85 Auriga Drive, Nepean ON K2E 7Z2
Direct Mail Data Processing & Merge-Purge Slider

Bells & Whistles

JSI offers a very large number of processing options that can be used to reduce your mail volume and/or improve targeting for your direct mail projects. And, when we enable these options, we can measure and report to you on their impact so that you can evaluate whether or not to repeat them.

Eliminate duplicates.

JSI merge-purge processing offers effective, and more-and less-aggressive methods for eliminating duplicates from files. We offer varying options to define a "duplicate" and we advise on which option is best given the quality of records in the source files that you use. See Merge-Purge Processing for details.

Quality control addresses.

Running address correction is an effective way of quality controlling addresses and improving the quality of poorly keyed records. We can return files of corrected addresses to you to be uploaded to your system and we routinely run address correction as part of our merge-purge processing. Accuracy Statements generated by JSI Address Correction can be used to demonstrate to Canada Post™ that your files meet minimum accuracy standards. See Address Correction for details.

CMA List Exclusions.

The Canadian Marketing Association maintains a list of people who don't want to receive unsolicited mail. When you ask for exclusions based on this list, we incorporate it as a "kill" list in a merge-purge database.

Eliminate matches to the "second name" where two names are included in name lines.

In a merge-purge, we often receive records containing two names (e.g. "Mr. A Brown & Ms B Jean White"). We construct a primary match key using the second of these names so that other records addressed to Ms White will be identified as duplicates. However, we can also construct a second match key using the first of these names. This second key will eliminate records addressed to Mr. Brown at the address for Brown/White.

Reduce a mailing file to one record per household.

Many households contain individuals whose names are dissimilar and who therefore are not recognized as duplicates when names/addresses are compared. For urban records, we can reduce your direct mail files to one per household, and we can "control" which of the records belonging to the household is mailed. Or, if you prefer we can limit to two records (or more) per household.

Eliminate addresses that are already on your house list.

If you want to prevent prospect mailings from going to someone who resides with a person already on your house list, we can run an address elimination. For urban records, we can compare your merge-purge output file to your house list using an address key to eliminate occurrences of house-list addresses in the output file.

Eliminate records that are missing a unit number.

Files of names and addresses often contain records that are missing unit numbers. In large apartment buildings, direct mail that lacks unit numbers ends up on the floor of the lobby. We can detect many of the records that are missing unit numbers and eliminate them.

Detect records addressed to corporations and organizations.

If you want to reach individuals and not companies, we can eliminate records whose name line contains the name of a corporation or organization.

Avoid (fund) raising from the deceased.

It is not unusual for us to receive donor records that are addressed "in trust" or "to the estate of." The records are created when bequests are made in wills and the receiving organization accidentally includes these records in files that they exchange. We routinely enable processing that eliminates these records from donor prospect mailings.

Apply forwarding addresses for people who have moved.

JSI’s NCOA™ (National Change of Address) software applies forwarding addresses for people who have moved and identifies people who have moved but who have declined supplying a forwarding address. Our NCOA™ software uses data supplied by people who have recorded change-of-address forms with Canada Post™ Corporation, data licensed by JSI.

Eliminate records that lack a name.

We can check name lines and eliminate records with blank name lines.

Smart Placement of Letter Carrier Presort (LCP) Data.

When we LCP sequence a file, walk codes, date check letters and container/bundle markers are attached to records. JSI has sophisticated placement software that examines addresses and "places" this information in the locations in the address where the information will "best fit" given Canada Post™ rules for placement of LCP data. We can simplify "downstream" imaging by supplying output files in which LCP data is already embedded in the addresses. Alternatively, we can supply the required information in "fixed fields" for later placement.

Eliminate “LCP Invalids”.

LCP (Letter Carrier Presort) invalids are records whose postal codesOM do not exist in the Canada Post™ files of postal codesOM that JSI uses to sequence records for Admail™ and Publications Mail™. These records don’t qualify for incentive rates and, if mailed, they are surcharged. If you want to mail these records, we can identify them and return them to you so that you can attempt to correct them. Or we can eliminate them from your mailing.

Target your mail to the postal walks in which you have greatest success.

Using your house files, we can identify the postal walks in which you are most and least successful. We can then use this information to filter direct-mail names and addresses selecting those that are located in the most responsive areas. In addition, if you are planning Unaddressed Admail™, we can use this approach to identify more and less productive postal walks. See Density Analysis & Unaddressed Mail for details.

Target by type of address.

You may already target by renting or trading lists that have the greatest "affinity" to your endeavour. We offer other ways of targeting based on address characteristics. You may find that your prospecting yields very different results between urban and rural records, or between houses and apartments, or between people whose mail is delivered to a street-location and those who pick up their mail from a box in a post office. We can help you target by analyzing your responses by type of address to help you select more responsive records.

Targeting records to improve response.

