£3,000,000,000 to update National Savings
Discussion
.
And likely to rise? And they cannot say (they don't know) where the spending has gone!
£50+ from every man, woman and child in the country!
Meanwhile the tax payer will be paying millions of pounds (£400,000,000) in compensation for errors surrounding deceased customers' accounts.
Just WTF has the "management" of this country come to? You really couldn't make it up!
Source: https://www.bbc.co.uk/news/articles/cp8637dnmn7o - scroll down to Daily Telegraph.
And likely to rise? And they cannot say (they don't know) where the spending has gone!
£50+ from every man, woman and child in the country!
Meanwhile the tax payer will be paying millions of pounds (£400,000,000) in compensation for errors surrounding deceased customers' accounts.
Just WTF has the "management" of this country come to? You really couldn't make it up!
Source: https://www.bbc.co.uk/news/articles/cp8637dnmn7o - scroll down to Daily Telegraph.
Edited by Cliftonite on Thursday 26th March 11:08
Surely this is non-event. NS&I is of part of the government so unless the money's down the back of the sofa the government's got it, had it, or whatever.
As regards the cost of updating their systems there will no doubt be others who can indicate what a "reasonable" cost might be. I can't imagine NS&I needs a system much different from any other bank of financial institution so they shouldn't need to reinvent the wheel.
If savers have lost track of their certificates, changed address or whatever it's not exactly NS&I's fault. Look at all the muppets who lose ££££ when they lose track of their Bitcoin.
As regards the cost of updating their systems there will no doubt be others who can indicate what a "reasonable" cost might be. I can't imagine NS&I needs a system much different from any other bank of financial institution so they shouldn't need to reinvent the wheel.
If savers have lost track of their certificates, changed address or whatever it's not exactly NS&I's fault. Look at all the muppets who lose ££££ when they lose track of their Bitcoin.
Dont worry, this kind of thing would never happen in the private sector hence why all the money is being poured into.. checks notes. The private sector!
Strategic Partners: NS&I is partnering with multiple firms to replace its aging IT infrastructure:
Atos: Continuing to run the core banking and payment systems under a revamped agreement until 2028.
IBM: Delivering digital experience, digital enablement, and cloud-native platform services.
Sopra Steria: Handling customer contact centres and operations.
Capgemini: Acting as a consultancy partner to aid in the transition.
Strategic Partners: NS&I is partnering with multiple firms to replace its aging IT infrastructure:
Atos: Continuing to run the core banking and payment systems under a revamped agreement until 2028.
IBM: Delivering digital experience, digital enablement, and cloud-native platform services.
Sopra Steria: Handling customer contact centres and operations.
Capgemini: Acting as a consultancy partner to aid in the transition.
There's another thread about this from some months ago. The fundamental problem is that NS&I got themselves in a position years ago where they were massively exposed to a single software provider. If that firm went bust, NS&I would be completely f
ked. And that firm has damn nearly gone bust. So NS&I absolutely need to move onto IT platforms that get rid of that risk. And that is inevitably going to be a massive project and a project that is impossible to plan and price in advance with any degree of accuracy.
That's the key thing to recognise. You cannot actually predict what work really needs to be done. You only discover that as you try to actually do the work. As a result of that you cannot predict with much accuracy how long one of these projects will take nor how much it will cost.
"How long will it take and how much will it cost?" turns out to be an unanswerable question, so it's really stupid to insist on getting an answer and even more stupid to then lose your st when the answer turns out to be wrong. Of course it's wrong. The reality is it was a stupid question to ask. Relying on the answer for your wider planning is asinine and profoundly irresponsible because you are deliberately ignoring significant uncertainty that you know exists. If you're not managing known risks, just what the fk ARE you doing?
Instead of focusing on inaccurate early forecasts, people should focus on delivering work EFFICIENTLY and then track actual progress and use that as a basis for making informed and educated guesses about the timelines and costs for the rest of the project.
ked. And that firm has damn nearly gone bust. So NS&I absolutely need to move onto IT platforms that get rid of that risk. And that is inevitably going to be a massive project and a project that is impossible to plan and price in advance with any degree of accuracy.That's the key thing to recognise. You cannot actually predict what work really needs to be done. You only discover that as you try to actually do the work. As a result of that you cannot predict with much accuracy how long one of these projects will take nor how much it will cost.
"How long will it take and how much will it cost?" turns out to be an unanswerable question, so it's really stupid to insist on getting an answer and even more stupid to then lose your s
t when the answer turns out to be wrong. Of course it's wrong. The reality is it was a stupid question to ask. Relying on the answer for your wider planning is asinine and profoundly irresponsible because you are deliberately ignoring significant uncertainty that you know exists. If you're not managing known risks, just what the fk ARE you doing?Instead of focusing on inaccurate early forecasts, people should focus on delivering work EFFICIENTLY and then track actual progress and use that as a basis for making informed and educated guesses about the timelines and costs for the rest of the project.
Panamax said:
As regards the cost of updating their systems there will no doubt be others who can indicate what a "reasonable" cost might be.
At one one of the spectrum, JPMorgan spends about $20bn a year on IT. That's not some massive fundamental re-engineering activity; that's just the normal business-as-usual spend. NS&I's business is simpler and narrower and a bit more bespoke, so you'd certainly expect it to be a lot less expensive to run than JPM. Twenty times cheaper might seem like a reasonable guess? And if that's just the BAU running cost, multiply it by a bit for a fundamental re-engineering project?
