“Chaos” in NSW Govt IT shared services


news A landmark report into the management of the NSW Public Sector commissioned by the state’s new Coalition Government has described how dozens of overlapping and competing systems and services providers have created “chaos” when it comes to the state’s current IT shared services paradigm.

The NSW Commission of Audit is a group led by senior NSW bureaucrat Kerry Schott, and chaired by influential businessman David Gonski. It was put together by the current NSW Coalition Government after they took government at the last state election, with a view to developing an overarching framework for the future of the NSW public sector. Last week, the Commission produced a report into their examination of the sector (PDF). The NSW Government also published its response (PDF).

Across Australia, State Governments ranging from Western Australia to Queensland and Victoria have struggled over the past decade to successfully deliver IT shared services to different government departments, with most states walking away from the paradigm in general over the past several years. In the report, the Commission outlines a similar situation in NSW. The state currently has a number of shared services providers — two serving the health and education portfolios, two more being established for the transport and attorney-general/justice portfolios, and two more that serve multiple departments and agencies — Service First and BusinessLink.

In the report, the Commission noted that the two biggest shared services organisations — health and education — were the most mature, although they needed “substantial investment” for technology upgrades and process redesign. But the story was different when it came to the others.

“Neither ServiceFirst nor BusinessLink currently meet customer expectations across a number of vital functions,” the report found. “Neither is truly integrated and in many instances the level of service is poor. Nine years after its establishment, BusinessLink is now in the process of integrating its three SAP systems. ServiceFirst has three legacy systems, inherited from its predecessor units. More important, its staff morale is low and customer satisfaction is poor. Clients of Service First are united in their frustration and dissatisfaction with the service provided. To be fair to the shared service providers they have been starved of funds and have not been able to modernise.”

The report found that many departmental directors-general could not use their IT services to conduct basic tasks, such as accessing aggregated data on their clusters’ financial position, review important information about their workforce, or even communicate with all staff on a single email system.

“The chaos relating to multiple systems is evident in the former Communities NSW cluster,” the report states. “In 2010 this cluster, with around 3,000 staff, was supported by an astonishing 15 corporate service providers, ServiceFirst and 15 in-house shared service providers. Altogether, it relied on 76 different corporate systems. This multitude of largely unrelated systems is not unusual. The Transport cluster for example, has 130 separate systems.”

The report added: “The poor support provided to front line workers is evident in all areas of transactional shared services: finances (e.g. invoice payments), HR (e.g. payroll processing), OH&S (e.g. claims processing), ICT (e.g. helpdesk support), records (e.g. archiving), property and facilities management (e.g. office leasing) and asset management (e.g. maintenance).”

One of the key issues in the state when it comes to shared services is the “lack of investment” in upgraded and standardised technology. “The Commission notes that much of this expenditure is on IT systems and there is a lack of confidence in the sector about this type of procurement in particular,” the report states.

Currently, corporate shared services functions are being consolidated within each super-cluster of agencies and departments being set up by the NSW Government. In particular, BusinessLink and ServiceFirst are to have their customers consolidated into a new shared services provider, NewCo.

The Commission of Audit’s key recommendation in this area is to highlight the importance of change management in migrating shared services to new providers/systems, as well as a number of more specific recommendations for each cluster. In response, the NSW Govt said its shared services reform efforts were proceeding along the lines recommended.

Separately, the report made a number of recommendations related to IT procurement — focusing, for example, on whole of government procurement efforts where possible rather than procurement by individual departments.

Well, now we can add New South Wales to the list of Australian states with disastrous IT shared services implementations. Already on that list: Western Australia, Victoria and Queensland. And I’m sure the other states are only an audit report away. These comments by Gartner distinguished analyst Andrea Di Maio in November 2011 on the global government IT shared services paradigm and the ongoing series of disasters in Australia in particular are now looking increasingly prescient:

“… it appears that shared services are having a hard time. Queensland, Western Australia, South Australia, Victoria have all had their fair share of issues with shared services, and this is happening quite consistently in other parts of the world.

There is no doubt that duplication of services and spending across different government agencies makes little sense. 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.”

In short, Australian Governments need to start thinking smarter about how IT shared services are provided, and looking outside their operations for help. The cloud computing deal which the NSW Government’s Trade & Investment agency recently signed with German software giant SAP for corporate business platforms may be a good start. Or it may not. We’ll see how well the project keeps on track. Because almost everything else in Australian State Government IT appears to be going wrong right now.

