Vic Govt ICT strategy analysis: CenITex split,
cloud adoption on the cards



analysis The potential break-up of troubled IT shared services agency CenITex and the opening of the door to government adoption of the new cloud computing paradigm are two of the most important themes written between the lines of the Victorian Government’s major new ICT strategy released yesterday.

Yesterday the Victorian Government published what many will expect to become a landmark document in the long and troubled history of the state’s use of technology to support its operations: Its new whole of government ICT strategy. The document comes with strong credentials; it was put together by former South Australian Government chief information officer Grantly Mailes over much of the past year, and features input from some of the finest public and private sector minds in Victoria. CIOs, vendors executives, analysts, the government itself; all of these threads have been pulled together in this one overarching policy.

Its publication could not have come at a better time.

To put it bluntly, ‘technology’ is virtually a taboo word in the entire Victorian public sector at the moment. A string of audit and ombudsman’s reports published over the past several years have demonstrated starkly that the State Government, frankly, is not able to deliver on almost any major IT project it undertakes without letting those projects run dramatically late and over budget; and a string of controversies surrounding IT shared services agency CenITex have similarly made it clear that the state also has fundamental problems with basic IT service delivery to keep its operations afloat. To put it in words almost everyone can understand: Right now, Victoria is bad at building IT systems; and it is bad at maintaining them and providing associated IT services once they’re built.

Will the delivery of the state’s new ICT Strategy fix those problems? Well, no, not overnight. It will take more than the mere publication of a policy document to do that, no matter how lofty its statements and how detailed its aims. In addition, the ICT Strategy isn’t just concerned with major IT projects and IT service delivery. The policy is actually an umbrella document containing the government’s plans in a huge range of areas; ranging from collation and organisation of government smartphone apps to the release of government data under open licensing; from privacy to procurement, the ICT Strategy goes virtually everywhere. Much of what is contained in the ICT strategy is what I would define as ‘housekeeping’ — stuff that a competent government should be doing anyway, without anyone really needing to tell it do so.

However, there are still several underlying themes in the ICT Strategy which presage fundamental shifts in the way the Victorian Government does IT project work and IT service delivery as a whole, and these should be considered separately.

The first of these is the potential break-up of troubled IT shared services agency CenITex.

If you read the Australian technology media at least sporadically, you should be aware that the agency, formed several years ago from the merger of a number of other IT shared services divisions within the State Government, has its share of problems. A recent Ombudsman’s report damned the agency for serious improper conduct in its procurement processes (the police were also brought in to investigate the situation), CenITex has suffered a number of issues with delivering basic IT services (such as email) to its client agencies, and the group last year went through substantial redundancies. Ultimately the state government sacked CenITex’s board in April last year and is currently considering its future.

What that future might look like was first outlined in an economic statement issued by the Victorian Government just before Christmas (PDF). The document stated that the State Services Authority had conducted a review of CenITex. “Its findings will inform decisions about how to best deliver shared IT services across the public sector,” it noted.

However, it was a set different paragraphs which really set tongues wagging in the technology sector. It stated:

“The Government will scrutinise the efficiency of all government agencies, focusing initially on those providing mandated services to business or regulating the private sector for a fee or charge. The quantitative reasonableness of fees and charges will be examined against enhanced cost recovery principles, aiming to drive down charges through efficiencies. The Government will also examine whether and which (if any) internal services can be more efficiently provided by the private sector or made more contestable. Initial candidates will be CenITex and the Victorian Managed Insurance Authority (VMIA).”

This statement appears to imply that at least some of CenITex’s work — which broadly consists of providing basic IT services to departments and agencies — could be taken away from it and handed off to the private sector. And the economic statement also went further, stating:

“Government will use market mechanisms to drive down unit price and increase information and communications technology (ICT) system and service innovation. Government will favour shorter contract terms and open standards to increase competition and guard against technology lock in or single vendors securing a disproportionate share of government business.”

