A long time ago I worked as the System Designer for a medium-sized engineering company, and the immediate requirement was for a database that would contain full details of how every product was manufactured (parts lists, sub-assemblies, details of each stage of production of each part, and so on) This was to be used for Production Planning as well as for Costing.
While arranging the data-input though I discovered that the number of changes of such data (as recorded on "Change Notes") was extraordinarily high, but that a high proportion of such Change Notes did no more than cancel a previous one. (Literally, the complete text of such a cancellation might read "Do not implement Change Number XXX")
I discovered that whenever a Change Note resulted in new tooling, new drawings or whatever, those involved habitually kept the old stuff aside for up to a year, so as to be able to merely switch everything back when the relevant change got cancelled, but with the hardware and software then available, keeping a record of the effects of every change in such a way that any given change could be reversed in a similar way was simply not a possibility. Even if the full effect of any given change could have been unambiguously identified, there would often have been all kinds of issues when more than one change had affected a single product.
It then occurred to me that my employer was spending large multiples of my salary each month in getting hundreds of unwanted changes made and then scrapped, and with a little more investigation I discovered that those involved often knew very well that a particular change would be revoked but assumed that every change they saw had been ordered by the Technical Director, while in fact no check was made on the feasibility or cost etc of any change until after it was implemented !!
I introduced a proposal that would have subjected each proposal to at least a cursory check before it was cleared to be implemented, and a meeting was held to consider this.
The Technical Director vetoed the whole idea though, because although it would have saved the company a multiple of my salary, it would have meant that the documentation ran to one extra sheet of A4 paper per change, a total of about 500 sheets per month, and as the Technical Director put it "We really must avoid wasting money on stationery!"
It was only when I got back to my office that I realised that what I should have done was to offer to pay for all the extra paper myself (500 sheets/month) in exchange for (say) half of the savings that would have been made by the company.