How many more failed IT projects is the government concealing?

Government and IT systems.  It’s a big hole that you  just keep pouring money into.

A major government IT project is three years late and nearly $30 million over budget.

The first stage of the Joint Border Management System (JBMS) – merging the computer systems of Customs and the Ministry for Primary Industries – was supposed to be finished by the end of 2012 at a cost of $75.9 million.

Secrecy has shrouded the problems which beset the project, but the Herald can reveal the budget has soared to $104.1 million, and the system is now expected to be operating by the end of next year.

Projected savings of $535 million over 10 years will now take 15 years to achieve according to papers released under the Official Information Act.

Crisis talks were held weekly between the chief executives of the agencies and IBM, the documents show, and high-powered legal advice was obtained from Crown Law and the Chapman Tripp firm.

Ministers of relevant departments, the Department of the Prime Minister and Cabinet and the State Services Commission were also briefed on the problems with JBMS.

A court case was avoided after 10 months of negotiations.

10 months of basically no progress, and 10 months of reworking the specs, 10 months of increasing the scope.

10 months of justifying more money.

A briefing paper for the Cabinet said the “parties have now agreed to a mutually acceptable arrangement that will ensure … the JBMS is delivered in the most cost-effective way that manages the risks associated with a sizeable information technology project”.

The specific problems with JBMS and the agreement with IBM remain secret because of heavy redactions in the released documents.

But funding for the first stage of the project was lifted from $75.9 million to $89.7 million last year.

The briefings released to the Herald show a request for another $14.4 million was approved this year because of “funding pressure”.

The delays have been blamed on unexpected complications in integrating new and existing systems, which has taken longer than expected to develop and test.

A Customs spokeswoman said the renegotiated agreement with IBM was “commercial in confidence” and remained secret.

Part of the agreement is that IBM is locked in to provide support and “enhancement” of the programme for seven years.

Labour MP David Shearer said the JBMS project was on its way to becoming another Novopay, but the problems had been kept quiet before the election.

“This is nearly $30 million over budget and at least three years late. This is another Government IT project going off the rails.”

There are two problems with large scale IT projects.  They tend to be under estimated, and once started, the scope of the changes and additions causes a run-away condition where project managers continue to chase their tails.

Next, at some point, the blame needs to be apportioned, and they go to court.

More costs, more delays, more changes.


– Jared Savage, NZ Herald


THANK YOU for being a subscriber. Because of you Whaleoil is going from strength to strength. It is a little known fact that Whaleoil subscribers are better in bed, good looking and highly intelligent. Sometimes all at once! Please Click Here Now to subscribe to an ad-free Whaleoil.

  • Dave_1924

    I seriously think the Government needs to stop hiring the same people and vendors to do its IT projects and programmes. The results are never pretty

    Poorly defined Scope, under estimation, scoop creep and business owners wanting endless change: INCIS, Novopay, JBMS….. similar hymn sheet all instances.

    And what’s going on with the MoJ development of integrated workflow systems to support more efficient courts? Didn’t Chester deliver bad news on that front as well in the last 12 months

  • The Guy

    The program managers do not know how to spec software – and hence to CYA they spec the kitchen sink. Double insurance is to contract with IBM et al, who use software and processes designed last century. One day, somebody will propose a radical solution .. lets find NZ’s best IT talent, particularly architects who are world class, and for every govt solution we put together a 20 – 100+ person team. 50% owned by the team, 50% owned by the govt. Implement in half the time and a fraction of the cost. Then sell the stuff overseas and NZ benefits.

    • exactchange

      Rod Drury proposed a not very radical PPP for the new IRD megasystem. Last I heard IRD said they would think about it but had privacy issues. Fair enough. But surely that is just a problem to be solved, not a show stopper.

  • exactchange

    Two main problems as I see it, both primarily internal governance responsibilities. The suppliers will be blamed, as it’s easy though probably unjust.

    1. Managing project and programme scope. Basically a risk issue.
    2. Maintaining people continuity, as much as possible. Suppliers will try to get away with people changes and have to be smacked down. Internally it doesn’t help if programme and project staff keep changing. Sometimes that’s because just quietly not everything is going perfectly. For the replacement, it’s not easy to adopt someone else’s baby, but way too easy to change direction. Again a risk issue.

  • NotLen

    Got a legal problem. – I know call Chapman Tripp. That will solve it.

    Put Tom and Jerry onto it.

  • Slijmbal

    IT overruns are de rigeur for large bureaucratic projects. Admittedly, government departments are bad and local government even worse at it than corporates but it is a almost a pre-determined outcome.

