History’s most expensive software fails

Projects that run over-budget are the bane of every industry. Governments have been toppled and companies have crumbled in the wake of poor project management.

Software is meant to make people’s lives easier. Governments and businesses frequently invest large sums into creating new technologies to improve the lives of their workers or to save money. But all too often development costs creep higher until they’re out of control. And when something that’s supposed to make things run smoother actually gums up the works it can be especially frustrating.

The things that send costs soaring beyond expectations are unique to every situation, but issues like cost overrun, scope creep, and poor time management are often to blame. The party responsible for sending projects over-budget often change from one project to the next. Sometimes it’s the contractor who fails to deliver on a promise, sometimes it’s the customer who doesn’t properly describe what they want, and sometimes there are outside forces that forms a perfect storm that sinks the project.

Let’s take a look at some of the most spectacularly botched, impressively mismanaged and expensive software boondoggles in history.

Scope Creep

“Scope creep” refers to the continuing growth or frequent changes to the objectives of a project. This usually happens when the scope of a project isn’t properly defined from the outset. It’s a potential problem for most projects, but is especially risky for larger endeavors.

Australia’s Standardized Business Reporting

In 2006 the Australian government set out to streamline government reporting requirements for businesses. A report released that year found that regulations put an unfair burden on local businesses.

Finally, in 2010 the government set out to reduce the costs of reporting for businesses by AUD$800 million over six years. In the time since, the entire project has cost between $800m-$1bn. Still, only 2%-3% of Australian businesses have taken up electronic reporting.

At best that’s spending $800m to save $24m. That’s a -3233% ROI. Wow.

Most criticism of the project is pointed at its growing scope and poorly defined objectives. Despite that, the project has not been cancelled.


Stock trading grew exponentially in the 80s, putting pressure on the London Stock Exchange and its reliance on paper communications.

TAURUS (Transfer and Automated Registration of Uncertified Stock) was meant to automate transactions on the LSE. The plan was that the program would reduce costs and time drain while increasing convenience.

Unfortunately, the scope of the project was never properly defined and costs grew out of hand. A number of regulatory hurdles, lack of oversight, and underestimating the complexity of the problem led to its abandonment in 1993. A full decade after it was started.

Total costs for the failed project ended up being in the £75 million range.

Health Problems

Managing the healthcare platform for an entire country is no easy task. Governments who have had decades of experience still struggle with the burden of providing a quick and easy service.

It’s no wonder then that some of the biggest money drains have come from the healthcare sector.


Healthcare.gov is an example of a software project that didn't go according to plan.

Perhaps one of the most famous examples of project overruns ever is the HealthCare.gov website. The health insurance site is operated by the U.S. federal government and is a cornerstone of the Affordable Care Act. It acts as an exchange for people buying private health insurance while offering subsidies to people who fall within a certain income bracket.

The software was rolled-out on October 1, 2013 as planned, but was impaired by serious technological glitches. By some estimates only 1 per cent of people managed to successfully sign up in the first week of operation and President Obama openly admitted to the site’s disappointing performance.

Cost overrun was one of the biggest issues with the project. The design of the site was contracted out, with Canada’s CGI Group the biggest player. CGI’s original budget was US$93.7 million, but that ballooned to $292 million ahead of the launch. Overall costs for the build reached $500 million ahead of the roll-out, and eventually cost American taxpayers $1.7 billion.

A 2014 non-partisan report found that the government did not provide “effective planning or oversight practices” in developing the site.

Cover Oregon

HealthCare.gov was set up for the 36 U.S. states that chose not to run their own sites. Other states chose to set up their own exchanges. But there was one case where the development of one of those exchanges ended in accusations of racketeering and an FBI investigation.

Cover Oregon was setup to allow Oregonians and small businesses to purchase insurance at federally subsidized rates, but since its inception the site has been plagued with technical problems.

The site was developed by the Oracle Corporation and managed by the State of Oregon. It was supposed to be operational October 1, 2013, but as of mid-December could still not process applications. With the January 1, 2014 deadline fast approaching, Governor John Kitzhaber advised residents to mail in a paper application.

The board decided to scrap the program in April of 2014 and opted to use the Federal website.

In the wake of the fiasco, Oracle Corporation sued Cover Oregon, claiming their software was used without payment. In the same month the state of Oregon sued Oracle for breach of contract and “racketeering.”

In March, 2014 the FBI launched an investigation into possible criminal activity involved in the development of the site. A grand jury issued subpoenas in May of the same year. In 2015 a judge threw Oracle’s case out of court in spectacular fashion.

NHS National Programme for IT (NPfIT)

Problems with healthcare software are by no means a phenomenon exclusive to the United States. Not to be outdone by their Atlantic rivals the British attempted to build a program in 2005 that would centralize the medical records of patients while connecting them to general practitioners and hospitals.

Data security was an obvious concern for patients and doctors alike with skeptics charging that the government wasn’t demonstrating value for money with the project.

The project began hemorrhaging money and by the time the program was shut down in 2011 it was nearly £10 billion over-budget. Members of Parliament aptly described the endeavor as one of the “worst and most expensive contracting fiascos” ever.

A Matter Of Timing

Delays and time issues can cause projects to derail.

When it comes to software development, time is nobody’s friend. Without careful planning problems can crop up at any stage in a project’s development, sending it over-time and over-budget.

Expeditionary Combat Support System (ECSS)

Military cost overruns aren’t uncommon. But the U.S. Air Force’s attempt to streamline its operations between 2005 and 2012 may be one of the worst examples.

In fact, it was characterized by two high-ranking officials as “one of the most egregious examples of mismanagement in recent memory.”

The goal of the Expeditionary Combat Support System was to consolidate and replace more than 200 legacy systems. But after spending US$1.1 billion on its development, the USAF found there was no significant improvement to military capability and it would cost an additional $1.1B to reach even a quarter of the original scope of the project. They cancelled it in 2012.

Development of the system was originally contracted out to our old friends at Oracle Corporation in 2005, but was later supervised by Computer Sciences Corporation.

Digital Media Initiative

In 2008 the British Broadcasting Corporation set out to modernize its operations by moving to a fully digital, tapeless production workflow. The project was expected to cost £81.7 million and save the BBC around £18 million in production costs.

The development process dragged on for five years and cost the BBC £98 million between 2010 and 2012 alone.

The project was initially outsourced to Siemens with consulting from Deloitte, but was besieged by technical problems and delays from the outset. In 2009 the contract with Siemens was cancelled at a cost of £10.7 million. The project was then brought in-house and rebranded as Fabric.

Ultimately, the BBC chose to abandon DMI. A number of higher-up employees involved with the project lost their jobs. The BBC Director General called it a huge waste of money, while one BBC Trust member called it a “complete catastrophe.”

Completely Avoidable

By properly scoping, keeping tabs on progress and breaking the project down into manageable pieces you can avoid having your projects go over time or over budget. It’s no coincidence that all the projects listed above are government connected in some way.

Every project at Gigster has a dedicated Product Manager. Don’t let your project go over budget or over time. Hire Gigster today.

Tyler Trumbull

Ty splits his time between Canada and Mexico. He’s been writing for Gigster since early 2016 where he really enjoys learning and sharing clients’ stories. He plays banjo in one of Mexico’s only country bands, wishes he could write like Thomas Pynchon, and is generally a fan of the Oxford comma.