back to article Warning flags were raised over GDS farm payments system – yet it still failed

The scrapped Common Agricultural Policy IT system was flagged as being at risk of failing by a government project watchdog last year, due to the “inherent high risk" of the showcase digital project. In March the digital interface of the £154m project to provide CAP payments to farmers was scrapped in an embarrassing U-turn, …

  1. Lee D Silver badge

    So is Government honestly trying to tell me that if we got 150 IT experts together, paid them a million pounds each to work on what was effectively a website with payment system, for a year, that it would be impossible (or even unlikely) to have got a better system?

    If that's NOT what they're saying - then why didn't they just do that? And you'd have £27m left over to actually supply the hardware for the first year or so of operation.

    That's just a ridiculous amount of money.

  2. Richard Jones 1
    WTF?

    Did Anyone Understand the objectives and Required Methods?

    The mess does suggest that there was a limited understanding of where the project was heading, what it had to deliver and the methods to be employed to get there. Mapping is not exactly a new requirement, we have had cartographers for hundreds of years and digital maps are pretty much old hat, if maps were a key requirement how come it was not understood early on and the right skills were not brought in at the start to identify the tin cans littering the path forward.

    Or, just as likely the required skill set was consulted, pointed out the potential issues and were then sidelined as not being team players.

  3. Kubla Cant

    Why do government projects fail?

    I'm genuinely interested to know.

    We all make a habit of deriding the incompetence and failure rate that bedevil givernment IT projects. But you have to wonder why it always happens. Is it failure to analyse the problem domain fully? Poor design skills? Incompetent developers who don't deliver to spec?

    If the answer is inadequate skills, why? The days when public sector work was badly-paid are long gone.

    I used to assume that scale must be a big factor, but that doesn't seem to be a legitimate excuse here. There are 300,000 farms in the UK, so it's big, but not overwhelming. And financial institutions deal with much larger systems without major issue (except RBS).

    The excuse here seems to be "The challenges for a system like rural payments were integrations [around] legacy, third-party supplier...", but that's the case for most systems. It's vanishingly rare to build a system that doesn't have these features.

    1. Lee D Silver badge

      Re: Why do government projects fail?

      Because the company that gets awarded the contract makes a lot of money even if it falls flat on its face and achieves nothing else.

      And the ministers that approve such contracts allow that to continue, with clauses that require payment to those companies even if nothing comes out of it.

      And the reasons for THAT, are that being in a position to choose a random company to allocate £100m to allows you to dine out very well for many months of "negotiations", then give the contract to your brother's next-door-neighbour, who'll somehow work out a way to give you 10% of it back once the fuss has died down.

      You think the Academy programme is about getting better schools? It's not. Every contractor, sponsor, and supplier involved in an Academy will be putting a little something somewhere. Think that "superheads" running these places are independent and have no declared interests? Think again.

      Extend to the NHS where it can cost £16 for a pack of AA batteries and you can only use "procurement-approved" suppliers.

      I speak from experience on the schools and from my girlfriend's experience on the NHS. It's not hard to extrapolate to the larger government organisations at all (military, banking, etc.)

      1. amanfromarse

        Re: Why do government projects fail?

        @Lee D

        Not in this case as it's GDS.

        1. Lee D Silver badge

          Re: Why do government projects fail?

          @amanfromarse:

          I wasn't aware that GDS built their own computers, wrote their own software, etc. rather than called in various vendors to bid for the options provided? Like, say, Kainos mentioned in the article? Not to mention those other "third-party suppliers" who had to co-operate to get the integration working.

          Just because it goes via GDS doesn't make it immune to the basic bidding process.

          This is a system for 300,000 farms, some of which will be jointly owned. At best, that's £500 per end-user. That's a disgrace, and they may as well just process them all manually and pay a guy £500 to do a week's paperwork for each of them (not counting what it costs the farmers to fill this out, hire advice, etc.).

      2. Anonymous Coward
        Anonymous Coward

        Re: Why do government projects fail?

        With regards to the comment about £16 for a pack of AA batteries, I work in NHS Finance and I was curious about how true this was (I know Procurement do a lot of good work although we do occasionally see some howlers). Anyway, I just logged in to the procurement portal to search the price list for AA Batteries, unfortunately the portal is stamped all over with the term 'Commercial in Confidence' so I can't tell you what the actual prices are but I will say that pack size is important (you didn't mention what the pack size was) and that the price per battery looks to be less than what I would expect to pay myself in the supermarket, with a variety of options around brand and pack size.

