Also this goes beyond just ERP - core banking packages like Temenos or Avaloq are the same. They have something called 'model bank' which is some default simple model of common expected behaviors, workflows etc.

But almost none of the banks actually work like that, apart form some small insignificant shops which can't afford such customization. Each has some unique stuff that makes it better than most competition in that area. Behemoths of banking world run their pretty unique things, workflows, specific jurisdiction reporting etc. so each of them have more or less massive customization on top of basic common stuff.

I guess it goes without saying almost constant flow of changes, either from business or regulators keeps hefty IT teams busy to maintain that customization and ensure things keep working. Also logically the complexity has only one direction to go, and so it goes, slowly but surely.

I am not even sure if business properly realizes all this, maybe they do or maybe they are kept in continuous cloud of bullshit from IT managers who of course want to preserve their jobs and importance (headcount, impact) within corporation. Anyway the prospect of migration to something else is something they (rightly) detest and want to avoid if possible, since costs and duration are massive, potential issues tremendous and backbreaking, and there is absolutely no guarantee of success.

Tax agencies might be the way to hammer this into organizations. If they demanded specific sorts of data for specific activities for audits everything else would quickly evolve to support that.

Seen a few projects which are equivalent to a tenant cutting off structural columns in a high/mid rise to install safes. Are people going to be punished for demanding impossible customizations or attempting them, or rewarded for "keeping the integrity" like magicalhippo and their client did, as mentioned in their comment?

Real question is why does it need customising… every council is going to have supply chain, payroll, asset management etc

We went live with our largest client to date by far recently. Initially they wanted a ton of customizations and integrations. However, since we're a relatively small company punching above our weight, we pushed back hard on stuff where we felt they could follow our standard flow and integrations which hundreds of other customers use.

After we went live they thanked us for this, as it had forced them to rethink their processes, and they were pleased it had led them to much better and more optimized ways of working.

Used to work for a large airline IT provider on departure-control software (check-in, boarding, etc.)

Each airline on-boarded would demand mandatory features that they couldn't do without. Keep in mind these features weren't free, they were billed market-rate development charges - though the company didn't make money off that, the core business model was usage-based billing.

More than 80% of implemented features would go completely unused.

Too much to build. UK government has shown it can build styling templates, payment and notification wrappers well. These are useful and achievable services.

Full on ERP would be a disaster.

The system is working as intended to those that matter.

It's interesting that the third-party consultants hired by the city to implement it aren't named in these articles and "Oracle" is taking the brunt of the blame, much like "Ticketmaster" doesn't care if you don't like them or not, you're going to use them. It's not just Oracle's fault for a 200m pound invoice, and I agree jumping from SAP to Oracle is probably one reason why the customization costs that much more than going from some smaller accounting system.

You have third-party consultants charging for addressing all the mistakes in the system, present on daily chats and constantly chasing them around for support. Oracle's always upselling you on the parts of the system you haven't implemented yet.

The actual municipal employees are making 75% (my own estimate) of what their industry peers make in salary so they're just trying to do their job as they found it, not attempt to replace the status quo with a "solved, open source, one-size-fits-all solution" like so many here believe various ERP systems are. Everything sounds better on paper.