Breaking Victoria’s IT fail cycle: What not to do


analysis The date of November 23, 2011, will go down in history as one of the blackest for the thousands of technology workers belonging to the many departments and agencies making up the Victorian State Government.

The State’s public sector workforce had already suffered many ignominies over the past few years. The cancellation of Victoria Police’s LINK crime database, following the departure amid allegations of contractual improprieties of its former chief information officer. The incredible series of delays, budget over-runs and shadowy claims associated with the gargantuan myki public transport smartcard overhaul, which were revealed in a series of damaging leaks.

Of course, the HealthSMART e-health overhaul, which had been quietly going off its rails for years, was also in the picture. And central IT shared services agency CenITex … let’s not even go there.

The state’s technology workforce was already aware of all of these problems — and many more — when dawn broke on Wednesday the 23rd of November 2011. But by the end of it they would be in shock all over again. That day last week marked the publication of a landmark report into ten of the state’s largest IT projects rolled out over the past half-decade and beyond. And what it found was remarkable. Total cost over-runs: $1.44 billion. Almost every project had been delayed, in some cases by up to half a decade. And in several cases, major, important IT projects had simply failed to deliver and been scrapped.

In the wake of the publication of this report, which analyst firm Ovum has branded the signal of “an emerging crisis of confidence in government ICT” in Australia, Delimiter undertook over the past week to put some thoughts as to what should take place next.

When almost every major IT project has broken its budget and its timeframe, and many have completely failed, after soaking up hundreds of millions of dollars of public money that could have been spent on health, education, cutting down crime or public transport, what happens now? Where does the Victorian State Governments and its technology workforce turn to? We’ll investigate the issue in a series of articles this week.

Don’t do it again
When we asked Australia’s top enterprise IT analysts what Victoria should do next, several of them made comments which reminded us of one of the fundamental precepts taught to all medical students as a first step when approaching any emergency medical situation: First, do no harm.

Peter Carr, the managing director of analyst house Longhaus (which has had a ringside view of similar state government IT disasters in its home state of Queensland over the past 18 months), was emphatic that the Victorian Government must not launch straight into the kind of comprehensive review of its technology operations that made Sir Peter Gershon famous (or infamous) within the Federal Government in 2008 and beyond.

“The last thing they need is a Gershon-style review,” Carr said. “They always end badly, to be honest.”

The reason for this, the analyst said, was that conducting such a review would inevitably lead the Victorian Government into repeating the cycle of technology underinvestment which led it to get into its current mess in the first place.

Carr said Longhaus had conducted research that tracked the performance of Federal Government technology operations over more than a decade — right from the establishment of the controversial Office of Asset Sales and Information Technology Outsourcing (OASITO) more than ten years ago under the Howard Government, right through to Gershon in 2008 and the fallout from the British efficiency expert’s recommendations over the recent period.

Over that period, Carr said, the Federal Government had gone through a clear cycle. The colossal IT outsourcing initiatives driven by OASITO — which saw major organisations such as the Australian Taxation Office sign multi-million-dollar outsourcing deals with companies like EDS — had been aimed at cutting billions out of public sector costs, but actually led to chronic levels of underinvestment.

Carr didn’t mention these examples, but it’s possible that it was such underinvestment which caused agencies such as the Department of Immigration and Citizenship, the Customs and Border Protection Service, the Australian Taxation Office and even Defence to be forced over the period since OASITO to kickstart IT systems overhaul projects worth billions — several of which went drastically off the rails and caused real-world impacts on industry and individual Australians, especially at Customs and the ATO.

The level of funds required always surprised the politicians involved, said Carr. “All of a sudden we’ve got a billion dollar spike in IT costs,” he said. “Forget the fact that we’ve underinvested for ten years. Suddenly some politician says: ‘They’ve screwed us again. Let’s do a Gershon review.'”

