{"id":12493,"date":"2026-11-04T16:43:20","date_gmt":"2026-11-04T16:43:20","guid":{"rendered":"https:\/\/myb2bnetwork.com\/blog\/?p=12493"},"modified":"2026-11-04T16:44:11","modified_gmt":"2026-11-04T16:44:11","slug":"why-outsourcing-projects-fail","status":"publish","type":"post","link":"https:\/\/myb2bnetwork.com\/blog\/why-outsourcing-projects-fail\/","title":{"rendered":"Why Outsourcing Projects Fail: 3 Case Files and the Data"},"content":{"rendered":"\n
\"Four<\/figure>\n\n\n\n

Why outsourcing projects fail is rarely a mystery once the inquiry reports are published. In 2007, Queensland signed a A$6.19 million contract to replace a health payroll system. By 2012, an audit put total spending on that payroll system through 2017 at A$1.25 billion. The contract looked small. The consequences were not.<\/p>\n\n\n\n

Every failed project has a version of that story: a signature, a deadline, a go-live date, and a bill nobody planned for. The three cases below come from public inquiries, regulators, and audit offices, so the facts can be checked. Together with one large dataset, they show the same four weak points again and again.<\/p>\n\n\n\n

A note on method.<\/strong> This is a desk review of documented evidence, not an interview study. I haven’t invented quotes or “anonymous interviews.” If you want the 15-interview version, the protocol at the end of this post shows how to run it so you can publish genuine primary data.<\/p>\n\n\n\n

What Does It Mean When an Outsourcing Project Fail?<\/h2>\n\n\n\n

An outsourcing project fails when the buyer doesn’t get the agreed outcome on the agreed timeline and budget, or gets it at a cost that wipes out the reason for outsourcing. It can fail loudly, as a public outage, or quietly, as slow delivery and rising invoices.<\/p>\n\n\n\n

Failure is also rarely one party’s fault. In each case below, the buyer and the supplier each point at the other, and the inquiries often find shared blame.<\/p>\n\n\n\n

Why Does Understanding Why Outsourcing Projects Fail Matter?<\/h3>\n\n\n\n

It matters because the pattern is predictable, and predictable problems can be priced into a contract. McKinsey and the University of Oxford studied more than 5,400 large IT projects<\/a> (budgets above $15 million) and found they ran 45% over budget and 7% over time on average, while delivering 56% less value than predicted. About 17% went badly enough to threaten the company’s existence, and each extra year of schedule added roughly 15% to cost overruns.<\/p>\n\n\n\n

Two cautions: that dataset dates from 2012, and it covers large IT projects in general, not only outsourced ones. Outsourcing adds a contract boundary on top of the same risks, which is why oversight and incentives matter so much.<\/p>\n\n\n\n

Case File 1: TSB Bank (2018) \u2014 The Supplier Nobody Verified<\/h3>\n\n\n\n

In April 2018, TSB moved its customers onto a new core banking platform built by Sabis, the IT arm of its parent, Sabadell. Up to 1.9 million customers lost online banking access, and some stayed locked out for weeks, according to the independent review by law firm Slaughter and May<\/a>.<\/p>\n\n\n\n

That review found the system was neither stable nor complete at go-live, and that only one of two data centres had been tested. An earlier IBM report, written after IBM was brought in to help, said it saw no rigorous go-live criteria proving production readiness.<\/p>\n\n\n\n

The bill was heavy. TSB reported \u00a3176.4 million in IT-related costs for the first half of 2018 alone, and regulators later fined it about \u00a348 million<\/a>. TSB’s board disputed parts of the review and said Sabis cut corners without telling it. The two sides still disagree, which is part of the lesson: if you can’t verify a supplier’s readiness yourself, you’re trusting a status report.<\/strong><\/p>\n\n\n\n

Case File 2: Queensland Health Payroll \u2014 The Procurement That Set the Trap<\/h3>\n\n\n\n