“At the infrastructure level, Government will progressively withdraw from direct ICT service delivery. The Government will determine the mix of insourced, managed and outsourced service delivery which is most cost effective, most responsive to business needs and which best leverages the expertise and opportunities available in the market.”

This statement is repeated in the ICT strategy released yesterday, and strengthened by other core principles, which outline how competition and market mechanisms will be used to drive efficiency and innovation in ICT systems and services, and how ICT services will “take advantage of industry capabilities”. “Commercial off-the-shelf software or outsourced services will be adopted in most cases,” the strategy states.

When you couple this direction with rumours inside the Victorian Government that CenITex is going to be sold off and the fact that IT shared services agencies right around Australia are not delivering on their initial promises … it’s not hard to join the dots. The Victorian Government is tired of CenITex acting like an odious millstone around its neck. The writing is on the wall for the agency to go through a substantial restructure and re-alignment as a minimum, or for parts of it to be broken off and sold to the private sector in the extreme case.

The second major theme within the ICT Strategy is the potential to adopt cloud computing services to resolve some of the State Government’s ongoing IT project and service delivery problems.

Cloud computing is mentioned right on the first page of the report in its introduction by Victorian Premier Ted Baillieu and Assistant Treasurer and Technology Minister Gordon Rich-Phillips, and it’s only a few pages later that the following paragraph appears: “The uptake of managed ICT services is increasing. The speed, flexibility and economies of scale offered by cloud computing are prompting organisations to rethink what needs to be delivered in-house. It is financially responsible for the Victorian Government to consider these alternative delivery models.”

Cloud computing pops up again in the strategy’s section on ICT procurement — where it states that agencies should, by July this year, “identify opportunities to pursue service-based ICT offerings” and, by December this year “demonstrate use of service-based ICT procurements”. Directly after that section, the report emphasises the message with a case study detailing the Department of Business and Innovation’s successful deployment of Salesforce’s customer relationship management solution, which is used by over 450 staff within the department.

“DBI’s Salesforce implementation is on track to deliver a 40 per cent cost saving over five years compared to the estimated costs of a custom-built option,” the case study concludes. Well. If that isn’t a clear message to department and agency executives and CIOs, I don’t know what is.

In the ICT project delivery section of the strategy, the use of cloud computing technologies is hinted at again, where it notes that Victoria needs to move away from ‘large, complex ICT projects’ that have been heavily customised, and towards adapting business processes to make the best use of existing market offerings. This approach fits in perfectly with the emerging cloud computing paradigm, where it is becoming much more common for customers to standardise their own business processes to fit with a standardised software as a service solution used by many similar organisations, rather than trying to modify the technology itself to fit their non-standard needs; a path fraught with peril.

Some groups within the Victorian Government are already following this lead. For example, the aforementioned Department of Business and Innovation recently went to market for a major Infrastructure as a Service (IaaS) solution, in a move which may make it something of a leader within the Victorian Government.

Taken together, these two major trends hinted at in the Victorian Government’s new ICT Strategy represent the chance for an extremely powerful and positive overall paradigm shift within the state’s public sector. Shifting emphasis away from its ailing internal IT service delivery mechanisms and towards more standardised services from external providers is something that Victoria clearly needs to do; it remains unclear if the state will ever be able to get enough IT talent to keep its IT service delivery on track otherwise; CenITex itself already makes extensive use of external contractors to keep it afloat.

In the long-term, the gradual adoption of cloud computing services represents a chance for the state to standardise its business processes and shift costs and energy away from maintaining its own IT systems and applications and letting others maintain them., Microsoft, Amazon, Rackspace, Fujitsu, Oracle, SAP, CSC, Telstra, Optus, Google … all of these companies have cloud computing offerings which they are better at delivering than any Victorian department or agency could do on its own. Adopting these platforms and taking advantage of these economies of scale makes complete sense in the long term and represents a chance for Victoria to break away from its cycle of failure in IT project delivery.

