The future of the mainframe

September 28, 2009 Leave a comment

For many years, mainframes have been typecast as the dinosaurs of the IT industry but the mainframe market is again showing clear signs of growth.

In their rebadged configuration as enterprise servers, mainframes are likely to remain an important consideration for meeting ongoing central computing needs. The battle lines are now drawn with other enterprise server suppliers, for a new round of competition in the central computing market.

Where do you see the mainframe heading?

Advertisements

Invest to save, or save to invest?

September 14, 2009 Leave a comment

This week, the QLD Government became the latest to announce a new state ICT strategic plan, Toward Q2 through ICT.  It is focused towards a ‘one government’ approach to service delivery and some investment will be required to support this more integrated service delivery model.

There’s a sting in the tail, of course: a review process will be developed and implemented with QLD Government agencies to deliver a 15% reduction in “the per unit cost of BAU expenditure”, to be achieved by 2013.

Two contrasting strategies are in play:  the first, initiated by the Gershon Review and followed by at least two states so far, is based on a “save to invest” model, whereby savings achieved in ‘business-as-usual” expenditure will be partly reinvested; the second, illustrated by Defence, is an “invest to save” model – funds are provided up-front for “efficiency projects” to generate larger, more significant savings downstream.

The jury is still out as to which is the right approach.  Certainly, Sir Peter Gershon expressed serious concerns about these key areas, having found that of 193 completed ICT projects he reviewed, only 10 (or 5%) “reported actual measurement of benefits and compared anticipated benefits with the actual benefits realised” (Gershon Report, p18).  He recommended much stronger program and project governance including planning for benefits realisation.

One year on, it would be interesting to hear a progress report on this recommendation.  The silence in respect of this aspect, from QLD with its new strategy, in NSW, where the focus seems to be on costs and Defence as it invests to save, is a serious concern.

Why has the NSW GCIO not been ‘MOGed’?

August 24, 2009 Leave a comment

The Machinery of Government (MoG) changes that created NSW’s 13 Super Departments provided a golden opportunity for the NSW Government Chief Information Office to move to the central agency that is responsible for not only controlling the NSW budget but is taking the magnifying lens to ICT expenditure across NSW government.

But to the puzzlement of many, it didn’t happen.

In Canberra, the equivalent office, the Australian Government Information Management Office (AGIMO) sits in the key policy agency of Finance and Deregulation and was moved there after its predecessor (NOIE) spent some years in the wilderness as part of a less influential Department.  As a result of that move, and a Minister, (Lindsay Tanner) who really ‘gets ICT’, it now has real clout.

The failure to be ‘MOGed’ into Treasury is even more curious in the light of rumours suggesting that the Budget-announced NSW Government ICT Review will be run from Treasury.  Speculation is that this review will have similar objectives to the review conducted by Sir Peter Gershon for the Australian Government (coordinated by AGIMO).

Seems that GCIO has been mugged, not ‘MOG’ed in this instance!  Was the new Department in which the GCIO sits given the name of Services, Technology and Administration as a salve for the GCIO staying on the periphery, rather than in the centre?

And what does it mean, to have ‘Technology’ in the title, anyway?

What is the power of 1?

March 23, 2009 Leave a comment

I’ve always been drawn to try to understand why those good ideas and concepts, the logic of which seems impeccable, never seem to cut it once they are put into practice. By cut it, I mean they never deliver the benefits and outcomes they promise.

It is fair to say that more than a few of these exist in the public sector. They don’t fail as such, but they tend not to deliver the same types, or quantum, of benefits as we see in commercial examples. Key factors, obviously, are the different incentive structure combined with different risk tolerances, the absence of profit and loss as a key metric, and governance that’s ultimately linked to the electoral cycle.

A classic example seems to be shared services. These score well in the private sector as a means of driving down costs. But in the public sector, as set out in the Gershon Report, the results are mixed. Let’s leave them for another day.

Consolidation is a key area where economies of scale can be exploited, and costs lowered by reducing duplication of systems, services, facilities and people. This is one of the areas that seems ripe for action in the Federal sphere, with the two largest DHS agencies – Centrelink and Medicare – the lowest hanging fruit. It’s no secret this is under consideration, with officials within DHS and the two agencies offering varying views of the future and a number of internal studies underway looking at the feasibility of consolidating systems, offices and so-on. Recently, co-location of some offices, housing both Centrelink and Medicare staff, has been announced.

The scale of the two organisations needs to be understood. Centrelink has annual cash operating costs of around $2,700 million; Medicare is substantially smaller, at around $700 million. So, if a consolidation strategy was developed to harvest, say, 10% in cost savings (a not insubstantial $340 million), there are immediate implications in deciding where these are borne. In the current economic environment, when these benefits are realised is also crucial: any approach which entails the Government with a significant upfront cost, with benefits some years away, may be unattractive.