    The original price is never real and the so-called Project Manager never has real control. I work in this environment and gave up managing projects as you were never allowed to actually succeed.

    Any project over about 10 or 20 mill should never be allowed as it will always fail.

    • symgardiner

      And there is the issue. Go over $10m and it will fail in a public sector context.

  • The Whinging Pom

    After working in IT for 30 years, managing multi-million dollar multinational projects, running departments for major NZ companies, all with a huge success rate, if I apply for public sector jobs I get told that they won’t look at me because I don’t have public sector experience!

    Basically I haven’t spent my 30 years in IT working in the culture of failure that seems to pervade public sector IT projects in both the UK and NZ (and I daresay elsewhere as well).

    • burns_well_eh

      Bingo bongo – you could try asking them “How has employing people only experienced in the public sector to run your big IT projects worked out for you so far?”

      Then watch the look on the face and cue the sound of crickets….

  • Dave

    However, before the project is even started, the scope needs to be thrown away, and consultants brought in to look at what is actually needed, is the internal process and practices world class, or is a junion architect scoping the design around outdated practices for starters. All too oftern the users need to change, not necessarily the system.

    • burns_well_eh

      The worst possible thing you can do is to add more consultants to the mix.

  • The question is also how many more are they already in the process of planning? A lot of the projects and programmes simply aren’t designed or thought through properly in the first place. You get projects to scope projects which have already been scoped several times before. They don’t want to solve actual problems, but instead have scope creep based on fantasies or ill-thought through ideas. This perpetual waste is linked to managers who aren’t able (either intellectually or due to higher interference) to actually start / implement projects. IT is complex enough as it is, but it’s certainly not helped by how it’s being executed.

  • Cambo

    One of the real problems is thinking a “NEW” system needs to be built when in fact dropping in a code free integration platform could be done for under 5 million max.
    The requirement was data needed to be shared. A new system did not need to be built.

    • jcpry

      Or did it? Was the platform on which is was running at the end of life? Was the development technology no longer supported? It is easy to say that a new system was not needed but there are often underlying factors that make it necessary.

      • Cambo

        Quite possibly some of their legacy systems were and are dogs. But that’s the point of integration isn’t it. To extend the benefits of your systems including legacy and to gather and disseminate all the data so it becomes shared information. The Philip Smith episode is a case in point.
        Field Name Surname
        Known by any other name? YES NO

        If YES – First Name Surname…

        Share the information

        Integration at the very least would allow old systems to be bedded out slowly whilst a newer product was integrated. An agile approach is a no brainer IMO

        • jcpry

          Not necessarily. I’ve worked on projects where the cost of integration is very high which makes the overall cost unsustainable.
          Plus you are beholden to the writer of the integration which can be a place close to IT he’ll.

          • Cambo

            Totally agreed on that point! Which is why we keep away from anything to do with hard coded integration. asking for trouble

  • Cadae

    One of the main problems is the accelerated RFP process that government organisations currently impose on vendors. Vendors are given too little time to accurately assess and price proposals.
    Integration projects are seldom only about IT – they are business process integration exercises which can have unforeseen repercussions on business operations, leading to scope creep and cost increases.

  • Nebman

    After seeing this sort of stuff happen over and over for 20 + years, they usually have two common denominators at the core of the problem;

    Namely at IT Manager woefully out of his/her depth but with visions of grandeur well beyond their ability and an exceptionally good software sales person.

    Neither of them will have project management implementation experience and they’ll both be gone before the end of the project.

    NZDF have exactly the same problem with their toys as well.

    • colinrippey

      Bosses buying things off the glossy brochures and mission creep were the main causes of this when was a worker bee in this area.

    • burns_well_eh

      Another problem is having an IT Manager anywhere near projects such as these. They should be run by the Business Units with an experienced Implementation Project Manager with experience in the real world and a mandate to get decisions made early and to hold people accountable for what they say they’ll do.

  • Disinfectant

    Hire the people who integrated the ANZ and National bank systems.

  • Fredd Dagg

    Unfortunately anyone who quotes realistically will not get the job, as someone else will always tell ‘The Govt’ what it wants to hear.

  • burns_well_eh

    While you’re at it, you could make some enquiries about MFAT’s 5 (yes, that’s FIVE) failed attempts to select, purchase and implement a property and lease management solution over the last 7 or 8 years. Colossal waste of time, money, effort and our tax dollars. Who knows if and when the current attempt will produce something worthwhile – all the signs are that it won’t.