Cold comfort to teachers who got paid late, but ERP software rollout had 'unrealistic' timeline A UK public authority responsible for about £1.1 billion ($1.43 billion) in annual spending damaged its ERP project — which saw SAP ditched in favor of Unit4 — by underestimating its complexity and kicking off with an "unrealistic timeline of 15 months," according to a public report. Surrey County Council busted its original …
"Local councillors (like MPs) are elected. They do not have to have any particular skills or experience, or be competent at anything."
This is correct, but professional officers are there to provide this advice and expertise. In my experience it is rare for councillors to simply ignore officers' advice, certainly on something as public-facing as this. Senior officers overruling the professional advice of subject matter experts is, of course, another thing entirely...
A problem occurs when the elected officers do not have enough baseline competency to determine whether the 'expert' advice they are getting is actually any good. All experts are not equal, and if you put two in a room, they are unlikely to agree. El Reg comments are full of experts disagreeing with experts. In the circumstances, I would like to see simpler solutions employed, airgapping internal and net facing systems, using simpler software and using less software. Printed forms that you can post are a cheaper solution than a digital system that can get hacked, fall over or cost zillions.
Probably and they've also probably been promoted. After all, these places are always run on the "Peter Principle" promote incompetence, to shield you from your own incompetence. Anyone that has any skill or knowledge gets kick around and they only stay because 1. They want to stuff to be done properly and hope they can make it happen 2. They don't have the confidence to go elsewhere.
Fucks me right off, to no end that this happens time and time and time and time again. Poor management, by managers who are out of their depth but kiss arse so are where they are, yet super shit at their job, push for unrealistic goals and when its pointed out "This can't be done, that's not how the tech works" and/or "This is outside of my JD, I'm not skilled" are just told "Get on with it, we sometimes have to do work outside our JD". Yeah and that's why this fucked up you cunt!
Sorry, went on a personal rant at the end.
After a quarter century of experience in IT project consulting, I find that that is easily the most elusive quality, especially in administrations.
In a private company, it is easier to find someone who actually wants the project to succeed - because it will benefit his (or her) department directly. That doesn't mean that it is automatically so however, I have found exceptions and private companies have internal politics that are just as obscure as the real thing.
But in public service, it is far more common to have a project leader that is absolutely not interested in actually leading the project. After all, he has no skin involved and his (or her) position will endure whatever the result.
That kind of mentality hinders the success of any project by default. In the private sector, it is less prevalent because you can, ultimately, always get fired, or removed from your position. But in public administration ? Never going to happen, even with people who have a history of failure.
That irks me to no end.
I didnt fly over anything, corporate leaders often have zero experience in the companies they lead... case in point in recent times Beoing leadership aka those that joined from MD did not have engineering backgrounds. Funny how these same leaders were the ones that destroyed MD and then jumped across to B and are driving it down.
How exactly iss someone who doesnt understand the core business or tech of the company possibly judge how to make decisions ?
""That kind of mentality hinders the success of any project by default. In the private sector, it is less prevalent because you can, ultimately, always get fired, or removed from your position. But in public administration ? Never going to happen, even with people who have a history of failure."
Your experience is what it is, but I have to say that my experience as a business strategist is that around four out of five large commercial ERP projects go hugely over-budget usually in the 40-100% range. In one instance, a company I worked for spent over £200m updating SAP, and scrapped the lot before it went live, tried again with a new SAP plan and spent over £400m (not to mention business on-costs and aged debt losses of a similar magnitude), and still had a system that was barely fit for purpose. As a result it had five straight years of losses totalling many hundreds of millions, and it got taken over. The acquirer migrated all the customers to a new ERP, and wrote off that second SAP instance.
> From what I read on these hallowed pages, they are frought with failure
They can work (I was part of one at my last job) and there is no one single magic bullet to make the project work, I believe in my case it was involving all the company departments in exploring the entire system on the INT instance well before go live and changing how we did things to keep the new ERP system as standard as possible, It is possible to do that but many people do not like changing how they do things.