The issue, Carr said, was that the savings made from initiatives recommended by Gershon-style reviews into effiency and effectiveness weren’t reinvested in technology projects to keep systems up to date. The current Labor Federal Government had initially planned to set up a billion-dollar fund for IT reinvestment with savings stemming from the Gershon Review. But it famously canned the approach as part of the 2010 election campaign, re-allocating the funds to other budget areas and immediately justifying, Carr said, Longhaus’ theory on IT underinvestment. The analyst doesn’t want the same thing to happen in Victoria.

A once-off colossal review wouldn’t solve Victoria’s problems, Carr said. “It all starts with a continuing and mature Government investment approach, which we can’t seem to achieve.”

Don’t call in Superman
The second solution which Longhaus recommended Victoria avoid is the so-called ‘Super-CIO’ approach which has served the state poorly in the past.

In 2003, Victoria poached high-ranking Defence IT executive Patrick Hannan to become its first whole of government chief information officer. This ‘Super-CIO’ role was set up with the explicit intention of reining in the State Government’s troubled IT operations and establishing a more commercial approach to buying, implementing and maintaining IT solutions.

Hannan declared himself willing to git the task “a bloody good go” and made some headway on issues such as broadband for schools, but quickly became bogged down in interdepartmental power struggles and issues with staff misconduct in the ranks. He famously never returned to his post after a Christmas break, taking the CIO of the Department of Education and Training, Ian Paton, with him.

After that point, the state had several other high-profile executives in its central CIO group, notably another ‘Super-CIO’, Jane Treadwell, who was plucked from Centrelink’s well-regarded IT management team but suffered a similar fate as Hannan. Another high-flier was deputy state CIO Steve Hodgkinson, who landed as a senior government analyst at Ovum. We’ll hear from him on Victoria’s IT woes later this week.

However, Carr points out that in all of these cases, the ‘Super-CIOs’ concerned failed to make much of a dent in Victoria’s problems — despite the fact that high-fliers such as Treadwell and Hannan tried different approaches to the same problem — Hannan being more technically minded and Treadwell possessing more traditional management skills.

“They’ve kind of tried the high-profile CIO thing and the central governance thing and it hasn’t worked,” he says. “What are we achieving, trying that again?” The problem, according to Carr, is that the problems in the Victorian Government when it comes to IT are “systemic”. Extending his logic, a Super-CIO wouldn’t just be wrestling with the projects themselves. They’d be struggling in a quagmire of embedded practices reinforced by multiple layers of bureaucracy. And that’s not a recipe for change.

Don’t focus on shared services
Another recipe for change which has been popular amongst Australian state governments over the past decade has been the notion that government IT service delivery could be consolidated into shared services centres — one or more per state government, to serve a number of agencies. Western Australia, Queensland, the ACT and New South Wales have followed this strategy over the past decade, as well as, to a perhaps lesser extent, the other states and territories.

Victoria’s already gone a long way down this path with the 2006 creation of several IT shared services organisations by the Department of Treasury and Finance. These groups merged in 2008 and were re-badged as CenITex, an organisation which provides basic IT services to mega-departments like Health, Justice and Human Services, as well as a number of smaller agencies.

Others in scope for transitioning to CenITex are Victoria Police, VicRoads and the Department of Education and Early Childhood Development (DEECD).

However, all is not well when it comes to IT shared services strategies in Australian state governments. Over the past year, the Western Australia and Queensland governments, which had been most advanced in their IT shared services implementations, have both broadly abandoned the strategies following a series of widespread which resulted in their own massive cost over-runs, extensive delays and abject failure to deliver. Gartner distinguished analyst Andrea di Maio — a foremost global expect on government IT service delivery — recently published a blog post in Australia noting shared services as a concept was under attack globally.

“There is no doubt that duplication of services and spending across different government agencies makes little sense,” wrote di Maio. “But experience shows that there is insufficient attention to the governance aspects (agencies want to have a say in how the shared services are structured) and – more recently – to technology evolution that is making some of the technology that is being targeted through shared services more and more commoditised.”

