How is it that all these ICT projects of the Dutch government fail?

They don’t all fail, but if it happens then it falls on… I think there are a few reasons for the failure of projects.

  • Lack of Knowledge ABOUT it in general with some government services (while other government services are much more-or exclusively-FOCUSED on it and you see fewer unsuccessful projects there).
  • This is linked to the previous point: great dependence on some major suppliers such as logic, Capgemini etc. These companies have a lot of experience in drafting tenders for public projects.
    They understand exactly how the game works.

Does the project show that something has been forgotten somewhere, for example, because the composing official is not an IT-er with 40 years of experience?i.That is a change point and falls outside the tariff agreement.

  • Poor project management on the part of the government.
  • Although one works a lot in projects within the government, managing IT projects requires some other skills that are not always well-attended. The government is doing a catching up here.

  • Weird demands from the government.
  • Sometimes tenders are written on a specific supplier.Or there are functional requirements in that side still hitting shore, resulting in huge cost increases.

  • And now suppose that such a supplier is working?
  • Who controls the work?
    It is not uncommon for consultants who work directly or indirectly for the same company.

    In some public administrations, half of the employees consist of consultants who simply cooperate with civil servants.These are the public administrations where the least projects fail, but at the same time it is also weird that the quality of the work carried out is controlled by people from the supplier themselves.

  • Difference in ‘ business model ‘ between the supplier and the government. The supplier has the business model ‘hour invoice‘, so it has an immediate interest to stretch the delivery and to use as many people as possible.
  • These hours are limited by the tender, but when you put sharp people on it you can write a lot of hours outside the appointments.

    The government’s business model is ‘ prevent room questions coming ‘, in addition to the desire to get a certain system.

    You can check where that shoe is going to wring, and that is in the early delivery and in limiting the costs.

  • Changing requirements by own legislation. Suppose you have a government service that suddenly has to store all citizens ‘ usage data for 7 years in a system.
  • But the original requirements also suggest that you should be able to restore a backup within 40 minutes.
    And that’s where the new privacy law is about to allow someone to be removed from that backup as well.

    The backup system you had in mind?Replace it with something that costs a few million extra.

  • Lack of consequences
    Do you think that the handful of fixed government suppliers are punished in any way for unsuccessful projects?
  • No… The government is too much dependent on them.
    So the next tender round will win them again.

    Let alone the government itself, a failed project will perhaps result in a small seating dance, but real consequences like people disappearing behind bars because of wasted millions in tax money are not there. First and foremost, many government projects in the ICT field are just going well.But if a project fails then that is extensively explained in the news while one usually is silent about the projects that go well.

    What is also involved is the enormous administration that ICT entails within the government.This is because these projects are paid out of tax money and therefore every dime that is issued carries a complete track of administration. An official who simply charges a 3 euro parking fee to his boss because he had to go somewhere to make a huge mountain administration and eventually even at the highest level someone can ask if those costs are rightly asked by the employer have been paid. Most companies are making such costs less crowded.

    But most companies also have a fairly simple hierarchy.The government does not. At various large companies such as banks, incidentally, too. These large, deep hierarchies are often a problem because every decision must first go up to a point where someone can decide on it before it can be brought back down so that it can be implemented. You then have a chain of communication that slows down everything and makes some decisions just too late.

    It does not help that the government should also always outsource these types of projects to the so cheap possible party while companies can usually ignore the quacks.Fortunately, the government often makes very high demands on these third parties, but that is not a guarantee that everything will go well.

    Discover the ICT of the Tax Office-working at the Tax Office

    DAarbesides, you have to think that e.g. the tax administration has been on digital since 1949 and therefore has 70 years of history IN its ICT archives.In addition, there will be several old projects that have been written very long ago and work fine not to be replaced quickly. All this in combination with lots of backups and redundant systems to cope with failures and with vastly stringent security requirements.

    But also with their vacancies you see that the tax administration (and the entire government) sometimes leaves a little to be desired.See this vacancy: .net developer.Traditionally, they are working with Java but they are looking forward to .NET. Unfortunately again version 2015, which is already somewhat outdated. And then they want a recently graduated HBO’er or WO’er with at least 1 year of experience and therefore experience with .NET… For the Junior position, luckily. For a Medior is 5 years of experience with software development required and HBO level. For a Senior you must also have leadership qualities…

    The problem is that in the meantime, the tax administration is lagging behind in technology for several years, while the better software developers want to continue to follow the latest developments.I do not want to say that there are only servants working at the Tax Office but within the normal business world there is generally much more to earn, if you are good.