Image credit: Michelle Dennis, royalty free


  1. Hmmm … I continue to muse over the best way forward. One way to think of this is that we have failed to construct sophisticated adult relationships between the central agencies and line departments and agencies. We lurch back and forth between regarding agencies as naughty untrustworthy children that need to have mummy and daddy do everything for them (and are surprised that they are resentful of parental ‘services’) and then if that doesn’t work we kick the little blighters out of the house and leave them to their own devices … but then have to take them back in later on when they run out of money and get themselves in a mess.

    The desired outcome is that agencies are mature independent adults … making sound sustainable decisions and collaborating well with, and supporting, other members of the family for everyone’s mutual success. It is all in the parenting style … we need a much more sophisticated and practical approach than the simplistic mechanistic advice served up to governments by the big consulting companies. Behind each of these shared services fiascos is a consulting firm that should be ‘value billed’ for the negative value delivered by their advice.

    When I say parenting, of course I mean leadership. We really need to reduce this mess back to a well-grounded understanding of the way the public sector operates and a pragmatic approach based on common sense appreciation of the ‘art of the possible’.

    The focus needs to revert to understanding your children’s (ahem … agency’s) individual needs and supporting them to equip themselves with the skills and resources to succeed by intelligent pooling of the family’s collective resources and selective outside assistance.

    Domineering and incompetent parents who take bad advice from the wrong people fail both themselves and their children.

    • I think this is an excellent analogy. In families, as with central IT divisions and ‘child-like’ agencies, you can’t just cut off the kids from parental support when they misbehave — you need a more nuanced approach to encourage the right behaviour.

      Maybe it’s all about education. With children, if you set the boundaries for what they can and can’t do, they will react well and grow, until they reach the next level and need the boundaries reset wider. Perhaps the key with departments and agencies is to start with baby steps. As they gain confidence, give them more power and control over their own destiny.

      One thing I would like to see is an injection of private sector CIO talent back into the public sector. Some of these guys, with the huge salaries they have commanded with major corporations, would have little to lose in a public sector environment and may be able to take harsher measures to enact change. All too often it seems like IT chiefs don’t have enough leverage in the public sector to insist upon good project governance and so on.

  2. One of the reasons that this always fails is the tendency to have the need to run everything

    Because it’s “The Government” they continually feel that every department is the same so they tend to dumb things down to the lowest common denominator.

    The issue is the further exasperated by the need for the shared service provider to start making decisions based on it’s costs and ease of use rather than the “Clients needs” . Leading to frustration on the part of the client businesses and the game playing that ensues.

    The shared service model across such a large collection of disparate entities with vastly different and specialised needs is a pipe dream. If they could hold the shared service model to just “Business Services” Email,Finance,Payrol/HR etc then there is a good chance it will work and produce savings.

    Saying the Government is not like saying McDonald’s not every department is the same. Even the legislation covering agency’s within some departments is different.

  3. This is very interesting reading for me in BC – Canada – a province that is moving towards Shared Services…and Dr. Hodgkinson – your comments are so very relevant to what I am observing.

    Thank you!

  4. Thanks Cindy … Renai is doing a great job raising visibility of these issues on this blog.

    I wrote an opinion piece on Canada’s shared services strategy last year:


    Essentially I think it is doomed from the outset (sorry!). The problem is that once such “bull in a China shop” decisions are made then you are stuck with 3-5 years of “Emperor’s new clothes” dynamics … you can’t criticise the strategy without being seen to be “part of the problem not the solution …” so everyone just sighs and mumbles under their breath while trying to minimise the damage and waits for the eventual reversal of the pendulum (again) when the government changes or things become so bad that the Emperor is indeed observed to be naked. [By the way, I sincerely hope I am wrong about this and some kind of a miracle occurs in Canada …]

    The core of the problem is that you can’t make a silk purse out of a sow’s ear. A tue shared service is a bundle of people, process and technology that is architected to responsively and efficiently serve a diverse range of customers. You can’t create this just by wishing it were so, and you can’t create it when the starting point is an underinvested outdated and dysfunctional mess of people, process and technology which might support one agency OK, but is not, and maybe cannot, be a shared services for multiple agencies. Here are my comments on a tragic case study of shared services failure in the Western Australian state government:


    This is not to say that shared services in government can never work – they can – but they are a very high risk and high cost strategy. The public policy disaster is that they are often put forward to niaive government ministers by self serviing consultants as a low risk strategy to cut costs … which they never are.

