UK government tech procurement lacks understanding, says watchdog UK government plans its technology purchases with limited assessment of technical feasibility, according to a spending watchdog's analysis of the £14-billion-a-year procurement of digital services. Among findings in the latest National Audit Office (NAO) report [PDF], which evaluates the government's approach to technology …
We work in the 'is this technology worth it' space and determine cost / benefits, impacts and issues - in great human fashion, no 'leader' wants to hear that their idea or concept has downsides or won't deliver all that was promised. Progress is measured in how much physical stuff has been visibly bought, delivered / installed, not in whether there's been upfront work to understand and de-risk the project. By the time it starts going wrong, the original 'leader' has moved on, usually promoted.
This post has been deleted by its author
It's an example of the general manglement malaise of focussing on outputs instead of outcomes. Understandable, though, as the former are usually a lot easier to measure than the latter, especially within the timescale of a government or business reporting cycle - and few people who make the big decisions are actually competent* to do so.
*Competent in terms of education (understanding the why's), training (having the skills) and experience (being able to put the first two together).
Reading the article the thought occurred to 'tell me something I do not already know'.
I was hoping that they would point out that a degree (even a first class degree form Oxford or Cambridge) in Classics, Mediaeval Philosophy, Ancient History, English Literature, Russian, Philosophy Politics and Economics, Modern History, Geography, Law, or Sociology does not qualify you in understanding computer system procurement. But then, what do I * know?
Same as the people who reject human affected global warming without any scientific qualifications at all, but who are really sure that it is all a plot to destabilise their country's economy, when they could actually go to any university with a physics department and do the experiments themselves and discover that CO2 really does cause warming, and then do the arithmetic of how much fossil fuel has been burned in the last 150 years and what that has done to the atmosphere.
OK, rant over, sorry
*B.Sc., Ph.D. in mathematics, plus over 20 years IT security experience - a no-hoper, really.
Actually, I 'politely' describe myself as "a gentlemen of leisure". Whilst the term "gentleman" is debatable, the "leisure" is not. I am also one of the 'inheritocracy'*, having received a very generous 'donation' from parents who are now sadly demised.
* see e.g., The book of the same name by Eliza Philly, ISBN 978-1-785-90858-3
A politician with the academic credentials and action on climate change is the Claudia Sheinbaum Pardo, the President of Mexico.
https://en.wikipedia.org/wiki/Claudia_Sheinbaum
One to watch/she's going to give as good as she gets from her incoming neighbour to the north.
It is not as simple as this.
Public Sector procurement is tied in with all sorts of other issues where the outcome is the principal point of the ITT is to prevent the organisation being sued by the companies with the unsuccessful responses. Then we have the utter insanity of the tender evaluation where you are not allowed to take you first hand experience of a company and use that to score down (or up... falls of chair laughing)!
This is how we see the same truly awful players in a merry-go-round of fat contracts and failures.
Now add in the the Private Sector will use every trick in the book to rip off the Public Sector as they see it as a bottomless supply of money and you start to see why there are so many issues.
Yes, we know there are problems round scope creep and all sorts of other funky stuff but in many cases these changes are forced down during the implementation by changes in legislation, either directly or indirectly from systems that have to be integrated or by undue delay.
Unbelievable amounts of time is spent gather requirements and getting sign off from stake holders. The result of this is that the requirements are often not fit for purpose.
IT Mangelement don't have much of a clue because they are more interested in covering their own arses than listening to the advice from people who understand.
I have yet to find any Project Managers in Public or Private Sector that truly understand what they are doing. All they are interested in is Gant charts and closing stuff off. I have seen whole chunks of a project closed off to meet timescales knowing that we are still working on delivering said section or waiting for the customer to implement a critical prerequisite.
What's changed in the last 100 years?
In The Tender Ship, Manhattan Project engineer and Virginia Tech professor Arthur Squires used Shute's account of the R100 and R101 as a primary illustration of his thesis that governments are usually incompetent managers of technology projects.
