Reading codes without errors

Data-entry errors cluster in a few predictable places. If you know the places, you can slow down for exactly those characters and keep full speed everywhere else.

The usual suspects

• 0 and O, 1 and l and I, 5 and S, 8 and B. On many fonts these look alike. Learn the pattern of the source: invoice numbers rarely contain letters in the middle, product codes rarely start with a digit.
• Transposed digits. 48213 typed as 48231. Your eyes read the pair, your fingers reverse it. Reading digits in groups of two or three ("48, 21, 3") reduces this sharply.
• Dropped characters at the end. The last character of a long code is the most often skipped, because you are already looking at the next record.
• Case and symbols. Some systems treat ka and KA as different values. Hyphens, slashes and dots count.

A checking routine that fits in the time

You will not have time to re-read every record. Instead, check the first and last character of each code before you press Enter. Those two positions carry most of the errors, and the check takes well under a second.

Group your reading

Read INV48213-KA-07 as "INV, 482, 13, KA, 07". Chunks are easier to hold in memory than a string of twelve characters, and chunking is what separates 6,000 KPH from 9,000.

Practise on the alphanumeric and invoice formats and look at which records you got wrong on the result page. The pattern is usually obvious after three attempts, and it is almost always one of the four groups above.