Lawsuits Show It Takes Two Parties to Spark IT Project Dumpster Fire

In the lawsuit, National Grid had originally sought [pdf] to “recover from Wipro the approximately $140 million in fees it paid to Wipro (consisting of approximately $71 million paid to Wipro during the Project and another approximately $69 million paid to Wipro after go live), and hundreds of millions of dollars in fees it paid to other vendors for post-go-live stabilization and remediation work.” National Grid ended up spending two years and some $600 million in addition to its original $393 million USFP development costs to get the system stabilized and working acceptably.

National Grid’s U.S. Foundation Program was intended to implement an integrated human resource management, financial, billing, accounting, payroll and operations support IT system, taking over the functions of a multiplicity of legacy systems then in place. National Grid claimed in its lawsuit that Wipro had, among other things: misrepresented its skills and expertise to implement the SAP-based system; created a system design that was fundamentally flawed and overly complex; failed to adequately test the system, and; failed to convert legacy data needed in the new system properly.

Once the system went live on 5 November 2012, it quickly became apparent with the first payroll run that the system was deeply flawed. National Grid workers either received incorrect pay or, worse, no pay at all for weeks. The situation was so bad that the Attorney Generals of both Massachusetts and New York eventually fined National Grid for the pay problems.

National Grid was forced to hire 450 contractors to help resolve the system generated pay issues, 200 more to unravel the supply chain issues the system created, and an additional 200 to support corporate financial activities like closing the company’s books. While it normally takes 4 working days to close the company’s monthly books, the first financial close took 43 working days, National Grid stated.

While it is apparent that Wipro failed to create the SAP-based USFP system as promised, the National Grid itself is not blameless, either, as a scathing 265-page audit [pdf] released in July 2014 makes abundantly clear. For example, despite “problems with the system and company readiness” prior to the go-live date of 5 November 2012, National Grid’s USFP management team decided to launch the system anyway in order to avoid the costs of postponing it again (the go-live date having been delayed three times already).

The system also went live mere days after Superstorm Sandy hit National Grid’s operating region, which taxed many critical ERP system functions to the max as the utility’s crews and contractors were sent out to repair the damage Sandy wrought. Adding to the trouble, many of the National Grid employees who were supposed to use the new system missed their training due to the storm.

The mismanagement of the go-live decision was just one in a long list of mistakes, the audit report explained. These gaffes included: selecting Wipro in the first place, as a basic check of its claimed expertise in U.S. utility markets would have shown it didn’t have any; the development team failing to communicate with the users of the system to find out if it worked or if they had the information they needed to do their jobs; National Grid’s management and development team underestimating the complexity and risks involved in the ERP system development; the existence of a “good news” management culture within National Grid that minimized the risks that existed; the use of an ineffective testing regime that missed obvious errors, and; the lack of effective governance and oversight of the project at all levels of the organization, stretching from the development team to the Board of Directors.

While it is debatable whether Wipro could have successfully contested National Grid’s lawsuit against it, if the case had gone to court there is no doubt that National Grid’s managerial competence or lack thereof would have taken another beating.

Another long-standing lawsuit over a failed ERP system implementation was also settled recently, this time between Bridgestone and IBM. Readers of the Risk Factor may remember that back in 2013, Bridgestone filed a $600 million lawsuit [pdf] against IBM for allegedly delivering a defect-laden $78 million SAP-based invoicing, accounting, and product delivery system that went live in January 2012. Bridgestone claimed it ultimately incurred more than $200 million in damages and hundreds of millions more in reputational losses.

At the time, IBM quickly and vigorously fired back, stating that the problems with the system originated with Bridgestone’s decision to roll out the system before it was ready, which it warned the company about. Among other accusations, IBM also claimed that Bridgestone’s project management team wasn’t highly skilled or knowledgeable, and that there was poor project governance given that the company had replaced its CIO six times over the project’s development. In addition, Bridgestone had under-resourced the project, IBM said.

The lawsuit has slowly wound its way through the court system, including a period when a judge ordered both sides to work with a mediator to find a way to settle the dispute out of court. Basically, the judge was sending a not-too-subtle hint that it looked like both sides might be found at fault for the failure.

After mediation proved unsuccessful, the judge set a trial date for July of this year. However, for reasons not yet disclosed, both sides decided to end the case and pay for their own legal fees. Apparently IBM did not pay anything to Bridgestone in the way of a settlement, at least according to its latest Security and Exchange Commission filing.

Examining the claims and counter-claims in the various audit reports and court documents, it is clear that no one covered themselves in glory. Both National Grid’s and Bridgestone’s governance of their respective IT projects was woefully deficient, not to mention overly-optimistic. They both went live with their systems despite warnings not to do so, which is a common project problem [pdf], unfortunately. They also both believed their vendors’ representations about their capabilities to deliver that could have easily been shown with a bit of due diligence to be mere puffery [pdf] if they had bothered to look.

Furthermore, both Wipro and IBM were more than happy to take their clients’ money, even though they apparently believed their clients didn’t know what they were doing. The risk exposure this caused to both their clients and themselves, was seen as acceptable. Moreover, both Wipro’s and IBM’s system development practices were inadequate, especially in regard to system testing and project risk management.

All things considered, each party probably got what it deserved. National Grid and Bridgestone received initially incomplete and defective IT systems that cost them lots of time and money to fix, while Wipro and IBM have faced significant knocks to their reputations for not being able to deliver what they promised.

Source: IEEE Spectrum Computing