IBM says it “successfully delivered”
Qld Health payroll

41

news Global technology giant IBM has written to the new LNP Queensland Government claiming it “successfully delivered” against milestones agreed with the previous Labor administration with respect to the disastrous payroll systems overhaul at Queensland Health, which has already cost the state $417 million and will need another $837 million to fix over the next five years.

The project was first kicked off in late 2007, when Queensland Health determined there was a need to look at a new payroll platform to replace the previous platform, based on Lattice and ESP software, which had been progressively implemented from 1996. Partially as a result of the fact that the state had decided to standardise on SAP’s ECC5 and Infor’s Workbrain software across its whole of government operations, those same platforms were picked for the Queensland Health implementation.

However, the project, implemented by prime contractor IBM, Queensland Health itself and government shared services provider Corptech — quickly went off the rails as poor governance and the complexity of Queensland Health’s award system kicked in, with the result that many of Queensland Health’s 85,000 workers have gone without pay, or were overpaid, at various periods from early 2010, when the system went live. The issue was raised repeatedly during the recent Queensland State Government election, which the LNP won in a landslide, virtually obliterating Labor in the polls.

An Auditor-General’s report filed in 2010 found that all concerned in the implementation — prime contractor IBM, Queensland Health and CorpTech — significantly underestimated the necessary scope of the project. IBM had initially told the Government that a “relatively small” amount of functionality would be required to implement a similar new payroll system at Queenland Health as a previous build at Queensland Housing — despite the fact that the Housing rollout only catered for 1200-1300 staff, compared with Health’s then-78,000. IBM had initially prepared a statement of scope for the replacement of the project, providing an estimate of $6.13 million to replace the previous LATTICE system.

In a new audit into the project commissioned by the new LNP State Government in Queensland and published in June, consulting and auditing firm KPMG estimated that it will cost a further $836.9 million for the project to be overhauled over the next five years, adding to a total of $416.6 million which has already been spent on it since 2010. Just over $1 billion of that will relate to funds needed to keep the payroll system functioning. Currently, the platform requires just over 1,000 staff performing some 200,000 manual processes on an average of 92,000 forms to make sure Queensland Health’s staff are paid the approximately $250 million they are owed each fortnight.

Another $245.5 million will go into fixing the key issues which the system still has, and undertaking a systems analysis to determine the long-term future of the platform. And KPMG also noted that the total of $1.2 billion in funding did not include any costs associated with a potential reimplementation or upgrade of the system, any contingencies associated with the implementation of fixed, or any additional fringe benefit tax costs which could arise from waiving overpayments, rather than recovering them.

This week, new LNP Health Minister Lawrence Springborg tabled a letter which IBM had sent him on 5 September this year. The letter is not available publicly, but Springborg quoted from it extensively in parliament. It states:

“As Minister for Health we appreciate your interest in the Shared Service Solutions Program which IBM entered into with the Queensland Government in 2007. In particular we understand your interest in the difficulties that CorpTech and Queensland Health experienced in administering the project.”

“As you may be aware, IBM successfully delivered against milestones agreed with the Queensland and concluded the implementation of the project on agreed terms. IBM consistently delivered beyond the scope of the contract to assist Queensland Health to identify and address concerns with its payroll process. We delivered within the governance structure established by Queensland Health and CorpTech and outlined in the Auditor-General’s report.” IBM also issued a similar statement to Delimiter this morning in response to the issue.

Springborg noted that the letter went on to indicate that IBM would like to meet with him to discuss the issues, but the Minister had decided not to accept a meeting at this particular stage, as he did not yet have all the information available about the process – for example, he wasn’t sure whether the previous Labor Qld Government had indemnified IBM, or to what degree issues of “contributory negligence” could be pursued.

With respect to the milestones which IBM said it had delivered against, Springborg told the Parliament that he imagined “they must have been pretty ripper milestones”. “IBM is actually indicating to us that it consistently delivered against the scope of the contract,” he asid. “What sort of a dodgy contract allows something to blow out by 20,000 per cent? I would have thought there would be a whole lot of benchmarks and processes along the way that were not actually met.”

LNP parliamentarian Stephen Davies told the parliament that from his perspective, having been involved in the corporate sector before his career in politics, it was “extraordinary – in fact, it is beyond the pale” that it could be said that the disastrous payroll implementation could be said to have achieved its milestones as agreed. “It is just bananas. It still does not even pay people,” he said.