Of course it was a relatively small to medium sized company but the same principles are there as long as you have strong project management who can get things done. And we spent weeks in that room testing everything.
> around four out of five large commercial ERP projects go hugely over-budget usually in the 40-100% range.
Wow. In any other industry, surely the vendor (SAP, Oracle etc) would take steps to limit reputational damage by refactoring their products to match more closely what their customers are buying to reduce need for customisation, or retraining their channel partners, or preferably both?
Imagine if you sold large capital items eg warships and they turned out to need completely new ammo and battle procedures just because the customer wasn’t able to articulate requirements and bought a “standard” product?
>Imagine if you sold large capital items eg warships and they turned out to need completely new ammo and battle procedures<
That would be a great opportunity. You get to sell a change request for a couple more mega bucks and you can even argue it doesn't make sense to test the original weapons and might never have to deliver those.
I am pretty sure stuff like that happens with defense contractors as frequently as it does with ERP systems.
'Sue' in the UK is a civil matter not a criminal one, and it is about the terms of a contract of employment. So the remedy is damages - and to get those you would need to quantify the actual damage done. By the time anyone got a court date, the arrears would probably have been paid.
Reading between the lines in the OA (and icon for Lindsay Clark for covering all this local authority/IT stuff) I get the impression some kind of detail was missing that was required for payroll in the July/August period leading to an urgent request for information from local management. Most council departments would be able to scramble to get the information, but schools will be out at that point in the calendar (for weird historical reasons to do with farming). So impasse.
In a four decade career I've only had one non-payment of wages in an educational institution: grovelling apology and £25(*) automatic compensation for inconvenience plus offering to make up any interest payments on bridging loans.
(*) Well pre-millennium. £25 was worth more than a posh breakfast then.
The detail that was missing in July/August (and to an extent June) was available staff to work with the support team, because they were either 100% on exams or 100% on holiday. And no, they cannot delay exams or holidays 'cause they have to fit around the pesky school year.
So the point was a lack of foresight. Which to be fair exists in the commercial world - I found myself regularly prodding PM's along the lines of 'you know that release is scheduled for Christmas/Summer Hols/Easter/whenever when parents etc. will be wanting to spend time with family'.
Fortunately in this more modern age, most of the companies I work for have outsourced a lot of their development etc. to far flung parts that have different holiday patterns, so all but the worst planning can now be fudged.
This is where the genuine consultant comes into play; one who goes round, listening to the little people because they actually know what goes on, and presents it along with a substantial invoice. Because the information is now more expensive than it was it must now be more valuable and heeded. Price = Value.
They did but but it was during COVID. As the report says ". This had a material impact on the volume of late change requests submitted during the Design and Build phases of the project, which undermined the programme in its later stages"
With the effect that "Lack of engagement at this phase of the project meant that real learning and knowledge transfer on the part of users only started in earnest with User Acceptance Testing (UAT) in July 2021. Seven rounds of UAT would eventually be requested, reflecting the lack of early staff exposure and engagement with the system, and which contributed to the delays"
This has been a common feature of the various HMG projects I've been involved with.
One of them (the first) had such appalling management that, before myself and 80 other contractors were brought in to rescue the failing project, which was several years late and 10 million early '70s pounds over budget, said Civil Service management had been ejected and replaced with ICL's Dataskill people. Anyway, Dataskill and we assorted bunch of contractors and the better Civil Servant IT people got the project completed, tested and runnig live in 18 months flat.
As I've said previously, my experience on other Civil Service projects has been that the staff and lower grade of management on Civil Service projects tended to be quite good but the higher manglement grades workshy (one individual was considered a waste of space by his subordinates: he was only in the office 50% or less of the time. The rest of his week was apparently spent on setting up his retirement project: an antique business. The levels above him were more interested on scoring points off each other than actually working: when after 18 months the project I was employed on was awaiting sign-off and implementation, one of the other higher-ups succeeded in getting our 18 month project cancelled: apparently this was purely due to dominance games among higher management. Needless to say, out team of EOs and HEOs were severely pissed off, especially as the same thing had already happened to the project at least once before.