To be fair, the NAO arent chuffed with the situation and having had interactions with them do have suggested pragmatic governance in publically available documents, I thought good observations on the problems and how to address them. But no-one wants to listen to 'can we just have a think about this first'. See NHS IT scandal and the yet to occur AI everywhere nonsense announced recently.
> See NHS IT scandal and the yet to occur AI everywhere nonsense announced recently.
Don't forget the Covid "Test and Trace" app, which cost... <deep breath> ...£37 billion.
The Independent: Scathing report blasts ‘unimaginable’ £37bn cost of coronavirus test and trace system
(For comparison's sake, the cost of building the Panama Canal was estimated at $639 million in 1914. In today's money, that's the equivalent of $20.1 billion or £16.5 billion. The Channel Tunnel cost £9.5 billion in 1994, which is equivalent to £19.5 billion today. Combined cost = £36 billion)
Okay, I used imprecise language.
Yes, the app was a snip, at a mere £35 million. (How did the devs keep food on the table?)
The Test And Trace system was allocated a total budget of £37 billion over two years. But only 25.7 billion had been spent by June 2022, with a lifetime cost of £29.3 billion.
Nevertheless, it was still an astoundingly huge sum of money to spend on a single, short-term IT project.
BBC: Covid-19: NHS Test and Trace 'no clear impact' despite £37bn budget
BMJ: Covid-19: NHS Test and Trace failed despite “eye watering” budget, MPs conclude
A centre of excellence used to exist until torn down by the Tories, and continued by New Labour and Coalition, and largely doesn’t exist anymore except as part of the endless and unnecessary rounds of procurement framework vomiting out of The Cabinet Office and source of gov.uk.
https://en.m.wikipedia.org/wiki/Central_Computer_and_Telecommunications_Agency
Comms, Networking, Hosting, Computing Excellence, IT Strategy, Project Management Framework (PRINCE), Service Management Framework (ITIL) etc….
With deep irony - Indeed sounds very similar to this weeks (ignoring the financially starved Alan Turing Institute) exactly what NuNuLabour were wanting with Sovereign Computing Hosting/Services and AI nous.
change in culture and approach.
Having worked in this space in the past, there are too many people making decisions on tech that they aren't vaguely qualified to even think about, never mind make a decision.
I could wax lyrical about a few of the programmes I was involved in, but I suspect they will be so far removed from reality that none of the learned El Reg readers would believe a government entity could be so inefficient! </sarcasm>
Wasn't it Niel Armstrong who commented on how he was going to space on something built with the lowest bidder ?
It worked out for him, thank goodness, but sometimes the government would do well to remember that cheapest isn't always best.
Something that costs more often has a reason for it (not always though, it's tricky).
I remember this quote from Armageddon movie:
"You know we're sitting on four million pounds of fuel, one nuclear weapon and a thing that has 270,000 moving parts built by the lowest bidder. Makes you feel good, doesn't it?"
Not sure if that was paraphrased from the Apollo missions, be great it it was.
Apparently from John Glenn's biography:
“I guess the question I'm asked the most often is: "When you were sitting in that capsule listening to the count-down, how did you feel?" Well, the answer to that one is easy. I felt exactly how you would feel if you were getting ready to launch and knew you were sitting on top of two million parts -- all built by the lowest bidder on a government contract.”
Something that costs more often has a reason for it (not always though, it's tricky)
Speaking as someone who has done a number of Public Body procurement cycles - you really, really don't have to go for the lowest bidder. What you go for is someone that can do the job and offers 'value for money' - which is up to you to define.
And when you examine the cheaper option, they (generally) lack one or more of the feature set that you require and are hoping that you are sloppy and overlook that. They are generally also the ones that sue/put in lots of FOI requests if they don't get the contract that we were offering. Also, generally, the weasel words in their offerings reveal that they didn't actually have the capacity currently but that the contract would enable them to do so..
