City council faces £216.5M loss over Oracle system debacle The total cost of Birmingham City Council's Oracle implementation disaster is set to reach £216.5 million ($280.4 million) by April 2026, according to a new audit report. Europe's largest local authority went live with the Oracle Fusion single platform for finance, payroll, HR, procurement, supply chain management, and …
At least someone understands. It’s not the product, it’s poor management of the programme. The product works but one major problem in all these councils is that there is a disjoin between senior personnel who think themselves above mere mortals who actually do the work….instead of one single meeting where all are in attendance, there are multiple separate meetings and little progress. I know because I saw it happening at another council. They fired those responsible and then rehired them!!!!!
Very true. And, worse, if you get consultances with no morals they go along with whoever is dominate in those meetings, even if its clear they are ruining the project. They go along with it until its discovered the consultant isn't helping so they move on. When they finally get a consultant that tells them how it is, they don't like to hear this so they don't last long. You have directors in roles they shouldn't be, don't understand the system and just want to put their touch on it so they can stick it on their CV. All this and no one asks the users that will be actually using the system.
The main costs are when you do a system this big, you give up your database to the 3rd party, Oracle, and they NEVER allow you access to your own data directly. If you want to change anything in the database you need to request it. Which then costs more money. This isn't unique to Oracle, happens whenever you give away your database. The people in charge never understand this, also doesn't help when you have people in charage who are listening to a shady consultant who's clearly "helping" the other side, and you start ignoring your own staff who see this happening.
And yes, I've also seen all this happen at a council.
Yes as in the situation we actually had 1, only 1, that was good, nice and told them how it was "No, this is actually going to cost you more money to do what you've currently doing blah blah" (we'd already told them that but we're staff so got ignored). Sadly, she left before she could do good as she got offered more money at the original job she'd applied for and she saw it was clear they weren't going to listen to her unless she followed along with their shitty plan.
Its one of the many reasons this council is almost bankrupt. Yet the director of IT that is responsible for this shit show is still in place.
"if you get consultances with no morals they go along with whoever is dominate in those meetings, even if its clear they are ruining the project"
It's not just the ones that lack morals. Consultants need to work with people that are willing to make decisions...otherwise whats the point in hiring them?
I work as a consultant sometimes and I prefer to work with the people that make decisions and get shit done...because the role of a consultant is to aid in decisions and to help get shit done. You can't work with people that are cowards that won't make decisions...fence sitters.
There are two types of managers. Those that want to make everyone happy and be regarded as a good boss (David Brents)...and those that want to execute the company vision and get shit done and delivered (actual managers).
Consultants are usually brought in when the company gets a bit stagnant with it's ideas and needs to move forward...it's a way to bring in fresh thinking and find solutions that are invisible inside the company bubble.
Councils are terrible because there are too many people elected to councils these days. Too many cooks and all that.
But I've witnessed a consultant come in and sit on the fence with those managers, until someone finally looked at the consultant and said "Hang on, you're not supposed to be sitting on the fence with us". He was just there to milk the pay cheque.
If by elected to councils you mean hires bought in by bent CEOs and/or directors then yes. If you mean councillors then action can be taken there as councillors are forbidden to interfer with the day to day running of a council. Although that doesn't help if the councillor is a mate of a director.
"Consultants need to work with people that are willing to make decisions...otherwise whats the point in hiring them?"
Can't upvote this enough. I'm coming to the end of a contract right now, myself and another engineer were hired to work on a project with no scope, no clear goal other than "DR" and no Project Manager or BAs assigned. As it's a hotdesk environment I happened to be next to the head PM one day not long after I started and she told me that the "project" was poorly defined that she was keeping well away and letting her juniors get experience because she knew there was nothing for her to do.
After 6 months the business has now realised they aren't ready for engineers and need scope and lots of decisions to be made at board level (RPO/RTO, list of Tier One business systems still undefined for example), so we're both finishing up as our contracts end. This is a private sector business, but I've seen the exact same thing in the public sector plenty of times too. Usually new systems designed with the input of a couple of managers who don't use the system, and don't listen to the concerns of those who do or let the engineers/developers/PMs/BAs talk to them.
Of course you never provide them access to their own data, and if you have any sense you don't allow them to host their own data either.