    The final point that I just keep hamering on about is that internal shared services strategies (high risk high cost) are being overtaken by cloud services strategies (low risk low cost). If cloud services exist that meet government’s needs then use them!!! Don’t even think about trying to turn that sow’s ear into a silk purse. I discuss this here:


    Shared services are primarily an organisational strategy … usually implemented without adequate control over all the organisational levers required for success. As it happens cloud services are primarily an organisational solution. I discuss this here:


    Governments just need to wake up to the simple fact that cloud services are just pre-assembled bundles of people, processes and technologies … shared services that actually work. Why persist with internal shared services strategies that are proven not to work when a better way forward is already emerging … and will become more mature over the next five years? The paradox for Canada will be that its shared services strategy will fail in the same time period of time that cloud services emerged as the ‘obvious’ solution to this problem. Analyst will lokk back and wonder why Canada didn’t see this coming …

    Here are some of my reports (publically available as whitepapers) if you would like more detail on how we see the cloud services opportunities for government ‘down-under’ in Australia:



    Sorry to be a ranter!!


  5. I take a different, but parallel view.

    IT services are (supposedly) solutions to problems.

    Shared IT services try and use the same solution to solve multiple problems.

    Unfortunately, the problems are sufficiently different such that the shared services do not solve all (or even some) of the problems.

    There are only three options:

    1. change the problems so that a common solution will work. This often means changing the way organisations work. Not an easy task.

    2. develop a solution that will solve multiple, different problems. Not an easy task.

    3. recognise that shared services seems to be a good idea but when you look at it closely, is not an easy task.

    Where I differ from Steve Hodgkinson is on cloud services. These are are just another set of solutions, but by definition based upon standards and commodities – and which probably won’t solve government’s many non-standard problems any better than any other solution. Especially when you factor in the new problems such solutions create. Contract and SLA management being just two.

    Where I totally agree with Steve is:

    “The public policy disaster is that they are often put forward to niaive government ministers by self serving consultants as a low risk strategy to cut costs … which they never are.”

    To which I’d add self-serving cloud vendors.

    • Fair comments Bernard … my main enthusiasm for cloud services is based on the logic that they can (actually) provide solutions for your challenges 1 & 2. By solving challenge 2 (standardised, multi-tenant, configurable, iteratively evolving solutions) you create the new agile thinking and behaviours that can start to provide solutions to challenge 1. Victoria’s DBIDepartment of Business & Innovation has used Salesforce as a common/shared platform for interaction management (CRM) and grants administration – and this is an excellent real-life example of a new paradigm of ICT in government. When cloud is the platform, agile is the behaviour (yes … really!):


      This is not suggest that cloud services can, or will ever, be able to address all or even most government needs … but they can address some of the more commodity like activities today and will become better over time. The direction of travel is pretty clear …

      The really interesting proof-of-the-pudding-is-in-the-eating will be NSW Trade & Investments use of SAP ‘s SaaS ERP solution as a shared services platform for 9+ agencies.


      If this works (finger crossed) it will be a game changer for government ICT and shared services.

      You are right to highlight that cloud vendors can be as venal and self serving as any other … we are already seeing companies like Salesforce recruiting ex-Oracle sales executives so the same-old same-old sales practices will no doubt play out. The difference, however, is “cloudy is as cloudy does”. Moreso than ever before it is actually possible to “kick the tyres” on cloud services before you commit to buying and to start small and evolve forward in an agile manner. The days of multi million dollar “bet the far, sign the contract and hope for the best” 3+ year waterfall method projects with no practical way out are hopefully drawing to a close.

  6. It does seem to be clear from the list of states who are experiencing problems that the failures aren’t necessarily due to incompetence. With so many states trying and failing either shared services don’t work or to get them working successfully is really hard.

  7. It is very clear that this strategy is not going to work in NSW just like it has not worked anywhere else. Shared services will not work when there is a top heavy structure, no staff development, no remote support staff, multiple operating environments, low staff morale, and no sense of ‘belonging’ for it workers.

    Yet they sail into their preparations for major failure (it’s already at a fail state now – pardon the pun) by merging bizlink and servicelast.