“The challenge is no longer to put one organisation in charge of delivering shared services, but to look at how to support agencies in efficiently buying the same service from external service providers.”

Longhaus’ Carr agrees.

The analyst points out that with an rough estimate of 500,000 to 600,000 IT professionals in Australia, and something like 130,000 along being contractors, it is unlikely that enough skilled labor exists for state governments to competently run their own on-shore IT service delivery centres at cost levels which they can afford to pay.

In Victoria, CenITex has been criticised for having too high a percentage of expensive contractors in its resources base, and has over the past several years been attempting to convert some of those IT professionals into permanent staff. However, Carr points out that when Federal Government agencies tried the same thing following the publication of the Gershon Review — which slammed the Commonwealth’s high reliance on contractors, a nasty little cycle occurred.

Federal Government CIOs were forced to let contractors go, the analyst says, only to have to re-hire them when important projects came up. “There’s nobody else to do the bloody job,” he says.
And the pain is already being felt at CenITex. Inter-departmental skirmishes are currently being fought between the shared services agency and some of its clients, who just want to get the job done and have been frustrated in some cases over the past year about the agency’s inability to do it. One can only imagine what the reaction was from CenITex’s government clients in mid-October when an outage at the shared services agency left thousands of government staffers without email and other IT systems for up to a week.

Whispers are rife in the Victorian Government that some agencies are bulking up their own internal IT teams to get work done on the sly that the official CenITex resources haven’t been able to complete. And CenITex isn’t even involved in IT projects, mostly — it tends to focus on standardised service delivery such as desktop PC support.

Ultimately, the Victorian Government’s IT problems are not going to be easy to handle … in fact, that’s probably the understatement of the decade. And both a long-term view as well as short-term tactical decisions will need to be taken.

However, what we’ve tried to outline today is where the obvious and immediate pitfalls lie for the state. If analysts and the past experiences of other states and Victoria itself are to be believed, the state will not find its way out of its current mess by conducting a huge, Gershon-style review, or by relying on Super-CIOs or shared services agencies. Those methods have been tried; and broadly they have demonstratedly failed repeatedly in other jurisdictions.

“Shame on the politician who endorses a Gershon Review for Victorian Government,” said Longhaus’ Peter Carr this week. “Shame on the politician who tries to bring in some external CIO superstar. A super-CIO isn’t going to fix it, shared services isn’t going to fix it. Hiring more IT public servants isn’t going to fix it.”

We’ve heard a fair bit from Longhaus on how not to fix Victoria’s IT problems. So what will fix Victoria? Next up, we’ll hear from Ovum analyst and former Victorian deputy CIO Steve Hodgkinson as well as from veteran analyst firm Gartner on what will work and has worked in similar situations. We’re also planning to look at who is actually responsible for ICT in the Victorian Government; putting some faces and names to the task ahead.


  1. So what is news? Is there anything unusual about this article? Surely the most ardent labor supporter and the most ardent liberal supporter can see that this is simply another mishap. It cost a few millions – but so what! Government always moves from one crisis to the next and always spend a few dollars along the way. The people don’t mind paying taxes to keep the likes of John & Ted (in the 1%) in their little hobbies! If they did have some objection they would do something about it. But they won’t. Will they!

    Anything for a peaceful life!!!!!!!!!!!!!!

  2. It’s not really that hard to work out why these big Govt projects almost always go off the rails.
    Typically a big multinational is picked as the outsourced provider (despite local companies who could do it better, the old maxim “nobody ever got fired for buying IBM” still rings true).
    The multinationals always have better contract lawyers and end up with contracts which suit them.
    The department (who are not blameless either!) typically will create an army of committees and contractors to represent the department in all engagements with the outsourced partner.
    The result is a total dilution of any responsibility, and paralysis through “design by committee”.

    • You have hit the nail right on the head with this one.I would add that most IT manager’s, CIO’s inside the government don’t know any better, so they keep to the old maxim in regards to IBM/Microsft etc…

  3. A lot of problems of this nature occur directly as a result of fear of failure. I know in our department we have a punishment culture where disclosure of a mistake is the last thing you want to do for fear of the consequences. So we cover those suckers up good and proper. Bury em deep is our motto!

