"Scientific method", as controversial as this phrase is, almost always contains the need for predictions - a theory will be scientific if it's (among other things, but at the very least) able to predict future phenomenons.
I'm reading Husserl's Crisis of European Sciences (haven't finished it yet), in which Husserl presents the accomplishment of the "Galilean technique which is called physics" as "nothing but prediction [Husserl italicized] extended to infinity". Then he goes on to state that actually everything in our life are predictions (where he also previously compares prediction to induction) - "seeing, perceiving, is essentially having-something-itself and at the same time having-something-in-advance... All praxis, with its projects, involves inductions [or- predictions]".
Now if we take Husserl's presentation of physics as "nothing but predictions", and life as always involving predictions, we can ask - is there any use for a method that doesn't involve predictions? [Note- I'm talking about natural phenomenons' research.]
Furthermore, if we're not going with Husserl's comparison of induction and prediction, or, if we simply don't accept that everything in life involves induction/prediction, but still take his presentation of physics, we can ask if a method that doesn't involve predictions can actually be useful and unearth certain aspects of reality we could not with the current methods that do rely on (or, requires) predictions?
[I suspect that the transcendental phenomenology that Husserl suggests, which I have yet to read but have a general sense of, is somewhat in line with this thinking, so I might just need to read ahead but the question simply struck me so I had to go here and ask. But in any case - even if Husserl's phenomenology is an example, what I'd like to recieve is either more examples (not necessary, if there are) with further explanation of the idea in the question, or a refutation of the question.]