    The problem is not the staff within these departments; they are talented and smart people – and they’ve been landed in departments run by bean counters. The problem is that they are being led on a text book journey, on a ship that has no sails, is leaking heavily, has engines of differing power on either side – so what you end up with is a heavy ship, sinking, going in circles, with no backup plan or sense of direction.

    Thankfully there is such a thing as a lifeboat, it’s called dissemination of these large, clunky, slow, irritating conglomerates.

    Pull the plug now, you’re wasting your time.

  8. The push for Shared Services, to me at least, seemed to hang it hats on the “dividends” it would gain. Typically there would be a list of ideals like uniformity and consistency of service across departments etc but the real carrot for governments was a perceived cut in expenditure. This core dividend (forget what they publicly said and spun to the electorate and staff – expenditure cuts was the big one) has never been achieved. The department ICT services don’t shrink commensurate to any new shared services layer coming on-line. There may be an initial claw back but, given the known localised support issues and the like, the department ICT services eventually just come back to their original size – or indeed in some cases actually have more than their original size.

    [As a digression a similar argument can be made for government contract services where the in-house department staff size seems to inevitably grow – sometime from the very start even.]

    For the Shared Services product itself the issue becomes size and complexity. When each department (or even departmental region) does their own processing on localised servers then you end up with a web structure – a failure on one part wouldn’t necessarily bring down another. The hardware involved is also less likely to be stressed. Roll all this processing into one centralied processing unit and you have a statewide incident when a failure occurs.

    Similarly, all the departments use, say , an accounting product. invoices are generated and the like. In a true Shared Services model you would think that accounting services could be standardised and shared easily – centralised and run off a common application platform such as SAP. But departments often use a core business product to manage their primary service. Health has a PAS (Patient management System) targetted specifically at patient management. Now these products often have accounting functionality built-in – fully integrated. Is it wiser to attempt to introduce a different financial package on the back of this or to just leverage the full product they have already purchased?

  9. “with a view to developing an overarching framework for the future of the NSW public sector”

    It needs to be accepted by our elected reprehensibles that no matter how much we pay for the public sector, it is never enough. Attempting to spend money to rationalise the public sector and their systems is akin to drilling holes in the hull of a boat to correct a list. Get rid of two thirds of them and the laws and petty officialdom they represent and then IT projects for the remaining third, delivering truly essential services, becomes a lot easier and cheaper. Unless we break the toxic cycle whereby the public sector demands money to enhance the public sector in order to craft laws to demand more money to further enhance the public sector then Greece here we come!

  10. The problem of shared services is primarily not about IT systems. It is about the habit of Princes protecting their turf. Their people are “special” and have different requirements. They need a system which addresses their special needs because they are not like the rest of the world. Be it finance, HR, payroll, content management or any other facet of business. And it is definitely not confined to Government. The failures of Government are just more public.

    • The new generation of IT systems can, however, be part of the solution. You are right about “Princes protecting their turf”, but what if we just accepted this as a legitimate behaviour … i.e. agencies have a legitimate right to tune their processes to suit their business needs? What we need are solutions that work in this context rather than assuming that we can change the context … or drain the swamp of governance complexity in government.

      Over 100,000 different and diverse organisations use Salesforce to deliver a wide range of applications by configuring a standardised solution platform.This model allows the achievement of economies of scale while not requiring each of organisations to adopt standardised “one size fits all” processes. This is the future of shared services. The secret of success is starting with the right multi-tenant, configurable, platform. Then you can just focus on being a friendly and responsive service provider!

      • IMHO, governments usually have different requirements from those of the private sector. The service delivery agencies often do things that private industry just doesn’t do and there are different archive and security requirements. This usually makes integrating a shared service into the rest of the environment too much trouble.

        . Most of the smaller agencies are best advised to go with larger agencies and avoid all the hassles of contracts and SLAs. An MOU is much easier than a legal contract.

        wrt Salesforce, the last time I looked, I couldn’t find a government agency (state or federal) that sells things.

        Do you know of an example of a shared service offering that does standardised government things?

        The only one I can think of is the provision of stand alone websites.