“IBM itself is now saying that they met all of their benchmarks,” Springborg added. “If you signed a building contract like that, if you signed an insurance contract like that, if you signed any other contract like that, would you be happy if you had such dodgy benchmarks, such dodgy wording in your contract, whatever it may have been?”

Davies said IBM and other relevant contractors should provide evidence to the Parliament, along with the former Labor ministers for Health and Public Works, about the disastrous payroll systems implementation. Currently, there exists certain legal advice provided to the previous Labor administration in Queensland which the current LNP administration is not privy to. Springborg himself is able to view the advice, but not takes notes on or copies of the documents, or communicate regarding them to a third party. It is believed that the current LNP administration is still considering legal action against contractors involved in the Queensland Health implementation, such as IBM.

opinion/analysis
You really have to ask yourself what IBM was thinking here in making this statement. The Queensland Health payroll systems overhaul has been one of the most disastrous IT projects in Australia’s history, being eclipsed in the magnitude of the issues it caused likely only by the Australian Customs Service’s Integrated Cargo System overhaul, which went so badly wrong in 2005 that cargo started piling up on wharves around the nation during Christmas that year, with the platform unable to process imported goods in a timely manner.

Why, for any reason, would IBM issue a letter to a hostile politician, noting that it had “successfully delivered against milestones agreed with the Queensland and concluded the implementation of the project on agreed terms”? This, to me, just seems like IBM is asking for trouble. Better to have said nothing – or at least, to have said something much less inflammatory.

Of course, it is definitely possible to make the argument that IBM is accurate in these claims and that it is making these arguments for precisely the reason of making sure that its legal position is known. I haven’t seen the contract or contracts between IBM and the Queensland Government with respect to the payroll systems overhaul, and in these cases of IT projects gone wrong, it is very often the situation that bad scope definition at the very start of the project contributed heavily to problems down the track. By the strict legal definition of IBM’s contract, it may indeed be the case that the company delivered on all of its commitments to the Queensland Government. We just don’t know at this point.

But in terms of public perception, that doesn’t matter. The fact of the case is that this is one project which truly went off the rails in spectacular fashion, and in the public’s eye, all of the stakeholders involved – regardless of their actual guilt – have a large degree of responsibility to bear. In this context, IBM’s statement in its letter to Springborg comes across to the everyday man as the kind of rank arrogance which multinational corporations are famous for. And don’t forget: It is the Queensland public which will be paying for this huge and costly mistake – paying with their wallets. $1.2 billion is a lot to pay for a new corporate payroll system for one department, even one as large as Queensland Health.

41 COMMENTS

  1. IBM is the master of delivering exactly what a customer has paid for, in this case they paid for a POS and that is what they got!

  2. IBM Delivered exactly what they wanted to deliver the same as any other major Vendo.

    This realy highlights the fact that the Internal staff do not have the knowledge to understnd what they have signed up to/.

    In my previous life as a coinsi\ultant having worked with a majority of these companues and some that no longer exist they all know how to play the. “You have changed the requirements game”. Unfortuneately the internal staff do not play the game as well

  3. If it does not work, they did not successfully deliver!
    I’ve held funds back from ERP projects in the past when they’ve not worked as spec’d.

    IBM ought to be ashamed of themselves. Also, I am certain this is not the only large scale SAP payroll deployed, so why were there so many issues with this?

    • Customer requirements are the difference between a system that works properly and one that doesn’t – if the customer gets their requirements wrong then they’ll just throw away their money, it’s that simple.

      ERP is very hard to get right and the customer knowing their stuff makes all the difference – in this case they clearly didnt!

      I’ve seen 3 million dollar (small ones) ERP’s thrown away because the customer didnt have a clear understanding of their own requirements!

      • That is true to a point, but firms like IBM are hired because of their expertise. Surely they would be able to advise a client if what they are proposing and paying for will work or not and have a responsibility to do so. The onus cannot always be on the punter to know what the result will be. We all know that customers (including myself) can be idiots sometimes and it is the place of the IBM’s & SAP’s to tell us our expectations cannot be met. To blindly build something and keep billing for it when you knows it’s going to fail is not something you would expect from these guys, as after-all the stake they have in play is their reputation. So much so, that I excluded SAP from my ERP product specifically because of the Qld Health fiasco.