Response rates can be improved by the selective targeting of records. Merge-purge processing itself generates information that can be used to differentiate records; the match rate between the source list for the record and the house suppression list, the number of times a record has matched to other records, the type of list (house, rented, traded), whether the person resides in a house or apartment, whether the address is urban or rural. Data processing can add additional information; assigned gender, the probability of a record being within a given age-range, the donor density of the postal walk in which the record resides, the number of times a record has previously been mailed. Evaluation of response rates based on these record properties then allows the ranking of records based on expected response, and the elimination of those predicted to have relatively low response rates. See Applying Analytics to Prospect Mailing Lists for details.


JSI’s name parsing software allows us to break "name lines" into their component pieces. It identifies and populates six separate fields: salutation (Mr, Mrs, Ms, Dr, etc), first name/initial, middle name/initial, surname, suffix (e.g. Sr, Jr, III), and title (e.g. RN, PhD). We can supply these components as separate fields in output and the components can be used to build "personalization" fields to begin a letter. Personalization fields can be built based on options that you define. For example, your preferred form of personalization may be "Dear Mr Brown." Under this circumstance we will build the personalization field to include "Mr Brown" (salutation+surname) wherever a salutation (Mr, Mrs, Ms) is present. For records where a salutation is not present we will "default" the personalization field to some other combination of the name fields (e.g. the full name) that you define or to a standard greeting (e.g. Dear Friend). This processing will also flag those name lines that do not conveniently fit a "parsed" data structure ("Mr Anthony Brown & Ms B White", "JSI Data Systems Limited") and personalize the exceptions with an agreed default. Alternatively, we can personalize using first names, defaulting to the full name in the absence of a first name. See Name Parsing and Personalization for details.

Segmentation and panel assignment.

We use several sophisticated programs to allocate records to mail panels or to segment them based on their histories or properties. All of these programs offer high levels of control. For example, if you ask for random assignment to "test" and "control" panels, the resulting panels will all be statistically representative of the records that are eligible for the panels. We can split output into panels containing numbers of records that you preset, or based on proportions of total output, or using more complicated rules. We routinely execute complicated "output designs" that result in multiple test/control panels and that use a variety of criteria to assign records to panels. Criteria for assigning records to panels may include their source lists, gender, geographic criteria, other values in fields in input records, "random assignment," or some combination of these criteria.

Source Coding.

We can attach a source code to each record indicating the list from which it came and, if you are dividing the mailing into panels, we can code each record with a value indicating the output panel to which the record is assigned. You have flexibility on how you design these source and panel codes.

Gender Assignment.

Our gender assignment software uses titles (Mr, Mrs, Ms, etc.) and a database of first names to assign gender. Success rates depend on the incidence of titles and first names on the lists supplied. You can use gender assignment to create panels that contain people of one gender only, or to analyze list composition by gender, or to "track" returns by gender. We recommend that you do not use assigned genders in personalization (for example to address males as "Dear Mr. ….").

Upper/Lower Conversion or Upshift.

If some or all of your rented lists are upper case, we can convert so that the names and addresses appear in lower case letters with each first letter or initial capitalized. We can also convert all text to upper case if you prefer. Our upper/lower routines are "intelligent" and will handle names and address elements appropriately (McRae, not Mcrae; RR not Rr; PO Box, not Po Box; etc.).

Unique Sequence Numbers.

We can apply unique sequence numbers to output to assist your letter-shop in matching personalized letters to envelopes or to assist in "stop/start" recovery during name/address imaging or insertion.

Finders Files.

We can supply a "finders file" or "enrollment file" that contains records mailed, each with a unique I.D. number. By printing this I.D. number on your reply forms, you (or your data-entry supplier) can use the I.D. number to "recover" the name/address of the responder from the finders file. This approach saves on data-entry costs. To eliminate mis-keyed I.D. numbers we can also compute and add a "check digit" that can be used to quality control the data entry of the I.D. If your files are used to support inbound calling and your data contains phone numbers, we can supply a finders file that is unique with respect to phone numbers and that can be used to "recover" name/address records using inbound phone numbers.

Non-response kills.

If you prospect to rented or traded files, you may receive the same names and addresses repeatedly and mail to some records that never respond. We can build and maintain special databases that encode names/addresses and record the number of times that you have mailed to each of them. Then, we can use this database to eliminate those who have a history of not responding to your direct mail.

Phone number standardization.

If telemarketing is an integral part of your marketing effort or if you are testing its usefulness, we can quality control records allocated for phone calls. Our phone standardization software inspects phone numbers, formats them so that they are consistently presented, and qualifies them as having acceptable or unacceptable formats.

Special Reports.

In addition to our normal situation reports on "net names" available and our final report on sources of names used, we can provide special reports on any of the fields we create in the database. For example, if you are dividing the mailing into "panels", or if you want a breakdown by province, or if you have asked us to run Address Accuracy software and tag records as "valid" or "invalid", we can provide counts.

Post Analysis of Sales.

If you supply us with copies of your returns, we will match them to your merge-purge database and provide a report that tells you which lists were most and least responsive.

Admail, Publications Mail, and NCOA are Canada Post trademarks. Postal Code is an official mark of Canada Post Corporation.