        • Well, the thing is we won’t solve this problem with the same thinking that created it in the first place. The way I look at it is that creating the genuinely configurable and flexible multi-tenant applications that are needed to deliver a shared service is too expensive even for cashed up governments … but these systems are already becoming available as public cloud services. Don’t try and build things that you can already buy. Aparent ‘show stoppers’ such as data sovereignty, privacy, security, record keeping etc. are simply design requirements which can be addressed by creative thinking when there is a will to do so.

          You are right, however, that the currently available cloud services are not necessarily a perfect fit to every agency’s requirements and tend only to cover the relatively generic application categories … but … SURPRISE … this exact criticism can be applied to any shared service … the difference is that mature enterprise-grade cloud services actually work today for the services in their catalogue … they do what they say on the label.

          We need to change our thinking about this to adopt a strategic rather than a tactical perspective. The fact is that the shared services story in government is troubled, and we have been trying to solve it for decades … with no sign of generalisable success.It can work in particular circumstances, but is a high risk, high cost strategy … and the forecast is that it will remain a high risk high cost strategy because the organisational dynamics are pretty much unsolvable. Cloud services, on the other hand, are becoming steadily better, faster, less expensive and more trustworthy for an expanding range of functionality. This simple observation means that we need to start reorienting our thinking to embrace the logic of cloud services and start using them where they are suitable and appropriate in order to gain skills and experience with the new model.

          Shared services as a mandated socialist economy is dead. Shared services as a “let’s hope for a miracle so that we can make a silk purse out of a sow’s ear” is dead. Shared services as an easy low risk way to cut costs is dead.Shared services as a way to force agencies to use/adopt common processes is dead.

          Shared services as a way to drive innovation by providing agencies with better and more responsive services at a lower cost is very much alive. If an internal shared service can do this then great. If not, then they need to stop standing in the way …

        • With regard to… “governments don’t sell things” … and therefore commercial CRM offerings are not relevant. This is too simplistic and narrow a perspective and needs to be challenged.

          Many elements of government policy and services involve engaging citizens and businesses in interactions to provide advice, information and services and then record the interactions over time. This style of activity is well supported by many different SaaS CRM platforms … and anyway the more mature SaaS platforms have quite flexible and configurable functionality which can actually be applied to a wide range of applications.

          The Victorian Department of Business and Innovation uses Salesforce as a PaaS solution to support the management of its interactions with citizens and businesses for industry development policy programs and also as a grants administration solution. These are mission critical government processes which have specific requirements for data security and privacy, record keeping etc. DBI has done full risk management audits, privacy impact assessments etc. and has found that the SaaS model can be an appropraite and cost effective way forward – better than more traditional custom developed and on-premise solutions.

          It is all about thinking creatively and working the problem with your eyes open to both the opportunities and the risks.

  11. Government agencies have many business requirements and constraints within which their information systems must exist.

    Where do you place innovation and responsiveness in the hierarchy of priorities for government?

    Compared for instance with legislation, stability, equity, probity?

    One other comment. Shared services, on their own, may be responsive, but when taken with all the other things associated with an agency’s IT systems, the added complexity of using an external service can result in an overall reduction in responsiveness.

    Shared services are a great solution – to some problems. Governments need to make sure they have the sort of problem that shared services can address, without increasing the size and nature of the problem.

  12. I did a search on Wikipedia on what a shared service is. “Shared services refer to the provision of a service by one part of an organization or group where that service had previously been found in more than one part of the organization or group. Thus the funding and resourcing of the service is shared and the providing department effectively becomes an internal service provider. The key is the idea of ‘sharing’ within an organization or group”.
    From what I can tell despite Greg Pearce’s has great distain for shared services yet he’s about to plunge all NSW government IT into one of the largest shared service contracts in Australian history. I can’t really see any difference shared services and outsourcing to one company, the only difference is that it will be delivering services via the cloud, it will be under the control of one company and that company will have the same problems of juggling the sometimes very different needs and requests from the various agencies.
    I work for Businesslink, I know the men and women here work hard to try to deliver the best service they can, I believe if the audit of Businesslink had taken place two years ago the result would have been a lot more positive.
    So what happened two years ago that has damaged our reputation so? Well we were directed to implement Citrix Xenapps to most of our agencies as a cost cutting exercise, this implementation was called MVO (My Virtual Office) and it was meant to deliver applications cheaply and reliably to client’s computers.
    One the surface this would seem to be a good idea, (much like cloud computing sounds like a good idea) and Citrix Xenapps is a good product. The problem was the budget for the implantation was limited and it was rushed through despite the problems it was causing the agencies. Compounding this issue it happened during a major restructure which has caused a major brain drain from Businesslink IT, it’s almost feels like we have been setup to fail.
    One of the issues we face with products like Xenapps and I assume clients will have similar issues with cloud computing is it adopts a one size fits all modality, this is problematic when there are so many different needs in the various agencies. For example, despite what Citrix will tell you about a seamless desktop and the compatibility of applications the truth is that many specialist applications such as Zoomtext for example which is software for the visually impaired, or even something as basic as using locally installed scanners (which the agencies have hundreds) fail to work as they should using Xenapps and I suspect they will cause similar issues for the cloud.
    Unfortunately the writing is on the wall for Businesslink, I believe our days are numbered but I have serious concerns that cloud computing will be any better and if it proves that this cloud computing solution is failure then it will be a failure of monumental proportions.

