To give an indication, from personal experience, of the issues.
I worked on a large government programme where software changes were first raised at a software engineering review board; if successfully passed it would then proceed to a sub-system engineering review board; then a system engineering review board and finally a programme design review board.
At each stage about 10-12 different departments were represented who would raise questions and issues which would have to be actioned, reviewed and closed before proceeding to the next step. If things went without issue it might take a year, some changes were still being reviewed after 2 years - and this without the work actually starting.
On another programme changes were further graded by priority as the funding and engineering team size were both limited. Emergency and priority 1 changes would be added to the build list; priority 2 would be placed on a wish list; priority 3 would be dropped as it was accepted they'd never be implemented, always be pushed down by new emergency/priority 1 changes.
Once that was out of the way the changes then had to be added to a build list for a planned configuration change and into it's test programme.
And by the time all that had been done, the change may have been rendered either unnecessary because of a further software change, or another change required it to go back to stage 1.
The same went for documents. I worked on system interfaces and had to produce/update ICDs, the weeks and months involved attending boards, updating and changing minor changes such as acronym, data dictionary and name changes, let alone substantial changes.
We had an entire, substantial, team who dealt with nothing except configuration control. We even had involved processes for defining under which category a document/change should be listed and numbered and 4 levels of change numbering.