Doing otherwise is a nightmare in terms of support. The only method of providing access is through a clearly defined interface, imports, or other front end. Otherwise what's to stop them corrupting audit data, and when the customer asks you to prove who made a change, you can't.
Not only are there nightmares of having customers insisting they host their own database, and then be incapable of actually administrating it properly, I've even had teams at work decide to query a database without asking the team who designed it. Their query worked for six months, and then broke, because they didn't actually know how it worked and the design did not account for all edge cases.
Once a schema is customer accessible, it can't be changed. If an interface is provided, the format of the underlying data can be modified whilst keeping the interface consistent.
But its also a nice earner for the people hosting the database as they now charge you if you want full access. All they give is a copy and if you really need to make main changes to the main database then that will be an extra large fee that wasn't in the contract. Because the fuck whit head of service that signed it off knows fuck all about IT. And assumed "Its our data, we can have it changed any time we wish".
Sounds entirely reasonable to me. If the customer can really change data willy nilly it's necessary to run a validation against all data prior to applying it to the main database. That's an expensive service to offer. Defined interfaces are much more manageable.
If you're too stupid to read and discuss contracts, or think about interfaces and data extracts when leaving, you shouldn't be in business. Either you know what you're doing and expend effort to specify and run the system properly, or you don't and you pay someone else money to do it for you.
Only a fool will charge low prices, it provides no buffer against costly issues that will inevitably crop up, and many customers will not be understanding if their use of the system is costing more than the income it generates.
Yep! If there isn't enough tax coming in then simply borrow more money on the national debt and kick that nasty can down the road for future generations. So long as Councillor Piggy-In-The-Trough gets his salary and his mates get a nice fat contracts, who cares where the money comes from today, paying it back is someone else's problem!
> "initial estimate of £19 million to a projected cost of £131 million"
Someone in Oracle got a new Ferrari as a bonus for that one. :D
Honestly it beggars belief how gov't (local or central) IT projects can f*ck up so badly, and I should know, I worked on one that similarly ballooned to £200M but was then cancelled.
It was a shit show when I joined mine (over budget and already late) and when I left 2 years later, very little had improved, in fact a full third of the 20,000 or so requirements hadn't even been allocated to a team to look at yet, and that was 5 years into a 3 year project!
"20,000 requirements" was the issue of course, as it was likely with this Oracle implementation for Birmingham, as every department in the council pushed for the system to do everything they could ever possibly wish for, for the next 20 years, probably without any ability to actually understand what they were asking for or its implication.
All the while Oracle consultants would be rubbing their hands together with glee saying: "Sure we can make it do that for you, let me just cost up that Change Request."
This is the perfect project for Oracle. They get paid near on 10x their initial bid, but don't actually have to deliver a system that works, and get to blame it all on the customer: "They just kept changing their mind." Could there be a better business model than that?
"I doubt if Oracle are getting all the money, most of that 10x is probably going to the incompetent consultants who tried to implement a poorly specified system."
Why do you say "incompetent"? They may have been entirely competent, but cynical and self interested. Recognising from the spec that this was a clusterf*ck-in-a-box, they bid low knowing that the gravy train would arrive, and from thereon it was just a case of printing out the invoices for an endless circular series of incompatible changes.
There may be honour, but there's no money telling a client "your specification is overly complex and unrealistic, the planned benefits are pie in the sky optimism, and the project as conceived is undeliverable and doomed to failure".
I don't understand why local government always seems to do this (and I work in it, so doubly so):
"We need a system to do X task"
"Ok, well this one is the cheapest"
"But it doesn't do the task in the way we do the task".....
Now, most sensible people would at this point undertake a process review and see how they can change the process to fit the standard, off the shelf software, OR go with a supplier who's process more closely matches the existing process. However, local government conversations seem to carry on like this
...."That's ok, we can get them to change it"
"Wont that cost money?"
"Not much, writing software is easy"
Then follows a depressing litany of poorly defined requirements, working, reworking, lack of user acceptance testing, all of which brings us to where we are.
Even more depressingly is that frequently the IT team are only involved after the decision to spend loads on changing off-the-shelf software has been made and collectively despair at the mess they find themselves in.