    But that’s easy for me, I’m not a manager. If I was a manager and I had the actual responsibility for actually dealing with actual issues what would I do?

    Well I could start by reading a lot of columns which purport to know what to do. No offense intended here but it is true that when you don’t have the responsibility it’s easy to figure out what needs to be done, heck it’s even a well studied field of psychology – . Real responsibility though involves real people, real issues, real consequences. Most columnists and bloggers have a hard time keeping a simple comment feed in line and troll free let alone a government department filled with those IT people which have a reputation for being so mild mannered and agreeable in times of even the greatest stress.

    Managing a department isn’t too different from doing any kind of sporting activity. What separates a good cross country athlete/manager from the mediocre ones is how they’re able to respond, focus, and execute when a bear jumps out on the trail in front of you. Sure managers don’t have real bears (unless you’re lucky enough to be the manager of the bear exhibit at “Bear world” but I digress) but they do have distractions, hr issues, ir issues, mistakes, cost blowouts, and any number of off plan events. Their ability to deal with these issues while not losing focus is key. And the key is your focus isn’t on the bear, it’s on staying alive long enough to find a place to safely change your shorts.

    Unfortunately most management focus is on the bears. With reports on goverment waste constantly coming out what else can they do. The job description is no longer “Get to the finish line in a reasonable time frame and without ruining anybody’s life as we understand things happen” it’s now “Get to the finish line faster than humanly possible whilst also avoiding the buttload of bears we released into the race area and while a whole bunch of people watch from the safety of a cctv feed writing reports about where you went wrong and what you should have done instead”.

    • I think fear of failure is definitely something we need to get over as a culture. In the startup industry there is a growing awareness that failing is not a bad thing; you learn from every micro- and macro-mistake and continue, push on, move forward and eventually succeed. We need to take the stigma out of failure.

      I agree with your point that it’s easy to sit on the sidelines and snipe. I’ve worked as a systems administrator and done some business analysis stuff, but I’ve never been involved in major IT projects. But I do care about major enterprise IT projects and I do like seeing them succeed. I think there are few things of this level of complexity in our society, so they fascinate me.

  4. Hello Renai,

    Don’t know why you’re not reading The Infonomics Letter any more – but this month’s edition may interest you. I’d tackled the same repoort and provided a view of the issues through the governance lens of ISO 38500. My conclusion, and we don’t need a Gershon to tell us: Governance of IT in Victoria is broken. It can be fixed, but the fix requires action from the top. For the agencies that learn to govern their IT well, the future should be bright. For those that don’t, there should be no more funding for IT initiaitives.

    For those who are interested, The Infonomics Letter is at


    • Cheers Mark, I’ll try and get some time to check it out!

      For everyone else, I recommend Mark’s newsletter; he often has some interesting ruminations on governance in it.

  5. I have sat on the sidelines and watched any number of government IT-centric projects spiral into failure and doom. I have also worked in senior engineering roles on a number of $100m+ IT-centric projects both within Australia and overseas. All have been successfully completed on time and within budget (including agreed contract changes along the way). What is the single most important common factor that led to their success?

    Responsibility and accountability.

    When both the client and the contractor have clear and agreed responsibilities and accountabilities, and each accepts same, then projects work. The client agrees to a negotiated scope, the contractor stands firm on scope creep, deliveries are phased to build incremental capability, and “why don’t we add …” is banished to contract change proposals for delivery AFTER the baseline system is operational.

    It is not rocket science, it is responsible engineering and project management on both sides. It is also compentent managment of client expectations by the contractor.

    Douglas’ comments are spot on. It is not a case of contractors vs. public servants, it is government organisations needing to appoint individuals who have the authority to make decisions, are prepared to do so, and are supported by their senior management. (As an aside: For all the bad press they get from time to time, Defence are actually quite competent in this role and their success-to-failure ratio is orders of magnitude better than any other government organisation as well as many commercial ones. That said, they could also improve and have the structural insight to realise that.)

