85 Auriga Drive, Nepean ON K2E 7Z2
Name Parsing and Personalization Chart

Name Parsing and Personalization

Personalize your message.

If your direct marketing piece includes a letter or a response card you may want to personalize it. One of the first things recipients see is their name, and no word is more important than a person’s name. Let our Name Parsing and Personalization software help you speak to people in a way that will engage them.

Names and Lists

Most of the lists that JSI receives arrive with a "name" field. JSI stores most of the lists that it houses in the same way. A "name field" implies that all the component parts of a name appear in a single field in the order in which they would be produced in the name portion of an address. Storing names in an "open" name field is a common industry practice especially among magazine mailers and cataloguers. It allows for variability in the information stored which may contain varying combinations of salutations, first names or initials, middle names or initials, dual names, surnames, company names, suffixes and titles.

Here are some examples of the data that we might encounter in a name line:

MR. A. H. BROWN
ANTHONY H. BROWN
MR. A. BROWN & MS B. J. WHITE
MR. & MRS. BROWN
ANTHONY & BARBARA KATHY WHITE
ANTHONY BROWN JR
ANTHONY H. BROWN III
MR. A. H. BROWN, PRESIDENT
JSI DATA SYSTEMS LIMITED

Many organizations elect to store names in an open field in order to simplify data entry and to avoid operator decisions on the allocation of "pieces" to fields when names are originally entered. JSI can parse elements of the name line from supplied data.

Inspecting and Parsing the Name Line

The name line is inspected and those name lines that can be successfully parsed are parsed to six fields. Where no "piece" fits the field, the field is left blank. The six fields are:

  1. SALUTATION (MR, MRS, MS, MME, DR, COL, etc.). Note that M. (the abbreviation in French for Monsieur) can be parsed as a salutation only if a language field is available and the record can be identified as having a language preference of French. Otherwise, the salutation “M” is indistinguishable from the initial "M".
  2. FIRST NAME Contains first name or initial of first name if only an initial is provided.
  3. MIDDLE NAME Contains all middle names and/or initials between FIRST NAME and SURNAME.
  4. SURNAME
  5. SUFFIX (JR, SR, II, III, PhD, RN, etc.)
  6. TITLE (President, Manager, etc.)

Certain name lines cannot be successfully parsed within the layout described above. For such name lines, "default" parsing is used to allocate selected pieces of the line to particular fields while "remainders" that cannot be parsed are allocated to other fields by default.

Parsed names have a variety of uses. The "pieces" can be used for certain forms of personalization. For example, you may want to address those who can be so addressed by SALUTATION and SURNAME ("Dear Mr. Brown") or you may want to address those who can be so addressed by FIRST NAME ("Dear Anthony"). Parsed data can also be used to preferentially pick certain parts of the name line for insertion in variable text ("Yes, Anthony, you can receive your free..."). To support selection of records that meet your criteria, we flag and count records during processing.

What We Flag

Here are the flags that are applied:

  • Dual names, defined as any records that have AND, &, or more than one salutation, are flagged as "dual." This flag also allows us to recognize dual names and to recognize that FIRST NAME in the flagged record contains data other than a conventional first name.
  • Company records (if they have not been dropped) are flagged as such, again allowing us to "special case" these records.
  • Records that contain only one word (ordinarily recognized as a SURNAME) and no other words in the name line are flagged as such. Again this allows us to exclude these records later or to special case them.
  • Records that contain no words in the name line or that contain words none of which "look like" names (e.g. "MR L.") are flagged as having "no name" so that they can be excluded or handled appropriately.
  • Records are flagged as having "valid first names." To be flagged as "valid first name" a record must meet two conditions: (a) It must contain more than two characters in FIRST NAME, and (b) FIRST NAME must not contain "defaulted" data from a line that could not be parsed. The threshold of two characters is set because many records come to us with two initials that are presented without a space leading us to allocate the two initials to the FIRST NAME field: e.g. AH SMITH
  • Records that do not qualify as having a valid first name are inspected for a valid second name. This allows us to recognize names that are presented as an initial followed by a given name: e.g. F. JOHN SMITH. Records that show this condition are flagged as having a valid name in the MIDDLE NAME field.
  • Records that contain a salutation (MR, MRS etc.) are flagged as containing a salutation.

When we run name parsing in conjunction with merge-purge processing, we accumulate counts of the flags applied as we process records. We can then use these flags to select those records that will support personalization or other special processing of names.

Name Parsing and Merge-Purge

Where name parsing is being undertaken in conjunction with merge-purge processing, we need to know your requirements before we begin. We will then ordinarily parse all name lines as we process records during the merge-purge. At the end of the merge-purge, we will then select names or allocate names to panels based on your requirements and using name-pieces and flags that we have applied during processing. We can also provide counts (for example, the number of names with either a valid first or middle name) along the way. The specific fields supplied at the end of this processing will depend on your requirements. For example, we might supply a full name line for use in a mailing address plus a "personalization field" that is populated according to your requirements.

We can also perform name parsing and personalization on supplied files independent of merge-purges.

A note on salutations and language

The French salutation "M." (Monsieur) is indistinguishable in a name line from the first initial "M" (which might imply a first name of Michel or Michael, or Miriam). When parsing names, we can enable processing that inspects a language description or language code and treats "M" as a salutation in a name line where the language of the record is supplied as "French." Obviously, to enable this processing, we must be supplied with a language field and it must be fully or near-fully populated. We advise use of this approach only if all or most of the records that we will be handling and matching contain a supplied language descriptor. In the absence of a language descriptor, we default to treating "M" as an initial.

Consistency in the handling of “M” across lists is important because otherwise the same record will be handled differently if one edition of the record contains a language value and the other does not. For example, suppose M. Nadeau appears on one list with a language of "F" and on another list where language values are absent. If we enable language-based recognition of the salutation on one list but cannot do it on the other, we will treat the "M" from the first list as "Monsieur" whereas it will be seen as an initial on the second list. This may lead to us failing to recognize these records as duplicates. If your job requires name parsing and involves records where "M" may be used as the abbreviation for Monsieur, we recommend that you review how we are to handle these records before we begin.