        • If the customer doesn’t fully understand their own requirements how then can they communicate those requirements to a BA?

          I seem plenty of instances where customers thought they knew their requirements and when they got the system to UAT even signed off to go live and then found yet had missed major items or misunderstood major items in their requirements.

          It’s more common than you’d imagine and IBM are very good at giving you exactly what the contract says and not a cracker more.

      • “Customer requirements are the difference between a system that works properly and one that doesn’t – if the customer gets their requirements wrong then they’ll just throw away their money, it’s that simple.”

        +100

        I have seen this sooooooo many times, especially in government, that it’s just not funny any more.

        • True – but it should also obviously be the consultants responsibility to ensure that this is determined to be the case before the commencement of a multi hundred million dollar contract. Though somehow that seems to be avoided by both parties in any enforceable way. The companies’ sales teams go in with promises of quick and clean solutions and then it’s usually technically impossible to deliver upon a mish-mash of conflicting requirements.
          Add to this the consulting companies can afford much better contract lawyers than government uses and you’ve got a completely asymmetric and dysfunctional relationship that will always end badly for those footing the bills. Us, the tax payers.

          • Sorry but the complexity of ERP’s is such that unless the client has a clear understanding of their own requirements and clear goals then this is what you get.

            The worlds best Business Analyst can not make up for a client that doesn’t understand their own requirements and have clear goals for their ERP – this is reality!

          • Finance systems are easily as complex yet when handled in-house these issues don’t appear. This is actually more of a symptom of out-sourcing and it’s glossed over costs and dangers in large, complex government organisations.

          • Which universe do you reside in?

            In-house verses Vendor/ External really has almost zero influence as long as communication lines are maintained.

            It’s understanding a project, understanding the stated (and often unstated) intent and delivering to that. Things can go just as horribly wrong in-house as external. It’s a lovely myth that it “never happens” in-house – forever perpetuated, it seems.

    • Apparently, it’s the various awards that QH people are paid under.

      Pretty sure I read something a while back about IBM saying the awards themselves were overly complicated and that there were a lot of exceptions (Rate X, unless Y, but Z shouldn’t be paid if X > Y type spaghetti). I’ve heard stories where they (QH) would keep going back and asking for out of scope stuff to be looked at, and anyone who has worked on projects like that know how fast that can add up…

      I fully expect that IBM built exactly what they were asked to, and that the problem actually lies with the folks that asked them to build it that way.

      • The awards issue was a big part of it, it’s true — but there were a variety of other issues as well. From my point of view, it looks like almost literally everything that could go wrong in this project did go wrong at some point.

      • Almost certainly the awards were stupidly complicated… but very hard to believe that the same company that managed to beat a chess grandmaster cannot also handle a few special cases on payroll.

        I believe they even had an existing fully working system to measure up against; so they were not breaking any new ground as far as any algorithms or data capture goes. Please correct me if I’m wrong on that.

  4. “Currently, the platform requires just over 1,000 staff performing some 200,000 manual processes on an average of 92,000 forms to make sure Queensland Health’s staff are paid the approximately $250 million they are owed each fortnight.”
    When I see that level of automation it makes me feel so proud. Let’s hope Newman retains some of the public service, by the time this is completed they can be deployed to work on the health payroll. /sarc

  5. ” IBM had initially prepared a statement of scope for the replacement of the project, providing an estimate of $6.13 million to replace the previous LATTICE system.”

    I’ve been ignoring this issue by and large… but was really a $6Mn contract that’s blown out to $1Bn+?

    If so… where do I sign?!

    :P

    • It was a ~$40 million dollar contract. SAP replaced the Lattice product. WorkBrain front-ended that primarily as an award interpreter to audit and prepare the pay advice for SAP.

      I’ve mentioned on previous threads but just to repeat. SAPs environment ended up being underpowered and its largest issue was being fed rubbish data from WorkBrain. WorkBrain was a traditional award intepreter type system that didn’t do contracts so was converted by CorpTech to try to pretend to do contracts. It was also underpowered for the load.

      QHealth contributed by bringing an award system corrupted into a mess through the previous Union pay negotiations with the government of the time.

      There are other instances in play now on much simpler environments. In terms of numbers at least. I discussed the awards with an implementor for a large internationally successful award interpreter that picked up a contract in an aged care facility in QLD – still under the same awards though. It has been 2 years trying to get their rules sorted and it still fails in known award conditions.

