This CFO.com story (again!) is different, though, even though it has its shortcomings. Where it is probably spot-on is in the statement that preparers presently have little to no benefits from the implementation of XBRL as there are no tested ERP applications that truly integrate XBRL at this point. So all (complex) XBRL preparation will be done in a bolt-on fashion.
This is probably true for the time being, until the SAPs and Oracles of this world actually integrate XBRL GL into their products fully. On the other hand, the integrated approach of XBRL adoption is usually referred to as the most expensive one, thus the intermediate bolt-on solution will result in little additional expenditures.
The story gets much more disputable where it goes into the benefits for analysts. It quotes the FEI as poking holes in the other purported benefits of XBRL by insinuating that "The organization predicts XBRL could instead lead to analysts receiving excess information." This is an undue truncation of an admittedly complex, yet valid point made in the FEI's comment letter:
We believe we may be creating a situation where preparers will be providing more information than the analysts want (versus key information/data), later than when they need it (to update their models), thereby missing the real window of opportunity which is likely when a company releases its earnings for the quarter. Strategically, the long-term direction of this project needs to be determined and communicated − is it to upgrade the manner is which data is filed with the SEC or is it to provide (key) information to investors and analysts for their use?