown in, fresh from its last assignment on the continent.
The team started to implement the same changes to business processes to fit with the new systems - it was left to the finance director to spot the resentment among a group of staff testing the new system.
The director commissioned an "operational readiness" review across his department, seconding some of his staff to work alongside the consultants. His team was happy because members had contributed to shaping their working practices and other members of staff were also pleased that they had been consulted.
In these examples, the IT director or chief information officer should not have been left to sell a solution to the business.
Well-constructed training sessions can also secure support from staff for a new system. One model is the use of "lunch and learn" sessions, where staff have the opportunity to examine the issues raised by the new system objectively - like a case study - accompanied by a buffet lunch. Roadshows also have their part to play, but they can be poorly attended unless there is a compelling reason or incentive to do so.
A more effective approach is to secure senior business sponsorship from the beginning. This sets the agenda and provides the necessary guidance to IT and other technical specialists, such as the process design team. Underpinning this is the identification of all stakeholders at the earliest stages of any project and involving them throughout the lifecycle.
Even if the project cannot address all of their concerns and suggestions, the fact they have been communicated with and given a voice will, in most situations, engage them and facilitate their buy-in to the business decisions made.
Arthur Murphy is a director in PricewaterhouseCooper's programme and change management practice