My company does a lot of form-based human workflow. Comparable to
insurance underwriting. Semi-skilled pink-collar employees review
files full of different bits of paper, type certain data elements into
a computer system (but data entry isn't the main task, evaluation of
the overall form, and interaction with the evaluation software are the
main tasks ), and deal with exceptions. Sometimes this process will
require several iterations, may span hours or even a couple of days,
which could include 1-2 incidents of customer contact.
Over the past couple of years, after trying and largely failing with
OCR, we have been experimenting with KFI--key from image. This process
uses purpose-built software to guide the user through a heads-up,
eyes-forward process to key data from a form, into a data-capture
application. That data-capture application then feeds the
abovementioned evaluation application.
I am finally getting to my question. KFI has significant efficiencies
versus traditional KFP (key from paper). However, we have not found
RFI (review from image) to have comparable payoff so far. My
literature search shows that KFI is a well-known term, but RFI seems
unknown. So I am wondering what other companies have experienced.
As a confirmed paper-hater, I have a hypothesis. I grudgingly admit
that paper has its place--high contrast, fast random access. I believe
that for RFI to beat RFP, the viewer app must provide a high
value-add. More than the same old zoom-in, zoom-out, rotate, annotate.
I'm not quite sure what that value-add consists of, I have a few
ideas, but I have to believe someone out there has researched and/or
experienced this.
We would be very interested in findings. Either as part of Google
Answers, or perhaps as part of a paid consulting arrangement. If you
can provide a decent intial answer, we will pay $20 "earnest money",
and we can negotiate from there. I work for a big company, so paying a
few hundred dollars for good information would be a no-brainer, of
course there is potential for more. |