No Big Splash after dropping the Big Deal

Just read "Leaving the 'Big Deal'... Five Years Later" by Jonathan Nabea & David C. Fowler.

• Via OpenSIUC: http://opensiuc.lib.siu.edu/morris_articles/63/
• Via Taylor and Francis Online: http://www.tandfonline.com/doi/full/10.1080/0361526X.2015.1048037#

This article describes analysis on cancellation of three Big Deals five years later from two institutions.  Here are some of the conclusions with my commentary:

1. Demand for the content is not high enough to return to the Big Deal, since ILL requests for content which would have be covered by the Big Deal is about 10% of the downloads previously recorded.  I think the first part of the conclusion may very well be correct, but I don't think that ILL figures after cancellation of an ejournal collection doesn't map directly to real demand.  In fact, in extreme, the difference between the two could be interpreted in the opposite direction: access has dropped to 10% of what demand was previously.  Neither extremes are correct and the truth probably lies somewhere in the middle since neither number is a good measure of actual demand in my opinion.
2. Savings were significant, particularly considering the size of the overall budgets, annual increases and inflation, and comparing it to the monographs budget.  This is not really an analysis but a statement, but I'm not sure that comparing the money saved with the purchasing power for books is useful.  Journals and books are two separate parts of a library's collection and it's not immediately valuable to say that with all the money we saved buy journals in a specific way, we were able to buy books in a different way.  The article uses this comparison as an illustration only but I'm not sure it's a very useful one.
3. Dropping the Big Deals gives us more flexibility.  Amen.  I think this is one of the best arguments for doing this.  The most dangerous part of the Big Deal is the lack of flexibility and control a library has when participating in it.  I just wish that flexibility and control could be given a dollar value so we could compare.

Not much analysis was done for this but something's better than nothing, I guess.  I would have expected something about any patron feedback that may be lingering, specific collections picked up since cancellation, changes in usage data for other collections or resources, etc.  What do you think?  Have you gone through any Big Deal cancellations?  What impact have you seen or expect to see?