        Over the time I have worked in the NHS I would have to say that the worst cases I have seen have come from senior people who stamped their feet to get something done outside the normal process, like bypassing procurement to place orders directly with suppliers. That said I also appreciate that sometimes exceptions are needed and that stubbornness on the part of finance or procurement can make things worse.

        Most of us try our best but at the end of the day we're still only human.

    2. chris 17 Silver badge

      Re: Why do government projects fail?

      @ Kubla Cant

      the companies that bid for projects know the real money is made in the indecision of the customer and changing requirements. I've been on both sides of the coin and whilst its not an open discussion to deliberately confuse and cause indecision the opportunity is plainly there for those savy enough to play the game which they see rewarded in their pay, bonus and progression. Its often asked why the useless ones get promoted, its because they extract most cash from the customers.

      The PM's are usually useless with no idea what they are doing and are influenced by those that speak their language and not those with the skills to actually build and or rectify the issues. Running a successful project in their eyes is not always the same as what you, I , the public or their customer would expect.

    3. JimmyPage Silver badge
      Mushroom

      Re: Why do government projects fail?

      One word.

      Politics.

      Where reality and response are rarely seen together, as ideology gets in the way. The classic example of this, writ large, is the Alice-in-Wonderland approach to drugs, whereby the response is far out of proportion to the harm. Something experts (paid by the government) have repeatedly pointed out.

      With regards to IT, it depends what the original (political) premise was.

      It use to be held up as a paradigm of "government" IT gone wrong - the Nimrod fiasco. Rooted in the fact that the [Labour] government of the day had to posture about buying a "British" solution, even though (a) the RAF wanted AWACS and (b) AWACS was readily available, whereas Nimrod was vapourware. Cue a decade of bunfighting (3,000 + annual specification changes - or 10 a day), as the RAF insisted on benchmarking Nimrod against their requirements (which AWACs satisfied) and Nimrod kept failing. In the 1980s (when I studied it for a module in my degree) the £3billion wasted was unheard of. Nowadays £3billion won't buy half a non-working Universal Credit system.

      Oh, and no government project has *ever* failed. Cancelled. Respecified. Replaced. Renewed. Yes. But if you look carefully, no government project has ever failed, or been classed as failed. even GDS will be written up as "unsatisfactory", "unable to meet expectations", "unable to deal with updated requirements", yes, yes, yes. But "failed" ? Never.

    4. R Callan
      Boffin

      Re: Why do government projects fail?

      Because no-one has read the most important works on how the world actually works. They make assumptions that what they have been taught (usually in university by people with no real life experience) is actually true.

      The works in question are:

      1.. Murphy's Law

      2.. The Peter Principle

      3... Parkinson's Law

      4... Systemantics.

      Without knowledge of these basic essentials the people involved are only going to continue wasting other people's money.

  4. Anonymous Coward
    Anonymous Coward

    Hmmm

    There are bits of UK government whose sole reason for existence is to do mapping or to handle payments. Maybe a part of building Government as a platform would be to discover what's already been built. Even if it means using stuff that hasn't been written in Python.

    1. David Moore

      Re: Hmmm

      GDS is (mostly) a Ruby dev club. :-)

    2. Anonymous Coward
      Anonymous Coward

      Re: Hmmm

      Using legacy systems? GDS? Where have you been? If it isn't new and written in Ruby or Python it isn't cool. Get with the cool kids!

  5. Anonymous Coward
    Anonymous Coward

    Gov is sure to fail

    If you've ever worked in the public sector this isn't a surprise.

    Step 1) Underpaid under-qualified mediocre civil servants spend their whole time procuring contracts with suppliers who screw them over for lots of money and don't deliver.

    Step 2) Enter stage right the kids at GDS who think everything is simple and they build a Ruby frontend prototype in 2 weeks that serves the simplest need. GDS put it in Alpha and claim success. Underpaid civil servants are left to trying to make the whole thing work properly and deal with all the difficult stuff the GDS kids ignored because "the user need wasn't strong enough".

    Step 3) Service collapses, users hate it, and GDS say "not our fault, it is the civil service bureaucracy that broke it. Now do you want to try my Alpha payments platform?"

    1. Kubla Cant

      Re: Gov is sure to fail

      Underpaid Overpaid under-qualified mediocre civil servants

      FTFY

    2. launcap Silver badge
      Stop

      Re: Gov is sure to fail

      > If you've ever worked in the public sector this isn't a surprise

      Not entirely true - the body were I work^Wam employed has a good number of IT projects, delivered on time and to cost.