Simply put, council managers find it easier to spend money than to ask their people to work slightly differently. This may or may not be down to the kind of people low-paying councils are able to attract.
Not just local government, in my experience most software customers think like that. Few software companies or consultants will stand up to them, because the customers just find some other supplier who'll agree. By the time the shit hits the fan the supplier is too deeply embedded to easily change.
"Few software companies or consultants will stand up to them"
It's very likely the salesmen. There's a commission to earn, a target to meet and a bonus at the end of the month/quarter/year.
I've managed to walk away from one employer whose salesman assured a customer that the product would be a straightforward replacement for the existing S/W. I knew what the two databases were like & could see the one was going to work very differently from the other. I had enough trouble trying to graft some very basic security into ours in terms of restricting visibility according to user roles. I had no desire to be the man in the middle between the salesman and the customer on that one.
I worked (freelance) at a blue chip company you'll have heard of, and was once asked by a Sales Manager to come with him on a sales meeting to see a prospective new customer, to answer technical questions.
"Sure, I'll come." I said. "But I won't lie for you."
"Errrr, ok." He replied. "Maybe you can sit this one out."
He never asked me again! :D
Possibly because the LGPS is nice; I know, I'm paying into one as a member of professional services staff at a University.
It is also possible that many local government employees were unable to find suitably well-remunerated employment in the private sector because they were fucking useless.
True story. I was working for a - competitor - ERP vendor at the time.
Stakeholder - of a government entity - says to me: "Our corporate communication guidelines are not to use the term "Client", we prefer to use "Customer".
Me: Well, OK....
Stakeholder: "And your software uses unified, central, labels, to associate with data fields, right? So how about we just switch CLIENTID's to "Customer"?
Me: Great idea! Now you can open and review about 1600 different pages to see if the longer label does not cause any problems. If they do, cool, just customize the page, great fun next upgrade.
Or another moronic entity - a world-wide NGO - which wanted to change the data model to add a new EmployeeContract instance each time the employee got a raise. Because, yeah, that makes a lot sense. Mind you, they were a careful shop and ran a tight ship: each Monday they would fly in one peerless MSProject wonder-PM from Paris to keep their 3-years-late project plan on track. The consultancy was killing it on that little project.
Yes, IT vendors do screw up ERP systems on a repeat basis. But the first line of defense is to minimize needless busywork on these things. The second is to ruthlessly focus on need-to-haves, at least until the first go-live. The third is not to count your "projected savings" until they are hatched, full-grown and have stuck around long enough to be legal at the local pub.
The original poster had a point: Customer is a longer word than Client, so even if the change is technically easy, there may well still be cosmetic problems. Depending on how the position of columns is determined, you might end up with a report showing a clipped final column, overlapping column headings, or a column titled 'Custon'. It is not just screen presentation to worry about, where you might end up with a scroll bar to allow viewing of the now wider report, but paper presentation as well. What is the likelyhood that there is really no data transfer between departments somewhere that does not involve the sender printing out a report and the receiver typing it in again?
Part of doing proper internationalization is allowing for that on your screens and reports. If you've crammed everything up as tightly as you can you'll have problems even localizing to a language like German, where many terms tend to be longer than their English equivalents (not to mention accessibility issues).
True, but the Client vs Customer issue actually happened in French, which is already longer than English. Changing that label was literally 3 minutes work - replace, in the French metadata, the content of that field's default label. Presto, all done. But, as you point out, you really need to check that there is space for the new label everywhere the field is shown to users. And adjust if necessary. I could have easily told the customer which pages had this field, even the ones using the default label. Could even have trawled through the security configuration to limit this review to only the pages which were actually used by any of their users, if the final security access rights had been configured (unlikely early in an implementation). But I wouldn't have been able to compute spacing to flag placement issues.
It was so obviously stupid that even a techy like me could carry the day pointing that out. But what made our software configurable almost invariably dragged customer implementations into quicksand. The customers had the same tools and compilers as the our own (vendor-side) developers, so they could go to town modding. Part of the problem was also the sales cycle: our salesmen sold that flexibility. An undisciplined government IT Dept - or even more often, one too subordinate to the Accounting or HR Dept in corporate seniority - would never have balls to stand up to endless cycles of nitpicking, or sometimes just stupid data re-modelling effort. "Well, we've re-implemented your Department Tree" was something another years-late customer told me when I did converted some stuff for them and panicked when said Department Tree - an uber-critical component - seemed to be broken afterwards. Throwing more staff at these projects just makes things worse - it literally feeds the fires.
