FAQs
Quality means entering all of your data from source documents
without errors or omissions. We achieve this with thorough
planning, testing, staff training, technology, and quality assurance
programs.
Planning
Good planning up front avoids poor quality at the end.
Our Project Managers consult with each client to understand exactly
how the data
output will be used to improve your business. Then we work with
you to develop project specifications, output formats, and project schedules.
The
most important task is to determine exactly how you will use the
output to achieve an organizational goal. This includes understanding
your
deadlines and project budget. We examine the source documents and determine
the
most effective method to extract the information. We suggest
an output
format (database, text file, new documents, etc.) and write clear
data entry and project specifications and instructions for our staff.
Testing
Before we begin full production for any project, we perform
tests to assure the final results are exactly what the customer wants.
Testing usually uncovers any quality or production issues, and lets
the
customer see the actual output and how it can be used.
Technology
Computer software can help verify accurate data entry.
We use field validation, table lookups, consistency formulas,
and logic rules. Validation tables are used to assure that only acceptable
data
is entered.
For
example,
one project required entering local districts in a county
database. We set up a table with the acceptable district names.
Data
entry
operators only needed to type the first few letters and the
computer filled
in the rest. The computer forced the correct spelling. This
is a simple example. Others include checking zip code, city and state
against
the
Postal Service database or checking medical diagnosis
codes
against
the
AMA database. If your project has a specific database
of acceptable information,
our software can set "validation" rules and checks.
Another validity check uses logic rules (or common sense). A social
security number can only be 9 numerical characters. No more, no less.
If your
data is date related, we limit the possible rage of dates that
can be entered. Anything else is flagged to check for errors. When
appropriate,
we use spell checking or proofreading software (and humans).
We also use random checks of data entry accuracy, proofreading and
visual inspection of output documents or databases, and data sorts
with logic rules.
In addition, we carefully monitor the accuracy of each staff
member to improve
training and reduce errors or delays.
OCR with Verification
For source documents with clear printing and
consistent formats, Optical Character Recognition (OCR) technology
provides
fast, automated data extraction. We use multiple OCR
products to test for
the best fit with your project. In addition, the OCR
software can be set
to the accuracy "confidence" level
you require, and trained to recognize characters that are specific
to your source documents (there is an endless number of fonts,
typefaces, type
sizes, printer
output errors, smudged output, old, faded documents, etc.).
In any OCR
project, operators are checking accuracy results. If OCR
cannot accurately recognize at
least 90% of the data, it is usually more cost effective to
use double-key entry of the document.
Double Entry Keying
One proven
tactic to guarantee accuracy is double keying the data. Two independent
data operators type the same data. The
computer compares the results and notifies the operator if the
two resulting
databases are different.
An independent operator can then decide what the correct data is.
If the source documents are illegible, we identify the problem
and send
the
customer a list
of documents they should look at to get the correct information.
Training
We constantly provide training to new and long time employees
to achieve high accuracy. For the most complex projects, staff receives
training
to make decisions when entering data, such as identifying
unique forms or
searching a
document to locate the correct information among possible fields
or choices.
Quality Assurance
Quality means that the finished
product meets your accuracy, format and deadline requirements.
The data has to work
in your
system to give you useful information. We develop a quality
assurance program for
each project.
The QA system includes procedures and schedules
for data entry, validation checking, random quality checks, and output
delivery. The schedules
are entered into a tracking system by manageable batch size. Output
is
checked
in the
China data
entry facility and in our US offices before delivery.
[top of page]
When can you use OCR/ICR/Automated technology
vs. manual data entry?
We can use OCR/ICR/automated technology to capture data for many projects.
OCR allows the computer to recognize characters, lines, tables, or
graphs in a document instead of having a person view and type the data.
Software is also
available
or is developed to automate HTML/SGML and XML tagging.
OCR works best when:
- The printing on each document is clear with consistent separation
between letters and sentences, columns in tables, surrounding graphics,
etc.
Also, drawings, graphics or art need to have clearly defined lines. This
allows
the scanner
and OCR software to easily recognize each separate character
of type
or line of a
drawing.
- The goal is to capture all of the data on a document
(full text data capture) or data is a uniform place on
a form. For example, a form with
the name and address typed in the same place on each form is usually
easy for an
OCR
engine to recognize
and place that information in the correct field of the
database.