  6. Another crown jewel and lesson learned for all of us in the ERP Implementation business. An old saying is you can never clap with one hand and in this case IBM cannot be blamed solo.

    Instead of spending another 400 million, why not just go to SAP and get the best Payroll resource to come in house and implement the overhaul for less than 10 million.

    Or maybe time to give Workday a call.

  7. As much as I despise IBM as a company, this type of thing will continue to happen all over the country for one simple reason (well not that simple but….). The public service effectively attracts the dregs of the IT industry because of comparitively low pay and very slow recruiting procedures (I got a call from Defence asking me for an interview 7 months after I accepted my first role out of uni). When you have the lower end of the industry doing the requirements analysis, risk management and planning etc you’re not going to get a very good outcome.
    This type of work should be 100% outsourced. Minimal involvement from the public service.

    • “The public service effectively attracts the dregs of the IT industry”

      I’d agree with that at a management level, most of the good public service IT managers end up going commercial, while the ones that can’t cut it there stay.

      I don’t agree with your comment at a staff level though, usually they get good training and while the pay isn’t as good as commercial, the super, Flexible Work Year arrangements and a bunch of other stuff make it a “Lifestyle” issue more than a “More Dollars” decision.

      • Fair point, however, I would think that most IT staff in the public service, regardless of level (management or technologist) choose to work there because they place more importance on lifestyle factors rather than career/financial factors.
        This being the case, anyone who values lifestyle over career or financial security is not going to apply themselves or push themselves professionally to a level that you would find in the private sector.
        I’m generalising here, but private sector promotions are done on performance and/or political factors, where as public service promotions tend to be more based on the length of time you’ve worked in the public service so that NEED to perform is less of a factor in government roles (again there are of course exceptions to this e.g. there are some very talented, hard working and successful public servants).

  8. The Qld Health payroll system first required pay award simplification. This is not an IT issue. However, the Qld Health IT debacle can be traced back to the Qld Government’s delusional Shared Serves Initiative Policy and the compounding effect of abysmal execution of that operational model. How would it be possible to deliver long term savings by entrenching inefficient pay awards in an IT system? The policy and the pay awards were not addressed by the Qld Auditor as he is precluded from such comment.

    The rationale for SSI was pursuit of “economies of scale” achieved by reducing duplication and delivering standardisation etc etc. So, Qld Health offloads a working specialist payroll system provided by a Qld IT company originally called Concept Systems now Talent2 and replaces it with a “I couldn’t care less” multinational giants SAP and IBM. What do reckon is going to happen?

  9. It is common that an organisation will ask for X when what they wanted was Y.

    Unless they communicate and understand how X can become Y, no-one will end up happy. That’s a lot harder than people assume. Always.

    • Good point, but at the end of the day, who carries responsibility for that process?

      IBM is more than just a vendor, they offer consulting, handholding, integration, and all manner of services, so I think we might say they carry at least some of the responsibility. That said, governments are very difficult to deal with, and the Qld under the ALP managed some uniquely bad stuff-ups like management of Wivenhoe Dam, and then needing to restart enquiry into why they couldn’t manage the dam.

  10. This was all such an unnecessary mess and expense, pushed by vendors who will always put their own self interest ahead of the customers. The QLD government had a fully functional, whole of government, totally ‘Shared Services’ payroll management system run by CITEC and its direct predecessors perfectly well for decades. It was called PAPMS – the Payroll and Payments Management System – was written in-house, to suit all those little idiosyncrasies that build up in any large system over time, and worked perfectly well. (Witness, PAPMS is still in use with another state government who had not bought into the “replace everything with SAP” mantra quite as much as QLD gov had).

    Reason for replacement? ‘Old technology’, ‘not cost effective’, ‘hard to find staff with knowledge of how to maintain and upgrade it’, and at a last resort, let’s not use the ‘m’ word in a modern, progressive, IT culture – the mainframe… Yes, at the end of the day, QLD Government threw away a working peice of technology for dubious reasons to replace it with something hundreds of times more expensive that does not even work yet.

    And they try and tell us that mainframes are the expensive dinosaurs of IT, simply unbelievable what some people will swallow given enough sales pressure…

      • “Sorry, I didn’t get past that..”