You really need to start out with a Project Director who has the direct support the C-suite and the authority to tell the user stakeholders to buzz off when appropriate. Then that person needs to have rally good functional SMEs who understand the product, the general business domain and that particular client company's business processes well enough to know exactly how to configure what's likely actually not very special at all, using out-of-the-box functionality. Even have them change their business processes a bit, if really warranted. Customize and create new code artifacts only in the need-to-have areas, like data integration feeds: very little busywork for the techies. Train, interview - and cajole - the heck of the actual line workers, those projects typically only ask their supervisors' input. Strangely enough: having too little resources and too little time can actually be a wonderful stimuli - in the right hands - to get to a successful go-live quickly.
But it is all down to good project and expectations management and some of the worst performers in that league are the big consultancies whose incentives are billing more hours..
Even the initial guess (I'm not going to call it an estimate) of £19 million should have been an immediate red flag. What's that, at least 100 man-years or something? Any IT project of that scale is, in all likelihood, too complicated to succeed. Cut down your requirements and do something smaller.
Judging by the balance sheets of new entrant ERP providers, £5-10m.
The problem is that the public sector doesn't pay enough to attract and retain proper digital skills (or indeed other skills), and it doesn't reward people when things go right, or hold them to account when they go wrong. So there's nobody in house able to project manage a "green field" new council resource planning system, and more importantly than the PM and dev skills, nobody with the vision and energy to come up with the idea and drive it through.
The UK does have these sort of tech skilled and entrepreneurial people (we're one down, based on recent events), but why would they work in local or national government for a mediocre salary, when coding up and selling a new ERP or EPOS system will net them millions?
Good point. Maybe a "center of excellence" that dispatches teams to re-implement systems time and again for other government departments? There so much lost $$$ in those failures that you could pay them royally if you had to and once you've done something 10 times, the 11th is much easier.
When Obamacare went live, their sign-up website promptly went tits up, IIRC. Very experienced private sector/startup techies who were sympathetic to the notion of public health care - ie not incentivized to gouge - volunteered, took over from the consultancy and stabilized it fairly quickly.
Birmingham City Council is bankrupt
There has been a mass sell off of buildings owned by BCC to cover the cost of the shortfall it has. The Birmingham City FC owner has bought Birmingham Wheels and the surrounding area for an absolute steal because of this. There are plenty of traffic lights with broken bulbs on them. There are plenty of roads with abnormally bad road surfaces. My mom doesn't get her green bin collected even though she paid for it. The council won't come out and fix the fence their tenant broke on her.
But somehow, still, BCC are on the hook to Oracle for a monumental fuck up of an IT project and continue to hire Oracle experts who say the best thing to do is to keep going with it?
Fuck them all out of it. If the city is bankrupt enough to be unable to fix a pensioners fence panel then it's bankrupt enough to not pay the fucking bill for these cowboys.
The council are legally required to have a financial system, so they can file audited accounts. They do not have that. They are breaking the law.
They are legally required to spend the money to fix that.
Now, you could question whether they are going about fixing it the right way. (I would say not). But they do have to fix it.
Used to be, you would hire an army of guys with green eyeshades and sleeve protectors to just *do* it. Forget the automation.
In my county in the US, the county supervisors just signed a multi-year contracts with a new company to handle the residential trash collection. It has been, in the words of one supervisor, "an unmitigated disaster" with trash not being picked up on time for thousands of residents, going on several weeks. They are looking for ways out of the contract. But in the meantime, they have been redirecting existing county employees in the road maintenance department to pick up the trash, and then they send the overtime bill to the under-performing contractor.
Of course it does but Comrade Starmer and his Stasi need to be seen to be getting tough on the "mythical far right"(*) else the WEF won't welcome Kier back to Davos next year!