ATG said:
Instead of focusing on inaccurate early forecasts, people should focus on delivering work EFFICIENTLY and then track actual progress and use that as a basis for making informed and educated guesses about the timelines and costs for the rest of the project.
That's great, but politically unworkable as a party in power wants to be able to say "We're investing £xxx in improving services", and the opposition wants to have figures that they can use to beat the Government with when it inevitably goes pear-shaped. ATG said:
That's the key thing to recognise. You cannot actually predict what work really needs to be done. You only discover that as you try to actually do the work. As a result of that you cannot predict with much accuracy how long one of these projects will take nor how much it will cost.
I guess part of the problem will be messy data. You can stick that in any new system you like but it will still be messy. Working through that mess will be a very big spin-off task from the project, if they bother. They've probably preferred to sit on their hands and not go looking for anything unless and until someone complains. They will know how much money should be in the pot and they will know which people 99% of it belongs to but the catch is that 1% of a very big number is a very big pile of trouble to sort out, especially if names change, addresses change and so on.. IIRC there was a time when DVLA tracked VIN numbers but didn't tie anything together with engine numbers even though the engine number is shown on V5. These days they do have it fully cross-referenced but only make the information available to the police etc. No public access.
If you wanted to be radical, you could get rid of NS&I as an actual customer service provider and out source all of its work to a commercial bank, building society or asset manager. NS&I becomes a guarantor and overseer and a pretty normal commercial customer of a bank; a customer who wants the bank to help them to raise capital. The private sector provider could be motivated in pretty much exactly the same way that they are when selling anyone else's assets to end investors. Normal market forces and motivations should just work without creating lots of unavoidable perverse incentives. Yay, for privatisation. Where's Thatcher when you need her?
boyse7en said:
ATG said:
Instead of focusing on inaccurate early forecasts, people should focus on delivering work EFFICIENTLY and then track actual progress and use that as a basis for making informed and educated guesses about the timelines and costs for the rest of the project.
That's great, but politically unworkable as a party in power wants to be able to say "We're investing £xxx in improving services", and the opposition wants to have figures that they can use to beat the Government with when it inevitably goes pear-shaped. ks is the root of the problem. People should be held to account for doing that.Panamax said:
ATG said:
That's the key thing to recognise. You cannot actually predict what work really needs to be done. You only discover that as you try to actually do the work. As a result of that you cannot predict with much accuracy how long one of these projects will take nor how much it will cost.
I guess part of the problem will be messy data. You can stick that in any new system you like but it will still be messy. Working through that mess will be a very big spin-off task from the project, if they bother. They've probably preferred to sit on their hands and not go looking for anything unless and until someone complains. They will know how much money should be in the pot and they will know which people 99% of it belongs to but the catch is that 1% of a very big number is a very big pile of trouble to sort out, especially if names change, addresses change and so on.. IIRC there was a time when DVLA tracked VIN numbers but didn't tie anything together with engine numbers even though the engine number is shown on V5. These days they do have it fully cross-referenced but only make the information available to the police etc. No public access.
Typically an institution won't even know what systems they are currently using when they kick off a project like this. They literally won't even know what applications their users are actually running to keep the lights on. They won't have an accurate understanding of what the systems they know about actually do. And they won't have an accurate picture of what and how data moves between those systems.
One trivial example. Everywhere I've worked has had some way for users to manually amend the data in the bank's risk management systems and the bank's general ledger. Obviously. That inevitably evolves into people building their own spreadsheets to generate the amendments. And it may start with just little tweaks here and there to true-up the numbers, but it inevitably snowballs into material amounts of the bank's most basic data coming out of unmanaged, undocumented Excel spreadsheets ... huge numbers of these things. Creating the things becomes a job role. Why? "Because IT are too slow/expensive, so we solve these problems for ourselves. Aren't we clever!". True in part, but it slowly puts the institution in a massive, massive hole.
ATG said:
If you wanted to be radical, you could get rid of NS&I as an actual customer service provider and out source all of its work to a commercial bank, building society or asset manager. NS&I becomes a guarantor and overseer and a pretty normal commercial customer of a bank; a customer who wants the bank to help them to raise capital. The private sector provider could be motivated in pretty much exactly the same way that they are when selling anyone else's assets to end investors. Normal market forces and motivations should just work without creating lots of unavoidable perverse incentives. Yay, for privatisation. Where's Thatcher when you need her?
Thatcher privatised Girobank, but didn’t privatise NS&I, which says something.ATG said:
One trivial example. Everywhere I've worked has had some way for users to manually amend the data in the bank's risk management systems and the bank's general ledger. Obviously. That inevitably evolves into people building their own spreadsheets to generate the amendments. And it may start with just little tweaks here and there to true-up the numbers, but it inevitably snowballs into material amounts of the bank's most basic data coming out of unmanaged, undocumented Excel spreadsheets ... huge numbers of these things. Creating the things becomes a job role. Why? "Because IT are too slow/expensive, so we solve these problems for ourselves. Aren't we clever!". True in part, but it slowly puts the institution in a massive, massive hole.
Banks auditors absolutely hate that these days, EUDA's they call them (End User Developed Applications). They're forbidden by the auditors.Various papers reporting that the chief executive has been sacked
Edit: Telegraph https://www.telegraph.co.uk/news/2026/03/26/nsi-bo...
BBC are reporting he’s resigned, so the story above may change
Edit: Telegraph https://www.telegraph.co.uk/news/2026/03/26/nsi-bo...
BBC are reporting he’s resigned, so the story above may change
Gassing Station | News, Politics & Economics | Top of Page | What's New | My Stuff