Really, all methods of estimating intelligence (which is ill-defined anyway) are full of bullshit. The big advantage of IQ tests is that they have a different sort of bullshit, and may find intelligence (or lack of it) in people who don't seem intelligent.
Some time ago, children who did badly in school were usually considered stupid, with the usual caveats for those who overcame that and went on to do neat stuff. When IQ tests were introduced into schools, they revealed that there were a lot of children who were doing badly despite being intelligent, and often helped get these students into environments that were better for them. (Then, of course, some schools tried classifying children on IQ and related tests, and putting them into different tracks, which shows that not all educators are actually intelligent.)
You have to be intelligent, in some sense, to score high on an IQ test. (You don't need to be competent; I've known MENSA members I wouldn't trust with a burnt-out match.) Some studies have found a modest correlation between IQ and success, with the correlation going away at higher levels of IQ. Another person I know underwent a cognitive evaluation, and one of the things in several pages of detailed discussion was an IQ score. The tests are useful in some circumstances. Just don't expect a single number to tell you everything.