We start by dividing the whole job among the number of operators. Each operator gets a certain percentage of the job. Once the basic entry is done, the job given to each operator is then redistributed for double entry. Each time, after the entry is completed, the data is run through a spell check for spelling errors. On completion of the double entry the two documents are compared to detect and correct discrepancies. The completed data is then proof read by a supervisor. This is a 100% proof reading. Finally, we do a random checking of the proof read data by a second supervisor. This ensures a minimum accuracy level of 99.995% on all our work.