Doesn't mean we haven't had some turkeys but that's generally because that was all the procurement portal offered us and it was a case of "choose the least-worst"
Going for the lowest bidder, in public bodies (I don't know about private industry) is not about best value, it's just about lowest price. And that's because senior management want to spend as little as possible so that they can show how well they’re doing. . Projects are always underfunded - they’re even underfunded for the lowest bid, so that bits are omitted or ignored, to keep within an illusory budget. For example the lovely promise will include computer controlled door access. But when you ask where the budget for the computers controlling the access is they say, "don't worry that's a detail - it's be sorted out". And come completion date there will be a building with doors that have no access control - because no budget. Not to mention the promised floor to ceiling secure storage which has now become a standard cupboard with a wooden door. The security cameras at key points will have become one camera at the front and one at the back door. And so on.
persuaded Ministers that competent internal procurement and project management skills are unnecessary
And 46 years on from Margaret Thatcher this is what you get.
The question is given that the current Administration has a hugemajority (and therefore can do pretty much what they want, although it is usually a good idea to try for consensus if you want things to stay in place ) what will they do about it?
IDK. Hopefully something that will produce some enduring benefits.
>>And 46 years on from Margaret Thatcher this is what you get.
In another context, didn't Mr Gove mutter something about "....had enough of experts...."?
Transpires that if you don't have in-house experts you don't know what you are asking, so make mistakes when asking the questions and, susequently get shafted for the inevitable goal post moves.
Yeh it costs money to employ experts, who might be less than fully employed on projects, but it is an insurance against budget overruns due to incorrectly specified systems which probably works out cheaper than said overruns.
Icon cos not even Mr Holmes could have predicted the outcome /s
Not something I ever thought I'd do, but ... (deep breath)
In defence of Mr Gove, he said the British people had had enough of experts who were wrong.
Of course, I wanted the interviewer to ask Mr Gove if he happened to be 'an expert' on anything, but sadly that little dream of mine did not come true.
In fact, before the " make mistakes when asking the questions " bit they need to know what questions to ask. And that means both technical stuff of the "How do we integrate these different data sources" type and the "What do people do, why do they do it and how" questions.
Terry 6 > "How do we integrate these different data sources"
Well obviously. you need a computer for that.
and "What do people do, why do they do it and how"
Well now you're just being silly, expecting manglement to know what is actually going on and why.
Sorry. I'll get my coat, its the one with an abacus, my CV and a P45 in the pocket.
persuaded Ministers that competent internal procurement and project management skills are unnecessary
We actually have a pretty good head of procurement. Knows very little about IT but is more than happy to take advice from those of us that do while making sure that the contract language actually matches what the vendor had promised.. (not always guarenteed!)
Quote: "....Setting requirements for digital programs can be particularly difficult...."
Quote: "...contracts being awarded before the true requirement is fully understood..."
Ha....Twentieth Century Thinking hard at work!!
Wake Up...."requirements" are a thing of the past....what we need is "user stories" and a pretty wall covered in yellow sticky "Post-It" notes!!
Oh....and the idea that you might need some project managers, or a time line.......what you need is a "Product Owner" or two....
By now you've got the picture.....in 2025, IT development teams:
- don't have requirements
- don't use concepts like "architecture"
- really hate the idea of "management" (which is believed to be "interference")
- ...and because there are no "requirements", IT Teams have very limited ideas about testing
.........so is anyone surprised when estimates are poor and cost and time over-runs are frequent? Surely not!!!!!
Bring back the good old functional specs, I love them!
One source of truth for the Customer, BA's, Programmers, Testers, Service Management Team.
A lot easier than trying to work out the functionality of the system by trawling through 1000's of PBI's in Azure DevOps.
Private sector worker here….
We needed a new network monitoring system, so just 4 of us sat in a room for an afternoon and came up with a list of requirements for what we needed it to do.
