University orders investigation into Oracle finance disaster The University of Edinburgh has launched a review into its disastrous go-live of an Oracle finance system as it admitted suppliers have walked away over stalled payments. Information passed to The Register also shows university staff were forced to fetch their own supplies and charge them to expenses owing to a lack of stock …
Or,... the wrong people were paid (on time) to ask the wrong right questions to which the answer obviously is Oracle. The people in this set are never users or operators. You know who I suggest to belong in this set. Those who make the decisions are the ones without any factual knowledge about what they are deciding. For this set of people it is about being paid (on time) to prevent delaying decisions.
Been there, done that.
The only problem is that often bean counters decide, nor users neither IT.
"Paying for two systems at the same time?! No way!". And in the end, the costs are multiplied because tests weren't made, or other blunders (or lies) from supplier or software vendor, and everything collapses.
== Bring us Dabbsy back! ==
"When replace a working system with a new system, do not turn off working system until new system proven to work better"
^ This. When a system is working well, even with older software, someone will always come along to say that new and shiny will always work better and they thereby break the rule "Don't fix something that isn't broken" and it invariably ends in tears.
In accordance with the Peter Principle, the moron who decided that new and shiny was a good idea will escape the blame and will be promoted to create even more havoc at a later date.
Peter Principle would hold that this cock up is the result of a promotion. If it's proper Peter Principle, once you hit your level of incompetence you stop getting promoted; any further movement is based on an external factor - nepotism, cash, blackmail... You know, the classics.
And you don't run a non-US payroll system on an US payroll software, especially one made by Oracle...
They have absolutely no idea on how computers work outside of the USA, and will INSIST that dates are always entered or printed using the MM/DD/YYYY format.
And there is only one currency sign: $
"Oracle and Inoapps have so far declined the opportunity to comment".
And when they do it will probably only be to tell us their number one priority is to ensure the perfect on-time delivery of everything they have failed to deliver.
It seems to me the only people who have to say "our number one priority is..." are those who have failed to meet it.
This is almost identical to a incident at a large government laboratory I worked at. The admittedly aging accounts system was replaced by a shiny new Oracle system by switching off the old one and switching on the new one. Exactly the same issues -- suppliers were not paid who would then not sell us anything else until we paid up. The laboratory appeared on the front page of Computer Weekly and the management did get the privilege of appearing before the public accounts select committee. The minutes of which detail the full scale of what happened, which included a £100 million "anomaly".
This is also almost identical to the ongoing incident at a large corporation I happen to work for, but in this case we turned off a working (sort of) Oracle system and turned on a shiny new SAP system.
I have just finished talking to an outside contractor who provides a very important service for us, and while he was apologetic he won't be working for us anymore because he no longer believes our finance people's assurances that he will be paid on time.
Fortunately for me, I can make this someone else's problem. I also have a job interview next week.
Standard implementation across sectors proven to work. In this case with Sponsor being UK Universities as the sector group.
Seriously how different can Edinburgh University be to other universities that but needs a bespoke implementation. Same for councils, Govt departments, NHS etc
Payroll, HR, Planning, Procurement are fairly rudimentary processes with same interfaces to same partners sector-wide.
—
Also noted ‘staff forced to pay for their own taxi’s’ ?? Like WTF ??
You might think that, but you would be surprised at just how different Universities are from each other in their internal workings. Its a bit like saying Easyjet and Ryanair could just use the same sales, billing and loading system because they are both airlines - but they dont, because they are organised differently.
Because they made an explicit choice not to standardise.
The first rule of ERP club is dont modify the ERP.
The second rule of ERP club is turn down all invitations to join ERP club. Run for the hills if someone mentions finance transformation to you. Or more BOFH like, join the project for long enough to get it on your CV for buzzword bingo, but get yourself re-allocated to a "business critical" project prior to go live.
Obviously we don't hear ( here or in the news) about the ones that do work, are on time and within the quoted price.
But even so, as an outsider in these matters, it does seem as if it happens an awful lot and that these kinds of project are just too big and too complicated.
Is it that they are trying to get too many functions into one over complex, do everything, system?
Is there another way? Something more modular?
I have been part of one such migration back in the 1990's which other than a few relatively minor teething issues (you will always get some) went quite smoothly on go live day.
The two take outs I remember most from this were weeks of testing/training sessions with the supplier involving *all* departments of the business, and that as far as possible we adapted the business processes to fit the new ERP system rather than the other way around.
Same story at our university 5 miles down the road. An Oracle ERP/Fusion project gone horribly wrong. Long overdue on delivery and costs have more than doubled from the initial £5m tender. r12 still needing to be fired up in parallel for key financial reporting functions. Purchasing badly affected and even now none of our research groups can get an accurate picture of their finances. Seems to be a pattern here but unlike UoE our senior leadership have conveniently decided to kick any investigations into the long grass.
Or at least the US entity of my company.
Oracle ERP go live over December. Thankfully I work for UK entity. Less thankfully most of the tools I use are licensed by the US entity and Jan is our main renewal date.
Im resigned to the fact that all I can do is sit back with the popcorn....
A common result of people at the top being "sold" on the idea of a Software as a Service solution being the answer to everything.
Rather than take the solutions they've been building and developing for years and simply moving them to Cloud platforms and upgrading naturally, they took the decision to undertake a large, complex, multiyear program full of inherent dangers.
It doesn't matter who the vendor is: SaaS is not the answer to the question every time
I’ve been involved with ERP implementations for over 20 years now (started with the “wonderful” Y2K anomaly). The ones that I generally see go OK are where the business aligns with the processes in the ERP system. Where it goes to pot is when the idiots from Finance (or Purchasing, or HR) say “No, it has to work to my process”. Sales guy says no problem, budget remains unchanged, followed by bad analysis / decision making by the client. Throw in a bit of off-shore consultancy and the biggest wonder is why anyone believes this can possibly work.
But as people have said, those in charge of the project very rarely have experience or ability. But they do have one common trait – an ability to move off the project before it hits the fan.
This seems to be a replay of the situation in Oregon a few years ago. The state hired Oracle to create a user interface to the (then) new Obamacare health system. The software never worked and never went online. In the end, the state used the software provided by the federal government - which is used in Oregon to this day. Oregon sued Oracle for the $240 million wasted. The lawsuit eventually was settled for $100 million.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2025