IBM won the A$6.19 million contract in December 2007, with delivery due in August 2008. The system went live in 2010 with defects, and health workers were paid wrongly or not at all.<\/p>\n\n\n\n

A 2013 Commission of Inquiry found IBM should never have been appointed prime contractor and described serious failings in governance, oversight, and contract management. The numbers need care, though. KPMG’s A$1.25 billion figure covers total payroll<\/a> spending from 2010 to 2017, including roughly A$1.01 billion of ongoing operations, plus A$220.5 million to fix problems. IBM says its own fees were A$25.7 million, under 2% of the total, rejected many of the inquiry’s findings, and stated that the inquiry found no breach of law or contract.<\/p>\n\n\n\n

The state later lifted its IBM ban. The takeaway is less about one vendor than about how the deal was set up: a bad selection process and weak contract controls can lock in failure before any code is written.<\/strong><\/p>\n\n\n\n

Case File 3: Healthcare.gov (2013\u201314) \u2014 Moving Targets and Loose Oversight<\/h3>\n\n\n\n

A national health-insurance website launched in October 2013 and stalled within days. Auditors at the Government Accountability Office<\/a> later found that the agency had proceeded before requirements were stable and chose a cost-reimbursement contract type.<\/p>\n\n\n\n

Between September 2011 and February 2014, the two primary contracts grew by about 245%, from $86 million to $294 million, and total spending across 62 contracts reached $840 million. GAO identified about 40 instances where program staff authorized more than $30 million of contractor spending without following oversight procedures. The agency also withheld just $267,000 of a $12.5 million fee paid to the first contractor. That contractor disputed the agency’s claim that it hadn’t met requirements.<\/p>\n\n\n\n

The pattern: when scope keeps shifting and nobody enforces change control, the contract price stops meaning anything.<\/strong><\/p>\n\n\n\n

The Four Fault Lines: What Do the Cases Have in Common?<\/h4>\n\n\n\n

The Four Fault Lines model is an original framework for this article. Each case broke along one or more of these lines, and each can be tested before you sign.<\/p>\n\n\n\n

Fault line<\/th>TSB<\/th>Queensland Health<\/th>Healthcare.gov<\/th><\/tr><\/thead>
1. Scope fixed before signing<\/td>Ambitious timetable set for the switchover<\/td>Tight 2008 delivery date<\/td>Requirements unstable at award<\/td><\/tr>
2. Oversight based on evidence<\/td>Board relied on supplier assurances<\/td>Weak governance and contract management<\/td>Spending approved outside procedure<\/td><\/tr>
3. Proof of readiness at go-live<\/td>One data centre tested<\/td>Live with defects in 2010<\/td>Launched without verified performance<\/td><\/tr>
4. Incentives and exit leverage<\/td>Supplier inside the same group<\/td>State later unable to recover damages<\/td>Little fee withheld despite problems<\/td><\/tr><\/tbody><\/table><\/figure>\n\n\n\n

How Do You Use the Four Fault Lines Before You Sign?<\/h4>\n\n\n\n

Turn each line into a question your team must answer in writing:<\/p>\n\n\n\n

    \n
  1. Scope:<\/strong> Are requirements frozen for the first phase, with a change-order process and price for every change?<\/li>\n\n\n\n
  2. Oversight:<\/strong> Who on your side checks evidence, such as test results and access logs, instead of reading status reports?<\/li>\n\n\n\n
  3. Readiness:<\/strong> What go-live criteria must be met, and who can say no to launch?<\/li>\n\n\n\n
  4. Incentives:<\/strong> What portion of fees is held back until acceptance, and how do you exit if delivery stalls?<\/li>\n<\/ol>\n\n\n\n

    Our guide to the risks of outsourcing<\/a> covers the wider risk register, and our post on protecting IP when outsourcing<\/a> covers the clauses for fault line four.<\/p>\n\n\n\n

    Run the 15-Interview Study Yourself<\/h4>\n\n\n\n

    If you want genuine primary research on why outsourcing projects fail, here’s a protocol that produces publishable results:<\/p>\n\n\n\n