        That’s pretty much the issue. It’s all well and good developing in-house solutions, as long as the intent is to always maintain. Which leads to the next concern; the issue of long-term support. What happens when the entire brains trust for that solution has left, or retired?

        The answers are never good.

        Solutions are changed because it often comes down to the cost of maintenance. Software is no different in that respect. Try calling Microsoft for support on Windows 3.1. It wasn’t that bad at the time.. still perfectly workable now, right?

        • And this is why truly large scale commercial computing environments support 100% backward compatibility – the IBM mainframe environments that these systems used to run on are still supported, and will remain so. Object code compiled in the 1960’s still runs, unchanged.

          People without experience of these environments have no idea just how much is saved by NOT having to throw away your hardware and software every life cycle of an application – and that’s why these environments are still being used in commercially run places like banks. Only Government can afford to throw a billion dollars at something and not have it work!

    • “Reason for replacement? ‘Old technology’, ‘not cost effective’, ‘hard to find staff with knowledge of how to maintain and upgrade it’, and at a last resort, let’s not use the ‘m’ word in a modern, progressive, IT culture – the mainframe…

      Sorry but I just laughed when I saw that! Who is number in Mainframes ……. IBM, that’s who and their latest generation “System Z” machine is pretty damned spectacular!

      Btw, supporting legacy mainframe apps can be frighteningly expensive, especially seeing as they are usually written in cobol and cobol programmers are worth their weight in Diamonds!

      • I think you misread me – these are the reasons often citec for getting rid of a mainframe. They are usually false, and perception biased. Microsoft have done a fantastic job of convincing IT managers and executives that their offerings are enterprise class, which they patently are not.

        I’m guessing you meant that IBM are number *one* in mainframes… Actually, they are numbers 2, 3, 4 etc also, since the demise of the hitachi plug compatible system z machines, the last of their kind.

        And I think you need to check your facts about the cost of a Cobol programmer – if I could make serious money writing Cobol, I would. I presently teach people Cobol at university part time to help meet this burgeoning need as older IT workers retire – Cobol is childs play to learn compared with Java or Dot.Net.

          • I’d be very surprised (and happy) if I could find a job anywhere in Australia writing Fortran… PL/1 would be more likely, but I haven’t seen a job requiring this language for over a decade…

  11. Exaclty the same way every major bank and financial institution in the world does today. By hiring graduates and training them, offering a career path with security and good conditions – just like they used to 20 years ago before economic rationalism informed us that this was not productive enough.

    It’s debacles like this, and I can name many of them, that make an obvious lie of the entire rationalisation of outsourcing, deskilling and core compentencies that has occured in almost all facets of in-house IT support organisations in the past quarter century.

    Commercial Off the Shelf Software (COTS – heck, they even made an acronym for it!) does not always equate to ‘better’ – and this project serves as another good example of how true this is.

    • You are really expecting Government to be that forward thinking.

      It doesn’t happen. Graduate learns his stuff, then leaves to higher pay elsewhere. Rather than bring someone in asap to learn the process, government waits till 2 weeks after the admin has left before trying to fill the position. Then that person gets thrown in the deep end, becomes really good at what they do through pure trial and error, and then gets a job elsewhere for higher pay. Rinse and Repeat.

      • You’re absolutely right – and that’s the sad thing. I started my IT career in government back in 1985 – and we were well trained very well in a variety of technologies. About half of us left over the next ten years, the other half are still happily running that department – attrition was part of the hiring and training strategy, and it seeded both excellent government IT departments as well as private sector ones with goverment trained specialists.

        And again, it’s this kind of very expensie debacle that should encourage governments to consider being forward thinking – would have saved them (and that’s us, the tax payers) about a billion dollars in this case…

  12. Egos and ambition – both of them buttons that sales reps push skilfully – get involved. I have never heard of a senior manager getting promoted for saying at a job interview “I decided NOT to spend x-million dollars”.

  13. IBM clearly did not do a good job and had no concept of what is required for a payroll project. Qld Health had to pay an additional $2.4m to have postings to financials included in the project and IBM struggled to understand why that was even necessary. However it was CorpTech who were 99.9% to blame for what went wrong. They have a history of failures and struggle to even understand the roles of key stakeholders. They are managed via a bullying mentality and rarely promote competent staff to ensure that the doers keep doing. Most of their senior management have limited background in the use of SAP which is the main software they are responsible for.

Comments are closed.