After that we consolidated it down into a list of questions.
Then we found a number of suppliers who supposedly offered a suitable solution and we sent the questions to them and evaluated the replies before speaking to about 6 suppliers (some of which were not what we wanted) shortlisted it down to 2 and went for a proof of concept with one of them.
I suppose the only difference is that we were the people who would operate and administer the system so we did know what we were talking about…
I suppose the only difference is that we were the people who would operate and administer the system so we did know what we were talking about…
In the (UK) public sector that often seems to mean that you are disqualified from participating. Especially at the higher levels, where Oxbridge Old Etonians don't much trust the horny handed workers or uppity middle classes. We need to learn to know our place, which is not in the room where they're making important decisions.
"Current government processes from business case development to contract award do not work well for digital programs. Departments can present investment cases without a detailed assessment of technical feasibility"
That doesn't just apply to digital programmes. It's been a characteristic of UK and doubtless many other governments for years.
"Is the UK averagely bad or is there something about the UK Civil Service system that make it exceptional"
I've had exposure to UK and overseas government procurement and the UK isn't unique. There so often is a gap between government and private sector suppliers. I can best sum up as government is over optimistic about what suppliers will deliver for the price, and suppliers over promise. That's a big generalisation, but it's what I've seen over the years.
Coupled with the fact that failing to deliver is often more profitable than a successful project.
The "big names" fail almost* every time, yet they not only get the next contract, they even get paid extra to fix their own failure.
The small companies stop bothering to tender because they know they'll never get it.
* Citation needed
Government often know that to fight poor contractor performance in a court against well paid corporate lawyers would be prohibitively expensive and time consuming with ultimately little real chance of getting recompense appropriate for spending tax payers' money on it.
One could say that governments should equip themselves with the at least the same level of legal firepower as big business. The same argument could be made about technical skills as well. But it comes down to the same challenge that when it comes to salaries and the ability to attract and retain lots of high talent, government pockets aren't as deep (or as future focussed) as big business.
I can perhaps support that in microcosm.
As a small teaching service in a fairly small building we'd had our own cleaner and window cleaner for years. Local staff paid through the authority, but coming directly from our budget. They did a brilliant job, took pains to sort out the awkward details, like areas where we had kids being taught and so on.
Then came the corporate contracors- as required by the council. Which costed us more than we were paying, plus various extra central admin costs on top. They seldom had the same people for three consecutive weeks, paid them less than we'd been paying, but charged us significantly more. Failed to clean properly ( dusty carpets and porthole windows at best) and somehow weren't responsible for doing certain bits of cleaning ( like the kids' desks) that our cleaner always did.
Our local window cleaner didn't even bother to apply for the contract for our windows. It was just too complicated, and somehow the corporates were underbidding him even though we ended up paying more. Our local cleaner didn't apply for her equivalent job with the new corporate cleaning contractors because they were offering fewer hours and lower pay for the same job, even though we'd been charged more for the service. And our previously clean building and windows were filthy ever after.
Expect the UK to appoint a Digital Transition Czar to sort all this out in the next decade or four.
Dido Harding, Liz Truss or, drum roll, Boris himself. Boris can promise 'world leading' anything if he is fully funded in his role. Speak to his agent.
There is actually a really cost effective solution to this. Instead of paying £10bn for tech that doesn't work, we could pay the big tech outsourcers £5bn to do nothing (without any overspend). We could then go back to using paper, and still have a net saving. My only concern would be the great pen shortage of Summer 2026. Perhaps someone can ask Bic to ramp up supplies.
"The Public Accounts Committee has long maintained that technology can transform the way government delivers public services. Without a more strategic approach from the center, and a sourcing strategy that is fit for purpose for the digital age, the government risks wasting more money and squandering the opportunity to modernize the public sector."