    Nevertheless, many ICT projects seem to succeed at the tax office.Similarly to the rest of the government. These projects are relatively expensive, but because of the enormous amount of administration that comes with it. And in addition, the Government is reasonably easy if a project threatens to go over the budget. It quickly pulls out the plug.

    It is unjustifiably to say that “all these ICT projects fail”, because there is quite a lot of good for the government.Because of the size, and because it is about public money, negative results often get the news.

    If you look at the failure of the major projects, you can recognise two important pillars, governance and architecture.

    Governance is an interesting word.What it means is that there is monitoring of the implementation: Do people do the right things that lead to the right results? And adjust if this is not the case.
    If governance is poorly regulated, you often see two phenomena: no one is responsible, and everyone has an opinion (and therefore often also influence).
    In Buddhism There is a story about five different animals that are tied together with a rope.The rabbit wants to hide in a hollow, the bird wants to be in the tree, the stag feels safe in the bushes… And so it pulls all together and the strongest wins (for even). In poor governance, you see a constantly moving force field.

    The result of this is that benevolent people in the execution piece run on a lack of focus.
    And afterwards it is pointed with fingers, it was on this, that, or that, but never to us and certainly not to me.In other words: lack of ownership.

    Also another word.The essence of architecture is that there are reasons why things work as they work, and that these are frameworks.
    There is a law to protect the privacy of people (think of the GDPR) and therefore a system that processes personal data must meet certain requirements.You can also choose to take out software from one vendor instead of multiple vendors from a cost consideration, which prevents, for example, a part of the system integration (how to communicate IT systems with each other) issue or Simplified.
    However, what you often see in practice is that the architecture (the “Why”) is linked to the implementation (the “How”, the system landscape).If this happens the risk is great that it becomes “how” star. Because you are not allowed to change “the architecture” of course.

    The government has an enormously complex system landscape, which means that there are many different systems which are also linked to a greater or lesser extent, and these connections also apply restrictions.Think of a spider Web: If you give a tap against one wire, the whole web moves. And because this landscape, often confused with the architecture, is “star”, it is considered undesirable.

    Governance + Architecture
    Now it becomes really interesting.Because when I give a tap against a thread it moves the whole web. And if you know spiders a little bit, you are aware that they are very quick to see what’s going on. But what if this is a number of spiders that are stuck together with a string and each react to other vibrations?

    This is where it really hurts to do.Not only is the landscape complex, but everyone has an opinion about it. Because if A moves, B also has to move, and with it, C also runs the risk of moving.
    Then you get what is called with a nice word “stakeholder management”, where there are dozens and sometimes even more stakeholders.And they all want to be served, because they are all important.

    Is that all?

    I also see a number of more behavioural components, something we can also call culture.

    I also often see a huge level of ambition, not only with the government but also with other companies that are very dependent on software.
    If we make an analogy here: when you propose to Scania to completely replace the assembly line of the trucks, they are looking weird.
    This happens sometimes, I think this has been done recently, but this only occurs when all new trucks need to be made or if the end of the service life of the assembly line is reached (correct me if I’m wrong).
    Other changes are incremental, minor changes on the existing assembly line leading to improvement.And over time these are very big improvements, but it’s a profit of seconds here and seconds there. Sufficient small profit makes it possible to divide the assembly units differently, so that more work can be done within the same time.

    Not with the government.If we change, we will do it at once (think of the above rigidity), then everything goes on the kick.
    It’s as if Scania suddenly decides to start building boats.

    Making a large-scale change is complex.You cannot simply tinker with part A, as this may have an impact on part B. You can choose to loosen the link between A and B (in technical terms: you put an API layer or a service bus in between), but this takes time and money and is often not Included in the scope of the change. It is also an adaptation to the architecture (not really, but yes… System landscape and Architecture…) And of course that is also undesirable. If you had included in your architecture principles: “when transferring data in the form of transactions, a” handshake “between the two systems is expected to ensure the integrity between the two systems“, then a service bus had a good fit.
    Still it is not an easy issue, but at least you have realized the desired decoupling.

    This seems to be a technical issue, but it is actually an organizational issue.
    The system landscape represents the organization, the way of working.In The systems there is an automated process. And apparently this process is complex, because otherwise all these links were not needed.

