Although FineReader recognized much of the text on these congressional reports, the software had trouble with the poor scan quality and missed quite a bit.
Complete phrases we were looking for turned up nothing, but FineReader did pick up bits and pieces, allowing us to search and return some keywords.
The pattern seems to be the way the original document wrapped new lines within each text box, as the software failed to recognize these phrases as connected. It also seemed to get hung up on commas.
This may mean reporters have to search more broadly and dig through many pages before returning the results they need.