Those two short sentences reveal the fundamental essence of the continued guaranteed serial failure which will blight governments, both home and away, friendly and hostile, as technology marches on to transform everything.
Centralised ministerial government personnel tasked with ensuring maintenance and retention of established historical command and control protocols/levers/taxing measures are a crooked fit in a new digital age which has leading prime technology and technologists recognising and able to advertise and present wasteful squandering applications/projects/programs just as soon as government sources unfit as free agents of radical change for a brave new universatile digital age venture forth into the public domain with announcements of such proposals as may primarily be designed to reinforce the status quo
You may like to consider that now it is best to accept that technology is into transforming public services which deliver government rather than government delivering public services.
I suggest that the author of the report, and all government IT procurement senior management be sent the following web address with the requirement to read the articles concerning large IT procurements and attended comments threads before their next procurement:
http://www.theregister.com
<Bows>
Procurement, not only public sector but also private sector are a fucking nightmare! I can only guess that if you're too stupid to go into HR or become a CEO, you go into purchasing.
It's insane that I could spend 6 months researching, meeting vendors, looking into stuff, looking into what the environment engineers are trained in etc etc etc and STILL some knob who can't turn a laptop on gets to over rule my decision on what to buy, because some vendor somewhere that I've already decided doesn't meet the requirements has managed to meet some sla on the business case
It's utterly nuts. If this was done properly, NONE of the big vendors would ever get another contract. WHY do they still get paid more money when they fuck up?!
Every vendor is out there screwing the tax payer. Charging the same yet offshoring the staff to pay less. You can't say.."You fucked us last time so we're not letting you bid this time "
You should be able to say..you didn't meet the requirements, you're not getting another government contract for a decade
Even at the lower levels. When I first went into education management I also supported the IT for my team and a few others, in my skilled amateur way. When there was stuff to buy I researched, then contacted vendors and if possible negotiated the best price.
But eventually we were given a central IT department with a purchasing manager. We were forbidden from buying kit through a well known catalogue,which was fair enough as they charged too much for much of their stuff and I only ever used them in extremis but had to buy it through him instead. I actually welcomed that, on the grounds that a) I didn't have the time to do it, really b) he'd have a more extensive knowledge of products and suppliers as a full time professional and c) his buying power meant he could get better deals. The extra charge we had to pay for his services would, I truly believed would be more than justified by getting better stuff on better terms. Believed, that is for about 3 months.
After which time it became increasingly obvious that he hadn't a clue. When I'd asked him to advise on the best product for a given requirement he just told me the standard well known, generic one- and I already knew there were better, more appropriate ones at better price, but he just hadn't done any research or got any knowledge. And when we'd asked him to supply some other bits of kit he bought items that were more expensive than I could get from the shops down the road - and in case you are thinking "ah support contacts/warranties etc". No, the most egregious part was that we found out he was just ordering everything from the same catalogue company we'd been forbidden to use ourselves, and at the same published prices.
It's like a perennial driving test failer going for their 16th test and failing yet again
I always thought that there should be a "5 strikes and you are out" on the driving test - if you haven't mastered how to drive a car on the road, in traffic, by then then you'll be a danger to other road users [1] when you finally do pass the test.
Driving a car isn't a right.
I'd also recommend that anyone going for a car test *has* to have spent time riding a bike (motor or human powered) *before* they are allowed to take the test - unless they can't for medical reasons. It teaches you to be a lot more respectful of other road users when you realise that you are a hell of a lot more vulnerable than them!
[1] Especially bikes (motor and human-powered).
Every IT provider will be able to give a detailed financial breakdown of their product development, implementation and operational costs, along with costs to change.
No government department is able to give a financial breakdown of the benefits associated with any of their projects beyond "if we don't spend it, some other bugger will".
