Data Entry Still Key to Productivity

Because manual labor is the most expensive part of any data entry operation, maximizing productivity is of critical importance. TAWPI's Information Capture Professional (ICP) certification program places great emphasis on this area, as well as seminar sessions and educational content aimed at increasing

Industry standards for heads-down data entry operators average 12,000 keystrokes per hour. For transaction-based data entry operators, where there are more complex forms, data entry rates may only reach 8,000 keystrokes per hour. By using a combination of manual data entry and character recognition, rates can exceed 20,000 keystrokes per hour.

Regardless of how quickly information is entered, any loss of accuracy can undermine productivity gains. In his book, Measuring and Improving Data Entry Productivity, Arthur Gingrande estimates that it can cost up to 4,000 times more to detect and correct a data entry error than it cost to originally key it.

Tips for Increasing Data Operator Efficiencies

1. Modern automated data entry tools can eliminate up to 90% of existing forms processing keystrokes. Advanced keying schemes, when coupled with intelligent recognition, improve accuracy and significantly reduce operator fatigue.

2. Organizations can leverage hardware investments by using software to automatically read information whenever possible. Technology is consistent, doesn't tire and allows operators to better focus their energies.

3. The amount of operator effort needed will depend on the type of form being processed, so clearly understanding and defining a processing job will simplify operator tasks.

4. Individual data entry operators have different skills. To keep accuracy and processing speed up, give those new to the job, "easier" keying tasks, like phone numbers and dates. More complex keying tasks, such as addresses, can be given to more skilled workers.

5. Field or "snippet" keying, versus full form keying, makes the division of work (described above) possible. Streaming several of the same field types to keyers saves time:

* Users do not have to scan full forms to identify the field to be keyed.

* A keying rhythm is established because keyers know exactly what to expect.

* Fatigue from eye strain and repetition is drastically reduced.

6. Using intelligent recognition technology also provides benefits:

* The element of human error is reduced because character recognition provides a higher level of precision, allowing operators more time to determine if the information is accurate.

* Forms typically contain more than one kind of text and data entry can be simplified by using technology that combines multiple engines to recognize combinations of text types.

* Intelligent recognition software improves recognition performance by incorporating custom lookup databases, rules on field properties, context, and business rules.

* Using data entry tools that allow specification of fields such as the number of digits, range of values, or specific vocabularies for data lookup increases accuracy and improves operator efficiency.

* Form processing software that applies logic to data entry, dictates how information should be entered, reducing required manual entry.

A Laundry List of Efficiency Techniques

1. Use multiple recognition technologies to read all types of text including machine print, handprint, cursive, alpha and numeric characters.

2. Use technology that employs more than one engine to decipher information.

3. Use software capable of whole word recognition and character recognition.

4. Use verification techniques such as custom databases and templates.

5. Use cross-validation to ensure that data corresponds to other related fields.

To illustrate how the above techniques increase keyer efficiency, consider entering an address field. Typically there are 32 characters to be keyed. Advanced keying schemes can reduce 32 keystrokes to 15 keystrokes by incorporating a postal address database. By entering the 5-digit zip code, the state and city are automatically known. Next, when the keyer enters the street number, the possible street list is shortened because there are limited combinations of street number and street name.

Finally, a "look ahead" list (type-ahead buffer) can be used to fill in the street name after just a few characters are typed in. For example, by typing in "bra," the entire street name of "broadway" shows as a complete street name. Intelligent recognition can automatically read up to 80% of address blocks without human keying - and when combined with the efficient keying schemes described above - can eliminate over 90% of unnecessary keying currently being used with traditional data entry tools.