Leaders dont ask the little people because they are too stupid to know what to ask and unable to understand the any answers.
Most for example in this story wouldnt have a clue what an ERP system should or shouldnt do,...
@Dr Who "When a mitigating factor for huge time and cost overruns is that the project "... has culminated in the delivery of a functioning ERP system""
Only 18 months late and £11.3 million over budget for IT that counts as a successfully delivered project, and it only cost in total £57.9 million. Beers all-round. ;)
There still are, but they're increasingly not wanted. The powers only want to hear reinforcement of what they already believe, and pretty much all decision making is based on "consensus". Consequently, any anyone making a minority observation or warning about a hazard gets ignored.
Or the SMEs give realistic time scales based on experience working within Council but management or consultants are sure that it can be done faster.
The SMEs input is then acknowledged but overridden by management that give shorter timeframes to save time and simplify integrations or how they will handle existing processes to reduce costs.
Middle management then convince upper management they have a viable plan and upper management get Council sign off for the costs and implementation plan.
Then reality is introduced.
I bet you won't be able to predict where this goes wrong...
Note: some (many - my sample size is small and consists of large metro Councils) Councils are ridiculously complex because of the mix of services they have historically been involved in or are forced to be involved in by government regulations (I.e. integrations between health/justice/law enforcement/social care/housing to protect vulnerable people). There should be better ways of doing this but that's far outside the scope of an ERP system that you plan to implement within a 3-5 year period.
That only works if you know what deliverables you want, before signing the contract.
People never know what they want the software to do. At least not in sufficient detail to write a good spec.
Writing that spec requires significant time and expertise. Even then, it won't be completely right first time. There will be changes.
If you take a rigid approach, then the software developer delivers something that's not what you want but meets the letter of the contract. Alternatively, if you write a contract that says "final software will do what we want and we can change our minds" then no sane software developer would agree to that.
Not all councils do everything as other councils, some have fewer responsibilities than others.
Some councils have processes that are in-house, some have onsite contractors, some have off-site contractors
Even then, departments in different councils with the same responsibilities will have different workflows to achieve the same ends.
Very good.
Now can Surrey explain what they do that is so very different to what, say, Essex or Somerset does that requires everybody to have custom systems costing obscene amounts of money, are poorly specified, and seem to end up costing many times more for something that arrives late, if at all, and probably doesn't even do a bunch of things that it is supposed to do.
With the appalling rate of child poverty in the UK these days (*), this sort of thing is crying out for a formal investigation to try to understand how and why so much money is being wasted.
* - not to mention social services, civic services, education, mental health services, hostels, and so on for a painfully long list of things that have either undergone brutal cutbacks or shut down completely.
Amazing how th elikes of McDonalds can produce SOPs for cooking burgers etc, in nearly 40,000 restaurants globally, even allowing for local menus and so forth, delivered through a predominantly franchisee based business model, yet in the UK we can't get just over 160 district councils to standardise how they attach a wheelie bin to a truck. The excuse "we have our procedures and ways of doing things round here" isn't the excuse, it's the problem.
Councils always struck me as needing to be the ultimate franchisee business model.... with elections every four years to award the new contract.
No ERP rollout ever works. Ever. Mostly to generate revenue for such service providers.
I've seen colleagues in our US subsidiaries have to go unpaid for THREE MONTHS because of ERP screwup. Back home we still don't have decent control of project costs; with managers literally incentivised NOT to book time to their projects despite both legal and financial reasons that it's better that they DO.
I've seen shifts from Oracle to SAP to god knows what, and not one of them has ever worked.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
