A little disapointed

I also feel that this product is a little premature. I am running a reasonably fast machine with a gig of memory. I am using a non twain compliant fujitsu scansnap set to 600 dpi. Even so, the recognition process takes too long and seems to miss about 10% of the data. I have figured out that it helps tremendously to scan w-2's separately from your other source doents. Then you will only have to do recognition on a few doents. I have tried to do it the other way, ie try to let the machine figure out which doents out of 19 pages of source doents are the meaningful w-2's or 1099's. After a few minutes it eventually came back with an error message. Not very helpful. Fewer doents means recognition in the nature of a minute instead of 3 to 5 minutes.I have tried a couple of 1099's with no success. So far the only time this will save me is if I have several new w-2's with all of their e-file info to put in, or for a new client to fill in address and ss numbers. Too bad, it is a great idea if it works!