One last item of significant note within the ICT Strategy which it is worth keeping an eye on its the creation of Victoria’s first whole of government ICT role for a few years. This isn’t explicitly a chief information officer role, although it will fulfill some of the same tasks; it’s labelled as ‘Chief Technology Advocate’. Personally I like this term; it represents the fact that whole of government chief information officers around Australia haven’t been able to make much of an impact on fundamental IT project and service delivery, due to the lack of control they have over powerful departmental CIOs.

Victoria’s new CTA will advocate for positive change in the State Government, but they can’t mandate it; they will report on ICT initiatives; but they won’t conduct those initiatives themselves. I like this positive, forward-thinking spin on the whole of government CIO. It will be interesting to see if it can have an impact where Victoria’s previous whole of government CIO failed.

I said it yesterday, and I’ll say it again today: I strongly approve of the Victorian Government’s new ICT Strategy. It hits all the right notes, and has a concrete set of deliverables and timeframes for the state to deliver on. Plus, there is obvious incentive for the political administration to whip the public service into action on this one. Baillieu and Rich-Phillips can only harp on about ‘fixing Labor’s bungles’ for so long before the Victorian electorate will want to see concrete action on the Coalition doing so.

Image credit: Benjamin Diehl, royalty free


  1. Yep, the private sector will do it so much cheaper! Just the PPP’s. Most government agencies want the Limousine model but only want to pay for the poverty pack. This will be interesting

    The police investigation is focusing on the PREVIOUS management and not the current.

  2. I never cease to be amazed how shallow comment is on the Vic Gov. Information Technology by commentators. There are so many services that Government supplies some of which can never be provided by cloud solutions simply because of the need for privacy and security.

    There will always be a necessity for infrastructure even if cloud solutions are used. For instance Network, Identity Management are two of the more obvious needs but there are others. A government technology group will need to do far more than advise. Much of the letdown of CenITex was because of failure to ensure these crucial requirements were not properly thought through, is it possible the same mistakes are to be made again?

    Perhaps the Press which has done a dismal job so far of reviewing some of the less successful projects should wait to see the results of the strategy which was a very high level document instead of offering what sounds like political fawning to the Ballieu Government who do not have a great record in delivery so far in their short term.

    • It appears to me is people Think Government = Coles it’s one thing so it just needs one system
      The Government is a collection of in some cases extremely varied business.

      It’s a bit like telling Rio it can only use the same software and vendors Coles uses across the board.
      That was what CenITex tried to do

    • Yup … fair enough Neil … things tend to proceed in a “baby out with the bathwater” fashion from one extreme of over-centralisation to the other extreme of over-decentralization … twas ever thus I suppose. What we need is less bull-in-a-china-shop ideology and more focus on pragmatic solutions that are sensitive to agency needs and within the ability and financial means of folks in the Vic Gov to deliver.

      It will be a big mistake to outsource CenITex in one big problematic lump in a traditional knee jerk manner … which would be akin to kicking the troublesome teenager out of the house. This seldom works as a parenting strategy as the blighters tend to return later with even bigger, and more expensive, problems! There are many tragic tales of ‘big hairy outsourcing’ of government ICT in this country …

      Cloud services, though, are turning out to be increasingly cost effective for some applications, workloads and categories of data … and at least have the benefit of being available today rather than promised tomorrow (“cloudy is as cloudy does”) and granular … they can be tackled on an application-by-application and workload-by-workload basis in a much more agile manner. BTW, the economies of scale are real … I was in Washington last week talking to AWS public sector team and they stated that AWS has reduced its prices 25 times since 2006 (!) How many government shared services or traditional fixed-contract outsourced arrangements could ever hope to achieve this level of operational efficiency? The economies of scale of cloud services are real and material. In the US AWS is certified to FISMA Moderate, and is regarded as policy compliant for many categories of government business … and indeed is being used by many federal government agencies. AWS, and the other leading global and national cloud services vendors, will mature pretty quickly in Australia to offer similar benefits. Cloud services are not the the answer to everything, but they are a useful new tool in the toolbox.

