He then oddly goes on to talk about the buy vs build argument for things like Oracle/SAP financials/ERP - which are deeply hated by those that use them (granted, may not be able to make that kind of BigCo HR/financial stuff fun !) - but what is interesting is that those systems are not really off the shelf. They are usually multi-million dollar "implementations" and sometimes (seriously!) "re-implementations" of them - which cost orders of magnitude more then most IT bespoke projects - often due to licencing costs. This makes me think that the numbers make things look better then they are. For example: if the cost of an setup for SAP financials is 2mil, and 400K "implementation", then a blow out of 100% in the "implementation" cost is not, percentage wise, as big a deal as if it was a bespoke project (ie sans the licence cost) - as the bulk of the cost is a fixed licence.

This is also why building (as in skyscraper style) blow outs don't appear as huge - they labour cost is generally a relatively smaller portion of the project cost then the materials compared to IT projects.