With very
clear type or uniform forms, OCR can often achieve accuracy
rates above 95% to 97%.
To get higher accuracy, we have staff members visually verify
key information, or any field that is highlighted by the OCR software due
to lack
of "confidence" about
accuracy.
For many projects, poor quality documents containing
the source data does not allow OCR accuracy above 90%. In these cases
we
usually
discover that
double
key data entry achieves better accuracy at a lower cost
than OCR with human proofreading.
[top of page]
How are prices determined?
Data entry, indexing, file conversion,
and many other document management tasks can be very labor intensive.
These projects are well
suited to offshore processing because of the lower labor costs available.
Prices
usually
reflect the time required
to complete each unit of a job. Factors that affect prices
include:
- The quality of the source documents
- Output quality/accuracy requirements
- Ability to automate the process
- Total project volume
- Amount of document/file preparation or handling
- Turnaround time
Prices can vary considerably based on these
factors. For example, prices for data entry from documents with very
clear, typewritten
source documents can be as low as US$.40 cents per 1000 keystrokes
with accuracy rates
above
99.95%. Often programmers can develop automated
data entry techniques to reduce labor
intensive tasks and hold down prices. Lower
pricing is also available
for high volume projects or if lower accuracy is acceptable.
If the quality of the documents is poor, has lots of handwriting,
involves finding
data in different parts of the document, etc.,
then the entry process
is much slower. Prices will increase accordingly.
[top of page]
How do you receive documents and return
the data?
Source documents are first scanned to digital images (tif, pdf, jpg, etc.). For projects
requiring fast turnaround, the tif images are sent via the Internet or private
lines for offshore processing. For projects with very large volumes of images or longer turnaround deadlines, images are saved on DVD/CD or hard drive media and sent
to our offshore
facilities in 72 hours via common carriers
such as FedEx. Data is
returned in
computer file formats via email, FTP, or VPN,
or sent by common carrier on CD, tape, or other media you request.
[top of page]
How quickly can a project be completed?
CPI currently handles daily processing for clients that require
12-24 hour turnaround. Our offshore facilities have
a shift working during the
nighttime
in the US, so results can reach you the next
business day. We have the staff and facility flexibility to start
jobs with very short notice.
We commonly accept "rush" projects.
Before we accept any job we work with each customer to establish project
deadlines. One of our top quality goals is meeting the deadlines
you need to meet your business objectives. If we cannot schedule
a project
within your
time requirements,
we
will let you know up front.
[top of page]
How do you handle any problems that
occur during the project?
After we begin full production, problems can (and often do)
occur. The most frequent problem is that some of the source documents
cannot be
read accurately
or do not conform to the data entry/indexing
rules.
Our offshore staff
sends a
daily
progress report to our US project managers. Part of
the report is a list of problems. The US project manager
is responsible for getting
answers quickly,
making decisions based on prior work with
the client or contacting the customer to discuss how to best
resolve the problem. The
goal is to answer
questions
quickly to keep the project on schedule.
If the source documents have some poor quality
images, our
staff helps
you to find the problems. We keep a list of
illegible images or parts of an image.
The
customer's staff is then able to examine
the problems and supply the correct data. A second strategy is
to enter a code for any
illegible data. The
customer can then search the database of illegible
codes and make
needed corrections.
[top of page]
How do we work with an offshore
facility?
CPI has project management offices in the US and offshore
data entry facilities in Mainland China. This gives
you the advantage of local contacts for project planning and
reporting, and lower cost
employees for the
labor-intensive work. Our US staff handles
all coordination with
the offshore facilities.
In today's "virtual organizations" working offshore is not really different
than working down the street. Data is transferred
by the Internet or VPN. Data on
paper, CDs or tape moves worldwide by post office
or UPS. Meetings can be done by phone, email, netmeeting, or instant message.
Many
customers
choose not
to take time for face-to-face meetings. But, if
your job requires on-site consultations, we will be there. We are flexible
to work
in
the "office" that
meets your needs.
[top of page]
How can we monitor project progress?
For each project, CPI sets up a Web based project management
database. Work is tracked when sent offshore,
data is entered, output is checked for quality, data is submitted
to the US office for final
quality
review,
and then the output is submitted to the
client. If requested you can access the project monitoring database
for progress reports.
[top of page]