      My view is that we should value the useful aggregation of decision rights that CenITex has achieved in getting a big pile of ICT infrastructure and applications under a single general manager – so that proper strategic decisions can be made. From this point CenITex itself needs to treat cloud service providers as its best friends rather than as unfair competitors. CenITex needs to position itself as a service aggregator and broker to transition its infrastructure and applications to cloud sourcing – to lower its own prices and so that agencies don’t need to act unillaterally in a fragmented and sub-scale manner.

      It would actually be a travesty of common sense (and public value) for CenITex to go the way of shared services in WA and just be broken up and devolved back into agency-by-agency fragmentation … or even worse sold as a Gordian Knot of complexity to some unsuspecting IT vendor. Outsourcing CenITex won’t solve the root cause of many of the governance and funding issues – indeed they will likely become more intractable when discussion can only occur within the constraints of the outsourcing contract and with a table full of lawyers.

      CenITex has a CEO … let him decide what the best way forward is … at least he understands the landscape, what is under the hood (yeech!) and what his customer agencies are seeking. There actually is no magic wand here ….

  3. Yeah yeah yeah, pretend to break up CenITex – give it a new name, I’ll move desks a few times and nothing will change. Been in this game long enough to know it’s all smoke and mirrors.

  4. as there is no mention of the zombie apocolypse this is not the nick everyone thinks it is…..

  5. Quoting Renai. “I said it yesterday, and I’ll say it again today: I strongly approve of the Victorian Government’s new ICT Strategy. It hits all the right notes, and has a concrete set of deliverables and timeframes for the state to deliver on”

    Can’t say I agree with you Renai. This strategy is all a bit motherhood. It doesn’t say anything most other organisations wouldn’t/don’t say.

    If we look at the principles in the Victorian Strategy, they are:

    Policy and service delivery programs will use popular digital channels
    Policy and service delivery programs will be increasingly co-designed and co-produced
    Information will be shared, open and managed as an asset
    ICT-enabled projects will be staged and focused on managing risks and delivering business benefits earlier
    Competition will be promoted to drive efficiency and innovation in ICT systems and services

    They are just statements of the bleeding obvious.

    A good decision has a realistic and viable opposite. I don’t think they do.

    IMHO, there are too many loose and vague statements in the strategy. It doesn’t recognise many of the problems that they will have to face up to when implementing some of their actions.

    For example “Information and data will be shared across government and with Victorians”
    Is that ALL information? Or some? If some, what?
    What are the issues in sharing information on Citizens?

    The strategy does not explain that there are problems here or what they intend doing to resolve them.

    And one of their actions is “Update Victoria’s web portal to provide search-driven access to government information, services, directory information and mobile apps” by March 2013

    So they want to provide access to all government information by the end of March. Six weeks. Pull the other one.

    IMHO, it’s a typical ICT strategic plan put together by IT people. Mostly same old same old.

    And it certainly ignores the elephant in the room.

    50-60% of large ICT projects fail. Victorian government ICT projects have a poor track record. The elephant in the room is the question “what are we going to do differently so that we get a different result – preferably success, not just failure in a different way.

    • Hah! Good on you Bernard … Elephant indeed … in truth there is a herd of them hiding in the jungle. The good thing about the strategy, however, is that is at least an adequate reference point around which activity can be mobilised. There was a period of directional vacuum prior to the strategy’s publication … and now there is at least a period where at least there is some clarity of the general direction of travel.

      There are so many threads which have come unraveled in Vic Gov ICT over the past 5 or 6 years that it is too difficult to knit them all together in one go. Baby steps are required I think to rebuild some degree of motivation and confidence in the troops.

