It is important to understand how AutoMatch is applied in TMS during pre-processing. Here is the process:
- Firstly, take the segment and look if a 100% match can be found.
- If no 100% match can be found, perhaps a fuzzy match may exist? If the number in your segment is three digits or less, it is unlikely that a fuzzy match will be found.
- If we do have a fuzzy match, can it be improved by the use of Auto Match?
If, after step two no match is found, then the segment remains as a "new" word. Automatch is not implemented against these segments - only against fuzzy matches.