( Far-right my arse! It was mostly bored people looking for a fight plus ton of ratboys nicking suasage rolls from Greggs and bathbombs from Lush for their girlfriends! )
This may be an example of a more widespread problem. The executive branch of our various forms of government really see themselves as "in charge", with the elected members who are notionally "in charge" actually being an obstacle to progress. So it is entirely likely that the councillors were kept in the dark.
I have seen this myself at a public meeting for a planning enquiry (remember them?) where I overheard the Chief Executive of the council complaining to a friendly journalist (who, I noted, did not report it) that "the problem is, these councillors don't realise we have a council to run".
This is why things seem not to change very much. The people we vote for are not really "in charge" in any meaningful sense.
That's why said Chief Executive needs to grab any and all of the councilors that will be in any way so grabbed, and go for several beers/coffees with them, to educate, persuade and cajole. That's part of their job surely?
My local permanent staff keep pushing idiotic schemes that make zero fiscal sense and the local councillors are too timid to ask the hard questions - i.e. why are we leasing parking meters when the outright purchase cost is barely 2 years lease ?, (they then bought them and due entrenched local opposition they then reabolished parking charges again (much to the utter fury of the permanent staffer who needs his head removed from his derriere and made to realise he is answerable to the public, the public aren't beholden to him - arrogant rude and frankly infuriating little man
I've often thought that the way to fix this would be to have the council manglement on a basic salary with a bonus to be voted on by the electorate for delivering, to the electorate's satisfaction, verious services or projects.
> I've often thought that the way to fix this would be to have the council manglement on a basic salary with a bonus to be voted on by the electorate for delivering, to the electorate's satisfaction, verious services or projects
even better would be the idea from the Isle of Taga from David Eddings Tamuli books - there anyone elected has all their assets confiscated and added to the public purse - If they do well then they would get more out that they put in, screw up like here and they would exit bankrupt !
I would hope there are some other large councils who have a successful system, which could be cloned.
The cloning is a know problem - the education of all of the staff would be expensive - but not as much as a rewrite
Moving the data would be a problem - it is bound to be in the wrong format, and by now may be inconsistent or missing.
Am I to naive?
Why are they not all using the same system to do the same job in the same processes?
It's nonsense that individual councils have to source their own because inevitable one mess-up like this, plus all the interoperability problems between councils, means that it actually costs far more than just having one, single, integrated, shared system.
Of course, that "costs far more" means that Oracle, a salesman, a local councillor, some contractors, even more consultants, etc. are ALL making bunce from it, which is why it happens like that.
It's the same mentality that results in every council having different recycling rules. They all think they're special, and need to justify their existence by creating a custom solution for their particular needs. They scream if central government tries to impose a standard, while also complaining when it cuts funding for their custom local fun and games.
Most local government isn't fit for purpose, and that's not a problem limited to the UK.
Whilst your comment has some (considerable) validity, recycling rules probably aren't the best exemplar given that they will be dependent on the contract negotiated with the waste collection contractor. Now it you think there should be one single national waste collection contract, perhaps we could get rid of those variations....
This is the age old debate of should you have:
A) A centralised system that the whole country uses, cost efficient as it's designed and implemented once with a central IT function and using best practice from across the country. BUT imposes a top down view of the world that's slow and costly to change for any one council's needs and takes no account of local factors, likely spawning a host of smaller home grown satellite IT systems to close the functionality gaps.
B) A local system for local people, that meet the needs of the local council's issues and is easier and cheaper to change, and change can be implemented quickly. BUT means you have 60 odd systems across the country, all doing things slightly differently, duplicating effort and cost.
That philosophical debate covers not only councils, but hospitals, police, fire service, ambulance service etc.
Plus there is massive resistance to change from entrenched empires within these organisations, from people who will lose their jobs or positions if there was to be a change. Not to mention the back handers that definitely do go on.
> This is the age old debate of should you have:
Or option C: Core system for legally required functions with optional add on modules as needed by some individual councils
Yes it would need everyone to do it the same way but the cost savings would be enormous once it is done
Sadly we are more likely to get that pig that just flew past my window than this utopia.. and yes that massive resistance to change from entrenched empires within these organisations would make it a non-starter
Actually that's how they got into this situation.
Original system highly bespoked SAP install.
