After I told him about one of my patents, my brother asked me "How long did that take?" Here's how the "invention" process works in my part of the corporate world.
1) Think hard, write in lab notebook.
2) Show to coworkers, ask 'em what they think. If they go "huh," can't find any reason why it won't work, but aren't really sure it will, you're on the right track.
3) Fill out invention disclosure, about 3 pages' worth, of which two paragraphs describe the invention, and one is a photocopy of the lab notebook. One is signature lines, for you and your witnesses.
4) Work on something else for a while.
5) If it impresses the lawyers and upper managers as having merit (or sufficiently baffles them, I suppose), talk to a lawyer about the invention, show him (haven't had a her yet) a couple other pages from the notebook, maybe some hardware.
6) Work on something else for a while.
7) Collect the 30 page fax from the fax machine, figure out which order they go in, and wade through an incredibly obscure rendering of the idea you had, expanded to cover every imaginable variant that has any remote relation; if you've put 6 things together, that's about 6! (6 factorial, or 6 x 5 x 4 x 3 x 2 x 1 = 720) claims. Mark it up.
8) Work on something else for a while.
9-11) Lather, rinse, repeat
12) Get a call from the legal department, walk over and sign the application, sign away all my rights to the invention, collect $x (a modest, and I suppose confidential sum).
13) Work on something else for quite a while.
14) If lucky, the patent will show up a year or two later. Often, it gets kicked back and you have to go back to step 9, or maybe 7. (But do not collect any more $x bills.)
The invention part? Somewhere between a moment of insight and a couple days of pen and ink. Getting ready takes a long time, though. You have to muddle around in problems for a while.
My bio page has a list of what's come out the pipeline for me, so far.
Tom von Alten tva_∂t_fortboise_⋅_org