There is of coure a very important problem charities trying to do this kind of measurments of efficiency face. They are in a market for public perceptiom to get

There is of coure a very important problem charities trying to do this kind of measurments of efficiency face. They are in a market for public perceptiom to get donations, and (if we take the numbers the article cites for Oxfam), can you imagine the loss of donations and goodwill after a few newspapers and TV programs run the headline "Charity X spends half their donations on statistics instead of helping people!"? Even if the costs could be amortized over a few years, the charity would be dead by then. Perverse incentive factors, again. (And the news might be right, collecting in depth statistics about the help program might just be a pretext to give a lucrative consulting contract to the director's best friend).