New Top Bod " bespoking kills compatibility to upgrade to later release, we'll use Oracle straight out the box."
This might work if the old system is vigorously documented, all features carefully analysed to discover if they are actually needed and then mapped into the new system using configuration settings built into the baseline system IE no programming.
But to do this properly is a massive PITA (and I have no idea if it was even done at all).
Let's step back a bit. Other than name recognition why would you assume 2 companies that built their reputations on supplying system primarily for manufacturing environments (multinational mfg, of the likes of Ford in the case of SAP) would have a single f**king clue about how to run a UK local authority?
When I say it out loud does it not sound a bit retarded?
Correct
I ran the UK marketing campaigns for the two companies that were merged into what is now MS Dynamics.
Both were bought for their market share, not their mutual compatibility, but at least the Danish system had strong internationalisation capabilities.
Anyone who was involved at the time will know exactly who I am. Those who weren't won't.
"delivered, and, furthermore, due to the inability to monitor budgets, "
"academics, consultants, and activists"
"management, and customer"
What the hell is going on with El Reg these days? This is not Daley Thompson's Decathlon, you don't have to randomly keep hitting the comma key for no reason just to keep the website running.
Anyway, working for a software company I once led a project to implement a smaller non-Oracle erp system for a customer. Contract and specs prepared for the customer by a 'Big 4' consultancy. Off we went and got to first stage of user testing. 'But it doesn't do this!' It's not meant to we said. 'But it's in the contract!' No, it isn't we said., you bought the standard system. The Big 4 outfit had not legally bound the non-standard requirements list into the formal contract documents, it was just a wish list...It cost the Big 4 outfit dearly...
I've worked as a freelance contractor to a sub-contractor on a number of public sector contracts. Delivered to spec and on time. I had no visibility as to budget but certainly heard no complaints. Modularity for the most part meant that new functionality was added easily and cost-effectively. (The exception was one where my client's IT director, decided to specify how the the database should be designed which made it a bit less easily extensible. They should have shifted it over to the 2nd version we did for later contracts.)
Of course you never see any headlines about that. And then, being public sector, everything had to be re tendered after a few years... Not one of the ones I was involved with but one of the clients lost their flagship contract and from one I heard there were some serious shenanigans involved in that.
As always, great points.
If I could add one further: The users in these situations are not going to suggest processes that result in them losing their jobs. Consequently, there's little by way of incentive to be more efficient with the new system.
In any change, I've tried to group people up into:
People who want the change
People who will benefit from the change
People who will be disadvantaged by the change
You need a different means of engagement for each category and be clear about which groups & individuals are in each. Conflate them and disaster will happen. Malicious obedience is very difficult to undo once it's set in.
Most project set out with the assumption that everyone has the same intent in mind. This is rarely the case for the above reasons. As a consequence, requirements are captured in a way that assumes they all have equal validity, which of course they can't.
On top of this, there's the challenge of those within an organisation who are experts at what they do as they've done it for years. Sure they have hugely valuable insight into how it has been done. But that's not the point of the new system. Taking 10 your old processes and welding them into a brand new system is usually a bad idea. They cost a small fortune to customise into the last system and will cost a large fortune to weld into the new one. What's needed is a step back with someone who knows what that process looks like at it's most efficient, present that and then see how far from 'good' the clients wants to be, based on specific business / vertical needs.
As someone further up stated, this is not a path that's easy for a consultant to follow as you're instantly up against the established knowledge and as such you often have to back down to keep your job.
John Cotton, leader of Birmingham City Council, said: "Report after report shows that there's a national crisis in local government caused by 14 years of neglect from the previous Tory government, combined with major rises in demand and cost-led pressures."
Steady mate, there's a mere hint of your loyalties showing. Heaven forfend that there was any incomptant or greedy people in the Brummie council offices that contributed to this shambles.
The system also left the council unable to provide an audit trail or detect fraud for 18 months.
The council said it will comment in due course - Sent from the Councillors diamond-encrusted iPhone on his solid gold yacht.
Seriously though - "somehow" the countil top-execs are now multi-multi-multi-millionaires. And there's no papertrail or audit possible to explain how they ALL have 100x their salary.
int main(enter the void)... The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2025