      The strategy just lays out some common sense assertions which create some basic direction and an authorising environment for Grantly Mailes and his new team to come up with a cunning plan … without necessarily being constrained by anything too inconveniently prescriptive in the strategy.

      • It’s a bit sad when common sense and motherhood are a great improvement on the past.

        The strategy talks about current gaps in ICT leadership, governance and skills, but the actions they are proposing don’t fill me with confidence that they will improve matters.

        IMHO, governance on its own doesn’t achieve much. You can have all the committees, plans, oversights, reports, accountability etc etc as you can afford. That won’t deliver unless you have competent people in the right places.

        Statements like “Agencies will adopt best practice governance standards such as the international standard for corporate governance of information technology (ISO 38500).” look like “we must do something, here’s something, let’s do it”. To plagiarise Samuel Johnson, best practice is the last refuge of a scoundrel.

        A final observation, the word competence is mentioned not once in the strategy. Maybe there are two elephants in the room.

  6. With respect …. irrespective of whatever infrastructure management mechanism has / was or will be in place to service government’s ICT requirement, until the ICT strategy was published the ‘elephant(s) in the room were / are and will potentially remain if in my opinion, two things are not addressed:
    ‘We’re special and couldn’t possibly operate within a standard architectural environment or within the framework of common software applications without substantial modification, amendment, add-on’s and API’s to 25yr old systems which we can’t touch because we don’t know what might happen’.
    and .. with the commercial contractual capacity to work with the vendor community with a split of 10/90 % risk apportionment in favour of suppliers. This presents both an open invitation for vendors to load prices (pip pip ‘value for money’) and places a wholly unreasonable burden on internal staff who while ‘trained’ don’t have close to enough support from the designers, manufactures and sellers of the highly complex technology they are working with.
    What the ICT strategy suggests (not strongly enough I believe) is that for the first time in years, in advance of designing the technological equivalent of Dr Frankenstein’s monster which while as ugly as sin itself does match and accommodate the “dunno, it’s the way we’ve always done it” approach to applying technology solutions to existing policy and procedure, the question of ‘would it not be sensible to explore changing the way we operate (there have been advances in the last 20 years) instead of adapting the technology to fit our old and out-dated methodology’ is set to be asked. COTS unless you have a very good reason why not. The second and potentially bigger impact contained within the cloud proposition is that it represents a sea-change for government in context of sharing risk. Cloud, SaaS and hybrid variations of outsourcing models will have SLA’s which will carry significant financial penalties if not met. That CenITex was ‘sent into battle’ with ‘best endeavour’ as its performance guarantee was never going to work (and IBM, HP, Fuj would be exactly the same).
    COTS, Cloud and shared risk is what I read in the strategy which IMHO is an appropriate aspiration. What the clan chiefs will make of it and what level of persuasion by those suggesting it materialises in any substantive form to make it happen will be the ‘sporting contest’ watched by us casual bystanders for I’m guessing at least the next couple of years.

    • You have hightlighted one of the key challenges in government ICT Tom. I discuss this in an Ovum StraightTalk comment published last week:

      The gist of its is as follows: “It is time for senior executives to understand and embrace more agile approaches to sourcing and managing ICT. The way executives frame requirements and their expectations for ICT projects are significant factors affecting project performance. Blinkered, inward-looking, technology-centric requirements and traditional waterfall approaches often focus on the means at the expense of the ends. They seem proven and safe, but can lock agencies into long, costly projects, which too often fail. More agile approaches, in contrast, are mindful of the outcomes but flexible about the approach. They reduce risk by encouraging creative problem-solving and leveraging pre-existing capabilities – such as cloud services – to practical effect. This can seem adventurous for some executives, but can also deliver better outcomes by enabling more pragmatic and cost-effective trade-offs between business needs and solution possibilities. Leadership is required to see agile approaches as prudent and effective, and to authorize staff to concentrate on ends and be more flexible about means.”