There are also significant differences in productivity and efficiency metrics associated with the two agencies. Using highly aggregated (and thus quite crude) measures, Centrelink employees handled around $2.7 million in benefits per headcount in 2007-08, while Medicare employees handled around $5.6 million; Centrelink’s operating costs per $ of benefit payment are around 0.0038%, while Medicare is lower at 0.0021%.

Nonetheless, there is a strong case for consolidation:

  • It would create the best possible base for a single view of citizens, especially in terms of health and welfare needs. This would enable more efficient, targeted programs including specialist case management, resulting in more effective outcomes.
  • There is considerable duplication – of facilities, services, and locations – right through to management.
  • The consolidation process can be executed step-by-step and is, in reality, probably about as complex as some of the recent bank mergers (Westpac-St George comes to mind). Sometimes this complexity is overstated.
  • There is little to justify the continuation of two organisations other than history. And even that, because it is short, is not compelling in the scheme of things

Impeccable logic indeed! But before we get ahead or ourselves (and there are many people who believe that sometimes DHS does get ahead of itself), what are the barriers and issues to be overcome? There are many; most can be overcome with leadership, focus and determination; and none of them are really technical (and please feel free to add more, or comment on those listed):

  • Both Centrelink and Medicare are icons. Moreover, they are Labor icons, tangible of a demonstrated commitment to delivery of welfare services that now extends to disaster support and relief, and a universal health insurance system that underwrites access to health services for all Australians. Change Management 101 highlights the risks of tampering with icons, especially where widely perceived as delivering successfully.
  • Medicare, in particular, is part of Labor folklore. The original Medibank was the centrepiece of Gough Whitlam’s 1972 “It’s time” campaign and despite some sabre rattling, even successive Liberal/National Party Governments stepped back from dismantling it.
  • Medicare offices handle cash; Centrelink offices serve among many other clients, drug addicts. These two “facts” are spoken, never written and the obvious correlation never mentioned. These represent strong perception barriers – even to the extent that the Minister made it clear in the recent co-location announcement that “the new Medicare Australia office will give customers access to a range of non-cash related services…” Surely, in 2009, we can do better than this!
  • Both agencies deliver services that are central to the Government’s health and welfare programs. Any change must have a minimal – preferably zero – impact on the receipt of these by citizens. This is probably the biggest challenge facing the Government in any consolidation. In the political game, ‘getting it wrong’, with high profile cases of missed or erroneous benefits payments could easily result in electoral decimation.
  • There is a view that two agencies with capability to ‘compete’ for the delivery of government services creates a degree of contention that makes both more efficient than would be the case with a single agency. Data is needed to support this.
  • An inevitable outcome of consolidation, if it is implemented successfully, is a reduction in workforce numbers. The Government is unlikely to have much appetite for this – especially since an office rationalisation will affect suburbs and regional centres – during a recession, and around 18 months or less from the next Election

A consolidation of Centrelink and Medicare is hardly a zero sum game, but circumstances could easily make it such.

Who is leading Gershon’s cultural change program?

March 16, 2009 Leave a comment

Unless someone takes the initiative to drive the “major program of both administrative reform of, and cultural change from, a status quo [of agency autonomy]” – and does so quickly – the chance of achieving the $1bn savings in data centres identified are little more than a snowball’s in hell.
At the simplest level, successful transformation is achieved in four steps: making it essential, making it ready, making it happen and making it stick.  Gershon, Tanner and Rudd have made data centre consolidation essential, but we seem to be moving to implementation (making it happen) before we’re ready.

My observation is that most failed transformations (be it an ERP implementation, implementation of a shared services organisation structure, a merger, or “just” the consolidation of data centres) result from doing just this.

Throughout his report, Gershon hammered on about the need for cultural and structural change. The Department of Finance and Deregulation should not even think of starting to plan data centre consolidation unless steps are first taken to create the type of culture that enabled IBM to consolidate its strategic worldwide data centres from 155 to seven (and using 80% less energy).
Who is charged with this?

Those that listen don’t read, and those that read don’t listen

This will be provocative – but then that’s what blogs are for! Nothing like a gross generalisation, unsupported by any facts, to stir debate. So, if you feel so stirred, please comment.

(Oh, and I just realised that if I am right, I am probably missing half the audience I was hoping to attract and – worse still – have alienated the other half!)

So, this is what I reckon, based on a fair few years of observation — I should add that this is aimed at those people who sell, and I guess by corollary, those that buy:

  • There are clearly people who listen and respond to what they hear; and
  • There are those who gather information by reading, and respond to what is written.

Now, in making this broad statement, I know there is a well known theory of learning styles known as the VAK model that looks at visual, auditory and kinaesthetic (learning by doing or seeing a physical representation), but I don’t want to divert into that debate.

