|
Although it is an intelligent program, dedupeIT is not as good
as human beings at inspecting data to determine what the different elements
are. If it gets any field names wrong, you can correct them e.g. by right
clicking on the column, selecting Rename field and choosing the correct field
name from the list available.
The list of fields shown contains all the common data items that
are common in name and address files. However, dedupeIT only uses name (or
company name), address and zip/postcode in deduping, so a data item is not
covered by the dropdown list of fields, simply select a field name of Other. If
you need to dedupe using a different item of data e.g. an account number or
date of birth, you can buy a more functional product from the helpIT systems'
range, matchIT, to do this. For more information, visit www.helpit.com or
Access databases can contain several tables and queries. If
there is more than one, dedupeIT will ask you which table or query you want to dedupe.
Excel files can contain several worksheets. If there is more than one, dedupeIT will ask you which worksheet you want to dedupe.
For fixed width text files, dedupeIT sometimes fails to identify
where one field ends and the next field begins. If you need to split a field
into two, you can right-click on the field in question, and choose Split
Fields. Similarly, you can Combine Fields by right clicking on one of the
fields you want to combine.
This is probably because they are like this in your input file.
If the input file is correct, then it may be that dedupeIT has not properly
loaded the records in at the Import stage. If so, try saving the file in a
different format e.g. if it is an Excel file, try saving it as a tab delimited
file then opening that file in dedupeIT. Alternatively, a sample of the
data.
For residential data:
- Person (contact): This gives one record per person in your deduped file.
-
Family (last name): This gives one record per last name (surname) in your
result file. It is a looser setting than Person level, as it ignores the first
names or initials of the names.
-
Household (address): This gives one record per address in your result file. It
is the loosest of the matching levels, as it ignores the names completely.
For business data:
- Person (contact): This is the same as for residential data - it gives one
record per person working at the company. When matching at person level, the
company name is ignored.
- Company: This gives one record per company at the same address. If you are
matching at company level, the contact name is ignored.
|