    • Hmmm … will cloud services really be any better than other attempts at shared services? It is exactly the right question to ask. The solution, however, comes from moving our thinking beyond Cloud Technology (the building blocks of cloud services) and focusing on the services themselves – which are of course a bundle of people, processes and technology.

      Risks can be fundamentally reduced through a combination of (a) buying services that are proven to already be working … “cloudy is as cloudy does”, and (b) changing expectations and behaviour regarding the need for, and merits of, solutions to be customised to suit governments supposedly “unique” requirements.

      Part (a) is getting easier because of the emergence of cloud services. Part (b) is hard, but will get easier as agencies gain practical hands-on experience with the use of cloud services … and discover that the way to reduce the cost and risk of ICT is to confront the established logic that business requirements must allways drive systems requirements. Sometimes the least cost and least risk way is to use a standardised service with minimal configuration and then iterate forward in a flexible and agile manner … this really is the new way of things.

  13. As much as I like to read about cloud services, I think everyone forgets that governments and its constitute parts, are a part of political mechanism.

    Their model for delivery is closely aligned with the government’s operating models and the policies at the time. It does not have the same razor sharp and measurable focus of a commercial organization to ultimately deliver a profit or larger market share. Government has the fluffy delivery of a satisfied citizens, which is open to the eb and flow of public opinion.

    As such, government share service organizations need to address the governance and processes first for fitting a department into the one size fits all approach of most shared service models. yes modern software is configurable, but to a point. Beyond that the process need to change. Ask anyone with SAP experience. That is where massive shared service programs gets undone.

    I worked on the NSW Corporate Shared Service Reform program and in NSW Health which was mentioned I the report. All I can say is that politics always gets in the way of achieving an outcome.

  14. In response to TY:

    Hmmm … you are right … but what to do? The music has stopped, budget cuts have removed many of the chairs … where does an agency sit?

    My enthusiasm for cloud services as a better way forward for some services in some agencies is simply based on the observation that we actually don’t have a risk-free or financially viable alternative strategy.

    If one had a magic wand one would give each agency the ICT capabilities needed to be fully autonomous and the common sense to collaborate to create interoperable processes and systems so that information could flow across agencies and jurisdictions seamlessly (subject to legal and policy constraints). Sweet!

    You are right that the curious and weird processes of big ‘P’ (government) and little ‘p’ (organisational) politics and power games … with ad hoc short-term decision making … are the root causes of the failure of shared services strategies. The issue is that we have no way of draining this swamp. It is just the reality of the public sector landscape.

    The main reason that enterprise-grade cloud services offer hope in this context is that they are (in a relative sense) on dry land … removed from the swamp. The fact that they are radically externalised but configurable shared services means that they offer the prospect of working well enough to be getting on with while being sufficiently arms length to be sustainable. Agencies just need to see the value of this and start learning how to apply it.

    One thing to consider is that cloud services may well be the best defence an agency could hope for against the next crazy whole-of-government ICT strategy. A cloud service can deliver both economies of scale and innovation agency-by-agency on an opt-in basis. The service already exists … you decide if it adds value or not. Each agency can decide for itself … there is no need any more for the secret police to herd agencies into an unwilling and sullen collective … and then force march them into a shared service that may or may not work … or be sustainable.

    Agency-by-agency cloud service adoption within light-touch policy and standards frameworks feels like a more sustainable, adaptive, agile way forward to me than anything we have tried to date … sure there are issues and risks to sort out … but what is the alternative?

Comments are closed.