Fear, Ignorance, Laziness, Greed
My experience of consulting in large organisations is one of mixed emotions.
On the one hand, it appears that navigating the purchasing process is long and arduous (and occasionally: random). It takes a lot of meetings, proposals, cost-benefit analyses, persuasion and horse-trading to get your project or upgrade financed. A side effect of this is that those who are experienced (i.e. have gone through the whole mess once before) will minimise the amount of stress they are subjected to.
There are many strategies for this. In no particular order:
Overstating the benefits. Since nobody can measure things that haven't happened, and everyone expects some degree of exaggeration, project benefits will always be higher on paper than in real-life. The only people who really, truly believe (or at least: repeat) them are politicians and the P.R. people. However, if you want your project to be approved, you have to be ready to make some outlandish (but not too extreme) claims for the benefits it will provide.
Padding the proposal Everyone expects to get less than they ask for. Hence they include some sacrificial lambs in the shopping list of stuff, so that when they are told they have to cut 20% off the costs, there are some items that can be axed. The trick seems to be to not make the paring-back process look too easy, or the bean-counters will ask for more. The problem is, that sometimes these (obviously padded) costs don't get challenged.
The twofer You have a BIG project that is critical for the business. You also have a few pet projects that you'd like to get done, too. The trick seems to be to merge them all into one, indistinguishable pile of interdependencies, so nobody can question why you need a 10TBmedia server for the LAN upgrade project. That way you only have to go through the pain and suffering of getting one single approval, rather than many. Most bean counters' eyes glaze over when you try to explain to them the technical stuff - frequently they'll sign-off on your requests just to get you to stop talking.
The bigger, the better It's a curious fact of business life that the more you ask for, the less resistance you meet. Say that your project will cost £100k and all sorts of people will stick their noses in: questioning your costs, asking if it's really necessary, can it be put off to next year (i.e.: cancelled)? But ask for £30 Mil and they'll all assume that you're serious and the organisation will be doomed to failure, or a takeover, or become uncompetitive, if your project gets the can.
As a consequence most middle-layer managers will submit proposals for a small number of mega-projects, rather than for what they need in reality: which is a dozen or two projects every year split between new work, upgrades, revamps and the occasional bit of blue-sky funding that might pay-off in 5 years time. The problem is that the people who's job it is to vet these submissions are unable to tell which ones are vital and which are pie-in-the-sky.
On the other hand, the mixed emotions I feel is that I am often in a position where I see these vanity/over-complicated pieces of work getting approved (real-life quote from an earlier boss: "it doesn't matter, it's only an extra 60k") when I know they are mostly unnecessary - but then again, they do pay for my consultancy time.