Will they never learn? Poor project management derails Transport for NSW project
- It seems astonishing that in the 21st century and with thousands of project details available in the public domain that poor project management continues to dog public sector projects. Will they ever learn?
There would be few executives, government agency heads or politicians who wouldn't jump at the promise of saving over US$70 million in technology costs. Unfortunately the Australian state of New South Wales has discovered that achieving those benefits, requires rigorous project management.
Shortly after being elected in 2011, the Liberal government of Australia's richest, oldest and most populous state decided to amalgamate the state's transport agencies into a mega-agency of 25,000 staff called Transport for NSW, loosely based on the UK capital's Transport for London model.
One of the main arguments for amalgamations are economies of scale and a 2012 government audit shortly after TfNSW's founding found the mega agency had inherited 130 corporate systems which the accountants suggested could be rationalised for AUS$100 million, leading to US$73 million per year saving.
The Sydney based bureaucrats and ministers didn't delay, quickly deciding to standardise the super agency's systems on SAP and enlisting a range of international integrators to deliver the project. Japanese integrator NTT Data Business Solutions Australia won the first major contract with $1.13 million tender in 2013 to manage the project's progress reporting.
A few months later global consulting firm Deloitte won the project's biggest contract with a $70 million deal to integrate the ERP system by the end of 2016. Tbe following year Global US based data migration solution provider BackOffice Associates won a $12.8 million contract over three years to migrate the existing systems' data into the new platform.
By the time BackOffice Associates won their contract in early 2014, the project was already in trouble. As Australian industry site IT News reported at the time, the state's Auditor-General warned scope changes and project management personnel problems within the agencies threatened the project at an early stage.
Management have responded to some of the early challenges of the project, including project team changes and scope changes, with re-alignment of project plans and delivery timelines
observed the Auditor-General's 2014 report.
Two years late the Auditor-General's misgivings proved well founded with the state Parliament's budget estimates committee being told in 2016 how key contracts had more than doubled with Deloitte integrations contract had now standing at $148 million and the BackOffice Associate's data migration at $29.8 million. Unsurprisingly the completion dates had slipped well beyond their original completion dates.
Even more worrying for the NSW taxpayer was another Auditor-General's report into the department at the end of 2016 that raised more concerns about the project.
Remarkably, one of the largest agencies within Transport for NSW – the State Transit Authority which operates Sydney's sprawling government bus services – had been removed from the contract in order, as the Auditor-General reported “to manage business risk and the program budget.” Instead the organisation would instead update its existing Ellipse system.
Even without the buses included, the Auditor-General's office was still worried about the bureacrat's ability to deliver the project and was scathing of Transport for NSW's disregard of basic project management principles.
Good project management to deliver a quality system depends on balancing three key constraints: time, scope and cost. Increasing one constraint without affecting another, changes the quality of the system. The current implementation status shows the SAP ERP program has varied the scope, extended the timelines and limited the program budget.
111 change requests were approved since the program’s inception in January 2013 to October 2016, costing $82.0 million. SAP ERP program was allocated $58 million or 71 per cent of the cost. The remaining $24.0 million was borne by the agencies budget.
In 2015, 31 change requests were approved costing agencies $11.0 million and ERP program $27.0 million. 18 per cent of this cost was on two change requests.
A further 19 change requests worth $12.7 million were approved in 2016, with one change costing the agencies over $5.0 million.
SAP ERP program management advised that further changes to the system will not be accepted. Limiting the project scope when legitimate changes are needed could impair quality and result in a system that does not fully meet agencies’ needs.
While the Auditor-General's office expressed its misgivings about the contract, the minister and his heads of department put up a brave face when questioned about the project during last year's Parliamentary committee hearings, blaming most of the delays and budget blow outs on adding Enterprise Asset Management features to the project and other tweaks.
It still remains to be seen if Transport for New South Wales can deliver the project over the next year and whether it will achieve the touted 100 million Australian dollar a year savings. At the moment though it appears the Transport for New South Wales SAP project is going to join the ranks of embarrassing IT failures that Aussie governments are becoming notable for.
Software vendors, their resellers and integrators have, over the years, been rightly criticised for overpromising and under-delivering. This seems especially prevalent when pitching to the executives of government agencies keen to show financial discipline to their political masters.
In this case, the delays and overruns in Transport for New South Wales' migration to an SAP ERP platform are the more fundamental problem of an organisation not having the skills to scope or manage what was always going to be a complex project.
At the very least it seems astonishing that a long history of public sector IT implementation failure around the world should have informed Transport for New South Wales of the potential risks.
Any implementation that involves multiple third parties runs the risk of over-runs for the simple reason that the entities involved frequently find themselves butting up against one another, with the all too familiar finger pointing that goes with it. Unless there is strong top down management of the SIs, then guess what - taxpayers end up with the bill.
On its face, the level of due diligence undertaken prior to embarking on the project seems - how shall we say - inadequate. If TfNSW had undertaken sufficient checks, it would have quickly discovered that at least one of the SIs involved in this scheme is frequently cited as a party in a public sector project failure/over-run. It would also have discovered that eye watering though the costs may appear, change order requests on SAP projects are super expensive. Finally, the number of change orders is mind boggling for a project of this size. Who scoped it? Who agreed the scope? Again, it is all too common to find that SIs low ball projects to win the deal but more or less from day one are looking for change orders upon which they make their real profit.
Once again, the Aussie experience is a reminder that starting a project or choosing a product simply on the promise of reducing costs usually ends poorly.Full scope and execution project management skills are essential for success.
Around the world IT and project management skills shortages are becoming apparent. Understanding the complexity, costs and risks of a project are essential to delivering the promised benefits. Sadly, the TfNSW failure among public servants to understand is something for which New South Wales taxpayers are now counting the cost.
Finally, for those interested in this broad topic, it is worth reading some of the content produced by Eric Kimberling at Panorama. He paints an all too familiar yet often depressing picture of ongoing problems in ERP projects and especially around project management topics.