    Sad to say, my suspicions are that as well as general incompentence, many of these failed governement IT programs would fail strict probity checks at almost every stage of their award and implementation.

    The answer(s) are (in part) clear at least to me:
    – well scoped projects with limited, clearly defined outcomes for baseline capability delivery
    – acceptance of responsibility and authority by client and contractor
    – focus on phased deliveries to build final capability
    – acceptance by the client that changes cost money and time
    – acceptance by the contractor that changes cost money and time
    – post-ponement of extensions to outcomes until after baseline capability is delivered and operational

    What say you?

    • I couldn’t agree more, David — in fact much of my upcoming article on what can be done in Victoria will focus on accountability. It’s not rocket science, but it’s a facet of projects which hasn’t been emphasised in government recently.

      I also agree with your comments on Defence — I think they have a much higher imperative to make stuff work (the defence of the nation) and so they often just get things done where others can’t. Their more hierarchical structure may also help with this.

  6. I completely agree with the comments concerning accountability, however there is another key issue – the allocation of ICT projects to project managers, mainly contractors, who may be highly competent in Prince2 or other PM methodologies, but who do not have ICT qualifications.
    ICT project management needs both PM expertise but also domain knowledge and expertise.

  7. IT projects are risky and they always will be. They are risky in three ways (at least) which aren’t going to change:
    1. the customer/buyer is naive. They want something new.
    2. the tools and/or the product are novel in some way. This means R&D.
    3. business processes change must occur. The software can’t do everything, it’s a compromise.

    There’s nothing else like this. When I want a bridge built I have a pretty good idea of what it will be. The tools are hundreds/thousands of years old. The new bridge replaces a punt or smaller bridge.

    People funding IT projects need to appreciate that nothing can reduce risk enough. Instead, be ready to be more vigilant, spend more money, change and learn more than any other kind of project.

    • All true, and reinforcing the critical importance of a very strong governance regime that ensures proper attention to the balance of aspiration and capability, responsibility and accountability; and maintain focus on achieving the intended business outcome by doing all the work that is required, and not being seduced into thinking it’s only about the technology.

  8. Renai,

    I am afraid I see this quite differently, and I must confess I was shocked at how “off the mark” the Ovum response was. The Ombudsman’s report is first and foremost focused on an appalling track record of managing manage complex change rather than specific ICT issues. In fact, I didn’t read a single line of text describing the complex technical ICT problems that these various Victorian Government departments are grappling with. Rather, what I did read was that Victoria has a large portfolio of programs and projects to deliver ICT enabled change. Characterising major complex initiatives such as Myki simply as a ‘project’ , as the Ombudsman has done, is incorrect and masks the complexity inherent in the undertaking. Myki comprises many interdependent streams of work, each a major project in itself. The key to delivering a large portfolio of change relies on an understanding of the overall change so that the right capabilities, management approach and governance can be implemented at the project, program and portfolio level. This distinction is missing from the Ombudsman’s report, which challenges the likelihood of root causes being identified and the right solutions being implemented. The good news is that, despite the grumblings about the Gershon report and others in this article, the way forward is really pretty clear. Other government departments, at federal and state level, have undertaken targetted programs to benchmark exisiting capabilities and put in place capability development programs focusing on key gaps identified. Given the financial investment of $1.3 billion in ICT enabled initiatives, and the reported cost overrun of $1.44 billion, the direct benefits of improving portfolio management and delivery capability are compelling and clear.


    • hey Mark,

      thanks for your response, it’s highly appreciated.

      However, to be honest, I’m not really clear on what point you’re arguing. I think we can all agree that Victoria’s problems aren’t so much about specific ICT issues as they are about change and project management. But I’m not sure what you’re arguing about that. Could you maybe be more specific?



Comments are closed.