    The interesting question is: Should this complexity be?Because if you look at an organization like Scania, they know from experience: complexity is not efficient. It leads to disruptions and unnecessary actions.
    Complexity is often in small things.I use Scania as an example because this is a well-known Lean environment and a good friend of mine is working there.
    He told him that when screwing a part of screw-bityou had to swap, he had to switch back a few steps later before securing a new set of screws.By adjusting the order-its proposal, and that is a change made for Scania worldwide on assembly lines of the same type of trucks-an unnecessary step has been taken out of the process.
    Now you might think: Oh, what does that matter?But Scania has been able to significantly increase the number of trucks made in one day by many of these small changes. It is not an efficiency issue, it is a culture issue. People see what’s better, suggest to improve, and actively deal with it, both within the Assembly group and within the management.

    I know from my own experience that many problems in IT occur from unclear and complex business processes that one tries to solve in the IT rather than in the business process.It’s easier to blame IT if you’re a colleague to talk to or think about how things can get better.
    I generalize here, because I know that many people in the workplace have tried to honor and conscience and are trying to make improvements.And that this often fails. That is behavior, culture, within the organization, and is tricky to change.

    I have now appointed the topics of governance, architecture, ambition and complexity (behaviour, culture).Is this anything?

    If we read the news about the failed ICT projects of the government, then we see the millions flying past.And then we think directly about waste of taxpayers ‘ money.

    I want to put money in a different perspective.
    Do not see money as something you buy your daily groceries with and pay your rent/mortgage with.Take a look at money as a means to make choices.
    The amount of money you have in your bank determines which groceries you can do in the grocery store.It determines whether you are forced to shop at the Aldi, or that you can also go to the Albert Heijn and buy the top brands. The money you have, or want to pay, for your rent/mortgage, limits the choice for your property. Unless you-like me-are willing to live in another region.
    Money = freedom of choice.

    If we look at money in this way, we will see another tension field emerged.Because what says a budget overrun then?
    Somehow I gave money with a certain expectation, a certain result.And if the result is not, what does it mean? As far as I am concerned, only one conclusion is possible: wrong choices have been made.
    And if we put a significant increase in money, what do we do?Only one conclusion possible: we reward wrong choices.

    And then the interesting question is: who made the wrong choice?
    Did I not have the budget (money) to realise my goal?Was my assignment wrong? Has anyone misjudged the amount of work or complexity? Was The estimation okay, but have we started doing other things? Have I accidentally unknowingly ordered other things to go about doing it?
    How could we have avoided this?

    I think it is very good that projects are being shut down and blown off, even if that costs a lot of money.With this we set the tone for the future: we stop rewarding wrong choices.
    This will allow other mechanisms to be played, with people being addressed to their responsibilities (both in implementation and governance).
    And that leads again to other (hopefully better, or at least more thoughtful) choices.
    It is not a simple problem with one cause, there are various facets that all lead to minor problems.And together, just like if you open enough programs on your computer, your computer slowly and eventually unworkable will collapse. Every program in itself works well, it’s all together what causes your computer to crash.
    And then the question: What do you do?Less programs open or buy a new computer? Because that new computer is still needed once.
    This dilemma has the government too: when should you make big changes and how do you do it in a good way, and when can you make things work better by small adjustments?

    If I were to get this issue on my plate I would focus on two topics:
    1.Benefits Realization: What does it deliver on
    2.Risk mitigation: How do I prevent this large-scale explosion

    And these are not easy issues, not in the business world and certainly not in the politically (and not commercially) driven government.

    Because they work with expensive external consultants, who have no internal experience.It is now really time that we are tackling this European and no longer try to invent the wheel. Educate their own people and pay them well and have a contract sign that they cannot monetize their acquired knowledge, at those external companies. Not to use Oracle Consultants or SAP consultants in particular.

    Skilled people with sufficient knowledge are very expensive.Per hour, day or per project.

    The government is very economical.

    So there is selected on price.Hence.

    Almost always thrift is much more expensive than hiring professionals, but explain this to the contracting.

    This is mainly because the government does not have the knowledge “in house”, and therefore enables companies to design and implement it.

    Because there are virtually no consequences for the completion of an IT project and the companies that are engaged per hour, the motivation for these companies is to stretch the delivery as long as possible, instead of being ready as soon as possible.

    And if a project is completed, the more problems are involved, the more they need to come back to fix the eee (= more money for these companies).

    An additional problem is that different government departments always have new input on a project, which means that things need to be adjusted.

    Leave a Reply