New research: IT cost overruns, delays and contract terminations

A report from the European Services Strategy Unit, an independent watchdog agency, summarizes the scope of major public sector IT failures over the last decade. The report is titled, Cost Overruns, delays and terminations in 105 outsourced public sector ICT contracts.

Key findings presented in the report include:

• 105 outsourced public sector ICT projects with significant cost overruns, delays and terminations.
• Total value of contracts is £29.5 billion.
• Cost overruns totaled £9.0 billion.
• 57% of contracts experienced cost overruns.
• The average percentage cost overrun is 30.5%.
• 33% of contracts suffered major delays.
• 30% of contracts were terminated.
• 12.5% of Strategic Service Delivery Partnerships have failed.

The primary contractors involved with these projects were: EDS, Liberata, Fujitsu, IBM, Accenture, Atos Origin, Capita, ITNET (now Serco), Siemens, and BT.

The report offers the following conclusions about the failures it covered:

• Many projects are over-ambitious, complex and difficult whilst others are more straightforward.
• The private sector frequently believes its own hype and PR about ‘world class’ services and thus often overstates its ability to deliver.
• Too many public sector managers rush to use management consultants who predictably encourage outsourcing, yet their evidence and conclusions are rarely challenged or subjected to rigorous assessment.
• Clients or ‘commissioners’ are often under-resourced and/or do not have the required skills and ability to manage large projects through the procurement process.
• The procurement process is a high-risk strategy, heavily influenced by market forces in respect to who bids, the level of competition and private sector strategies to increase market share. Inadequate bids or loss leaders may result in tight margins and cost reduction strategies.

This research is important because it reflects analysis of historical data, as opposed to most studies of IT failure, which rely on opinion surveys. It's interesting to note the results correspond more closely to the Standish Group's Chaos Report than to other recent research. The Standish report suggests that approximately 60% of software development projects are challenged, which closely corresponds to the 57% cost overrun rate presented by the current research. In contrast, other recent research (here and here) has indicated failure rates closer to 30%.

Another interesting fact is that only 33% of projects studied suffered major delays. Since 57% of projects were over-budget, it appears that many projects compressed unanticipated work loads into the scheduled time frame. When unanticipated work is performed during a fixed schedule period, the risk of inefficiency and poor management rises.

This point is supported by the report:

Some additional costs are connected with widening the scope of contracts once they are operational although it is almost impossible to determine whether this is for positive reasons connected to service delivery, whether it is connected to inadequate specifications or whether it is linked to private contractor proposals motivated by additional income, or a combination of all of these factors.

Along the same lines, it's particularly worth noting that 30% of contracts were terminated, which I find to be one of the most significant numbers in the report. Cancellations often represent the worst kind of failure, where the project has gone so wrong it's not worth salvaging. Even if this number is slightly inflated, it's still an extraordinary level of catastrophic failure.

To learn more, I spoke with Dexter Whitfield, Director of the European Services Strategy Unit, which produced the report. He expressed frustration over promises made by IT contractors and government agencies:

What is said in order to justify those projects is a million miles from reality. Projects are presented as being much easier than they actually are. There's a massive gap in credibility in terms of what is said at the beginning in projects and what happens at the end. It's not quite as simple and easy as it's always made out to be.

For the reasons described above, this research represents an important summary of historical data related to government IT project failures in Europe, and is worthy of careful review. The report also includes a detailed bibliography, which will be of great value to researchers.