      • Steve,

        From your article: the business usually says something like:

        “I’d like to acquire an ICT solution to support today’s internally focused understanding of business requirements.”

        I agree that this statement will probably lead to a failed project. I also agree that what is needed is a more agile approach.

        I believe you are proposing a more agile approach to solutions development. You mention agile problem solving in the summary but this applies to problems in the solution domain, not the business problem domain.

        What I suggest is that flexibility needs to be introduced into understanding and solving the original business problem.

        The statement, above, implies that the business knows what its problem is, has properly understood it and has come up with a solution as defined by the requirements, and yes, they are often over engineered.

        In reality, the business has not done all that. Someone has come up with a “Good Idea” which gets turned into requirements and needs. All too often they have not thought it through, have not identified all the problems their “Good Idea” will create and no amount of agile approach can fix it. The requirements and needs are a well specified solution to a poorly understood problem.

        Unfortunately “agility” has acquired a specific meaning, one around the development of functionality. IT systems also have non-functional requirements that agile approaches tend not to address, neither do they address consequential problems like archive, back-up, DR.

        I prefer the term flexibility – or decision making in the context of uncertainty.

        I believe that flexibility should extend well into the problem space and not be limited to solution development, even if expressed as “capabilities”.

        After all, it’s solving the business problem that matters, the solution just costs money and causes more problems.

        So I agree with what you say about flexibility, but recommend that the business recognises that it is not very good at solving problems.

        • Hey Bernard, hmmm … well it is a bit of a generalisation to assert that “the business” is not very good at solving problems. I suppose the essence of my argument is that a more agile approach adopted at a more senior executive level and leveraging pre-existing solution elements (such as cloud services) is better because it creates shorter cycles of iteration between “what is our problem?” and “how can we solve it?”

          Perhaps the original understanding of the business problem was imperfect but the best way to discover that is by a process of iterative solution refinement – and cloud services provide a way to do this with minimum up-front investment and in shorter timeframes. I wrote a very good case study of this sort of behaviour based on the experience of the Department of Business and Innovation in Victoria and their use of Salesforce as a platform for citizen/business interaction management and grant administration:

          Actually, Chris Gilmore from DBI will be presenting on this case study at the 3rd Annual Cloud Computing Forum in Canberra on Wednesday.

          The point is that we need a more agile approach to executive thinking and decision making which leverages the increasing relevance and value of pre-existing ‘as-a-service’ solutions – hence ‘external-in’ logic (i.e. what solutions are available?) needs to replace traditional ‘inside-out’ logic (i.e. what do we need?” as the starting point in procurement.

          I’m not suggesting that no business analysis/scoping etc. is necessary – just that we need a better balance in order to remove the blinkers that are too often placed on projects too early in the problem solving cycle. As cloud services mature the benefits of working out how to leverage shared/standardised solutions as opposed to dedicated/customised solutions will become increasingly compelling.

  7. Tom,

    re: ‘would it not be sensible to explore changing the way we operate (there have been advances in the last 20 years) instead of adapting the technology to fit our old and out-dated methodology’

    I agree, totally. A statement like that would be an answer to my question “what are we going to do differently so that we get a different result – preferably success, not just failure in a different way.” That’s what’s missing from the strategy. That’s what would change it from motherhood to something useful.

    In order to explore changing the way they operate would require an approach driven by the problems the organisation faces. IMHO, it’s always a good idea if the IT department helps solve business problems, rather than cause so many of its own.

  8. Bernard, have a 14 hour flight to Abu Dhabi tonight so I’ll read the report again …. as I understood that which I proffered was actually in it …. or is this just what I wanted to see so badly I mentally read something which is not there??

  9. Tom,

    re: ‘would it not be sensible to explore changing the way we operate ….’

    To clarify, we = Victorian Government.

    and I said “In order to explore changing the way they operate would require an approach driven by the problems the organisation faces.”

    They (= IT department) operate

    The organisation = Victorian Government.

Comments are closed.