When it comes to selling to the public sector, despite anyone’s learning style, nothing really exists in the mind of a public service manager until it is written. So, while you might have a good meeting with a decision-maker or key influencer about a particular requirement that you are trying to sell against – the mood is friendly, and “the vibe”, positive – it generally means little in the procurement process unless it can be integrated into that process. That means providing some form of documentation in a format that is useful to the process, in describing the specification or other key attributes. A verbal discussion rarely does this, even if the recipient is an avid listener and note-taker.

Perhaps this is why many interstate sales managers ask “why don’t you just have a game of golf and make a deal while walking around the course?” The response of the experienced is usually a rolling of the eyes!

The obverse of this is the importance of understanding – that is, comprehending – the written word. Especially in a tender document. This can be challenging, especially if the recipient is one of those people accustomed to taking in what they hear rather than read, and the tender documentation comprises a complex set of files, many of which may have been cut and pasted from previous examples not directly relevant to the current.

And, while looking at the other hand, government officials often complain that suppliers “don’t listen”. What could that mean? Perhaps those who have mastered the art of speaking may not always be well placed to understand what they hear.

The obvious lesson here is to ensure you have a team of people, with a mix of skills around speaking, listening, reading and writing when working in a complex sales environment, such as the public sector. Equally, we must all recognise our own strengths and corresponding weaknesses in dealing in that environment.

Unfortunately, we all seem to forget this!

Is “whole-of-government” a myth?

February 22, 2009 3 comments

There are a large number of reviews, reports, strategy documents, ministerial speeches, consultancy reports… that talk expansively about the importance of, say, the Federal Government taking a “whole-of-government” approach to service delivery, administration, procurement and so-on. Alternative terminology refers to “cross-cutting” initiatives and, most recently in the Gershon Report, “pan-government” activities.

These are among those terms of “managerialese” that sound great – indeed, “appropriate” – until you start thinking about them. What do they really mean? What does a real “whole-of-government” initiative look like? And the big one: how might such an initiative be implemented? In reality, there are significant barriers.

First, there are the structural issues. Government agencies are established quite purposefully as “silos”; they are based upon what are known as “Administrative Arrangements”, which are, to simplify, the responsibilities arising from certain pieces of legislation associated with implementing the policies of the government of the day. These, in turn, are derived from the Constitution. So, for the most part, a government agency exists to do what it is required to do by law, and not to do anything that it is not empowered to do. While this might sound narrow — and possibly even bloody – minded, there is an extensive array of administrative law cases castigating ministers, agencies and officials for doing more that they’re permitted to do. From the perspective of public servants, they are constantly reminded their role is to serve the Minister of the day. This usually always excludes serving the Minister of the day in another portfolio.

Secondly are the funding issues. Appropriations of money are made to support the activities of agencies in implementing the policies of the government of the day. Indeed, it is quite difficult (probably impossible in the bureaucratic scheme of things) to appropriate money to do something that is not directly related to those activities. At the outset, this requires gaining approval in the Budget context – not for the faint-hearted – and subsequently there are the necessary checks and balances such as Parliamentary scrutiny and the Auditor-General to ensure money is spent against the purpose for which it was appropriated.

Thirdly are the incentive issues. It is almost a corollary of the “silos” created of agencies by the Administrative Arrangements that accountability and responsibility of public servants (in serving the Minister of the day) is focused internally on the achievement of the directly associated “key results areas”. There is no effective mechanism to consistently measure (and hence manage) cross-functional or externally-facing activities that are not closely related to the activities of the agency, based on its administrative mandate.

Perhaps the answer is through specific legislation. This has occurred in relation to law enforcement and intelligence agencies, but neither can be cited as exemplars of the process. It seems the experience has been enough to cool the ardour of legislators.

There is the model of empowering a specific agency to take responsibility for a “whole-of-government” activity. Typically, these start with unbounded enthusiasm, which quickly wains once the persistent inertia created by “coalitions of the unwilling” are encountered, wave upon wave. These being, again, those siloed agencies of focused responsibility. Sometimes resistance is overcome by funding incentives, but history shows the inertia is nothing if not resilient.

In the ICT space, the delivery of common shared services is often held out as having impeccable logic, the incentive of savings for each participant, combined with those magic words “efficiency and effectiveness”. Surely, collaboration will deliver a “win-win”? Alas, the reality appears far from the theory. It seems the problem arises just trying to agree on what might be “common”, “shared” and a “service”. When mandated, the outcomes can even be worse that the problem that was set out to be solved.

There has to be a better way. There’s clearly a need to do things better across government agencies and across governments. That means more thought must be put into the “what” and “how”, especially given the substantial barriers to success. But glib exhortations to take a “whole of government” approach must be rejected for what they are: obfuscatory strategies to maintain the status quo!