A couple of personal experiences:
I spend 30% of my time working in the energy sector, for whom Ed Miliband is the Secretary of State for Energy Security and Net Zero (since July 2024). Ed's qualifications for this role? A degree in Philosophy, Politics and Economics from Corpus Christi College, Oxford, and an MSc in Economics from the LSE. One of the institutions with whom I work is extremely frustrated with the pie in the sky thinking coming from the department which is actually sacrificing energy security on the altar of net zero. I won't go further to keep my blood pressure down.
Another of my main occupations is providing technical input to certain government fuel related projects. My company was taken off a job by a NGO who did not like our work and yet copied and pasted large sections from our reports into their new report template. Their costs are over 30% more than ours and the end customer doesn't like their output! The person making the decision to employ the NGO instead of our private company has no qualifications in a relevant discipline but is a civil servant with a degree in classics and an MBA.
I detest the way that government continues to blame "experts" and "consultants" when on many occasions they simply disregard the work carried out by these people and do whatever is politically expedient.
BTW; an expert isn't someone that you were at Eton / Harrow with or a fellow member of the Bullingdon Club.
Every time this comes up I keep coming back to the fact that the government could save a truly colossal amount of money by recruiting a solid and well-paid in house development resource. We can no longer pretend that software systems are not part of core government functions, so not having the ability to run them from inside the government is ridiculous.
It would save money and keep domain knowledge within the organisation that needs it. We've had forty years to see that the private sector is not inherently more efficient than the public sector and that competent and happy internal teams can do outstanding work. Build up the resource and make it happen.
The churn rate of Windows versions alone must be costing billions in public money. Nothing will change as government exists primarily as a mechanism by which the rich can extract public money from the rest of us. Any actual benefit is incidental.
AI BS will just increase costs and energy use, whilst making everything less reliable. Daft. Much of what we use computing for would be cheaper and safer to do on paper.
Downvoter here, and here's why.
On the face of it, yes you are correct - UK Gov is of such a size that it could keep a significant team fully occupied, that would keep skills in-house, yada, yada, ...
But think a bit harder and you come up with why that's not going to work.
To start with, those skilled people (all the parts of the dev team from business analysts through to the coders and testers) would be civil servants (disclosure, I'm a CS hence posting anon). CS pay is not renowned as being competitive with the private sector so you won't be getting all the sharpest pencils in the box. Then you take a step back and think of the adage that there's nothing that making it a government function can't make worse. And even if you surmount those issues, you will still have the problem of learning about a major change to your project, deliverable "next week", from a minister on the evening news.
As others have said, what we DO need are stable requirements, proper budgets, senior civil servants prepared to tell ministers publicly that "no, you can't sensibly run a project like that", and teams adequately equipped (i.e. the right people in the right places with the right skills) to properly handle the private sector suppliers - as in, write proper specs, hold the supplier to them, don't hand out more money for fixing failures to meet the specs. But that also means having higher ups who know the difference between cost and value.
And as already said, it's not like it's something new, cue Benjamin Franklin “The bitterness of poor quality remains long after the sweetness of low price is forgotten.”
Dominic Cummings had the right solution for the established problem which has lions being led by donkeys up the garden path to nowhere worth arriving at ........ PM's adviser invites 'weirdos and misfits' to work for Number 10
When/Where is the Great Comeback/StartUp Party due, Dom? It is long overdue and the natives are getting restless and listless. Don’t let the buggers grind you down and have you giving up the ghost and your every right to host/mentor and monitor the field of dreams.
It is not as if there isn’t an almighty army of stealthy invisible untouchable help able to deliver every heartfelt tactical wish and vital strategic need, is it.
UK government tech procurement lacks understanding, says watchdog
That's better.
Just replace "tech" with "hospitals", "HS2", "Nuclear Power" or whatever.
“Successful delivery of the project appears to be unachievable,” it said.
“There are major issues with project definition, schedule, budget, quality and/or benefits delivery, which at this stage do not appear to be manageable or resolvable. The project may need re-scoping and/or its overall viability reassessed.”
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2025