      The reason? We have our own Dev team. While much of our IT work is outsourced, we maintain our own in-house team of IT people that are pretty experienced and have authority to say no to the outsourcer when they are being silly.

      The ones where we do fail are generally where we are solely reliant on our outsourcer to deliver for us.

  6. Swarthy

    Here's a thought

    Maybe it should be made part of the contract that any contract for a payments system can only be paid via the system they are developing.

    I know it's not feasible, owing to numerous laws and business practices, but it's still a nice day-dream.

  7. Eclectic Man Silver badge
    Flame

    Govt procurements and failure

    Don't forget that most large govt. procurements are awarded after competitive bidding, in which several companies lose at considerable expense (I know, I've been on both winning and losing sides). So not all bidders want to obscure things or make them more difficult, at least not until after contract signature.

    One of the main issues in govt IT procurement failure is changing requirements, and politically motivated, but unrealistic requirments. the 'Universal Benefit' being one of the most unrealistic things to do in the time available. Most of the disasters I've personally been involved in were over ambitious projects from a client who really did not know what their existing situation was, where they wanted to be (in anything more detailed than a fluffy cloudy-thought-leadery-wouldn't-it-be-nice-if-it-all worked-like-this way) and wanted the contractors to just make it happen without bothering the client with all those annoying questions like:

    "what, exatly does your system do? What data do you have? What are the access requirements on the data? What applications do you run? Who can use those applications? What is your network topology? What servers, operating systems and other software do you use?" etc. etc.

    None of this is helped by sales teams whose main motivation is to win the business and get their bonus and leave the delivery team to JFDI.

    <and B R E A T H E >

    Sorry, rant over.

  8. Aristotles slow and dimwitted horse

    Awesome quotage...

    "Complexity had been one of the main issues"

    "Getting used to working in an agile way"

    Lol Kainos, just lol.

  9. D Moss Esq

    Quiz

    Who said:

    I go weekly now. I go to the meeting of the Common Agricultural Policy Reform Group. It's the RPA. It's the Rural Payments Agency.

    Why I'm so excited about that is because they've embraced agile completely. They're going with an agile build out of a whole new programme. That's going to affect everyone in this country, and how they deal with land management, all the farmers, all the people who deal with crops, all the data. It's going to create, I think, a data industry around some of that data.

    It's going to help us deal with Europe in a different way, and quite rightly we're building it as a platform. It's going to be another example of government as a platform.

    I'm on the Board, and I'm trying to help them every week, and GDS will be working very closely with them to deliver that.

    Answer

  10. D Moss Esq

    Kainos

    Some of the rural payments problems are put down to the users. Too old, not computer-literate, these farmers, and they live in the countryside, where broadband speeds are low.

    That seems fair.

    Others are laid at the door of Kainos, who provided the graphics software for mapping. And which prize did they win at Digital Leaders 100 yesterday? Industry Digital Leader of the Year.

  11. eurotyke

    Mainly a problem with the public procurement processes that are designed for the benefit of the suppliers in mind rather than making the best use of taxpayers money.

    Very hard to kick-out under performing suppliers, very hard to make penalty payments work, very hard to give work to suppliers that have proven they can do the job.

    Every procurement is a lottery.

    Also somethings just shouldn't be put out to fix priced tender, the requirements are just to weak or evolving. Fixed price on these means paying a premium for the supplier to take on your risk, or suppliers that win knowing full well they can't deliver, but that they will get paid in anycase. Plus they will still win more tenders in they future regardless of how badly they perform...

  12. Cupboard

    It was working!

    It was working! It really was! And it wasn't too bad to use!

    OK so some of the mapping features were a bit odd (you'd sometimes have to draw a line in from the field boundary to go round a pit and then back to the end of the field) and sometimes there were missing options (like "Pond" as a feature), and as more people started using it it got interminably slow, but it was working once.

    Some of the requirements were a bit... rubbish. Like the accuracy requirements meant that you'd have to measure hedges to <1cm accuracy across a reasonably sized field.

    For our farm, we used to have coded field names that fitted in the boxes on the forms, the web forms meant we could put the full field name in (so we did) then it got truncated back to a handful of characters on the paper forms, leaving us with loads of fields that have the same name now.

    We're very glad not to have a BT line for our internet though... it was just far to unreliable. We've got a wireless link with a local provider that's much more reliable, fast and responsive.

POST COMMENT House rules

Not a member of The Register? Create a new account here.

  • Enter your comment

  • Add an icon

Anonymous cowards cannot choose their icon