Personalized communications increase response rates and conversions. The key to success with such messaging is accurate data. Even a small amount of data can make a difference. With nothing more than a name and address it is possible to speak personally to a recipient and include local flavor in the message and illustrations. Adding seasonal influences in localized photos, colors, and design changes a campaign from generic to personal.
Once brand owners decide on campaign concepts, there are a few things to know about data files that will make projects process more efficiently. Conversely, some situations can add to project costs or affect performance. Personalization only works when data is accurate and trustworthy. Follow the guidelines below to help projects flow swiftly and flawlessly to recipient mailboxes.
DATA FILE PREPARATION GUIDELINES
DO submit data files in a consistent format. Data files allow us to identify different parts (data fields) for all the recipients in the list. This is generally accomplished with delimiters. Delimiters are special characters that separate the data fields from each other. This is quite different from a plain text file which has no special delimiters, only spaces separating the data. Though the data appears formatted into columns or tables in a word processor, text files are generally unusable for data. Preferred formats are tab-delimited or CSV (comma separated values) files. Sending data in an Excel spreadsheet is perfectly fine.
DO include important information about the data structure (how many fields and the column headers) and the number of records, along with submitted data files. This information eliminates guesswork and positively communicates the intended layout and size of the file, resulting in accurately rendered personalized messaging.
DO have someone who is familiar with the data review the file before submitting it. Every data record should start on a new line. Scan through the file to ensure data is aligned with the appropriate column headers. Each data record should contain the same number of fields, each field separated by a delimiter.
Choosing an appropriate delimiter is important. Though commas are the most common delimiters, some data fields in a file may already contain commas. Company names for example, often feature a comma just before “Inc.”. Imbedded commas may be interpreted as delimiters causing data to be misaligned. It is usually safer to use a more unique character as a delimiter such as a tab or vertical pipe (|) not likely to be present in the data itself.
Spreadsheet programs may strip leading zeros from fields, so always inspect number fields such as membership ID’s, zip codes and phone numbers. It may be necessary to change the field type to “text” and export again to prevent this truncation.
DO Provide your print vendor with match criteria. If using multiple data sources, use distinctive values to match records such as account numbers to ensure the proper match. Always use something unique. Names or property addresses are poor choices for match keys as files may contain more than one possible match.
DO inform your print vendor in advance if a previously-used file format has changed. Data structure is critical to accurate personalization, so if anything has changed please let them know.
DO export a display field rather than an index field. Database and CRM index fields can contain values such as “Hartford, The” when you really want to print “The Hartford”.
DON’T send name and address data formatted as labels (such as three across).
DON’T mix name formats in the same file. Files with some records as ”last name first” and some with “last name last” may be difficult or impossible to normalize. Pay close attention to suffixes (Jr, Sr, etc.) and professional designations (Dr., Rev, etc.). CRM’s and databases treat name elements differently for internal alphabetizing and search.
DON’T allow comments to integrate with real data. Customer service agents have been known to enter unprofessional notations into secondary address lines or middle name fields as warnings to future CSR’s. Slurs and insults could be unwittingly printed and delivered to your customers.
DON’T plan for text, graphics, or salutations based on the sex of the individual addressee unless data accuracy is assured and a gender field is present in the data. Genderizing software is never guaranteed to be 100% accurate. Many given names are ambiguous.
DON’T use a single CSV file to transmit data with a “one to many” relationship, such as a list of transactions, accounts, or property addresses that can vary in number and must be printed on the same document. In most cases, you must separate this type of data into multiple files. Consult with your print vendor for the best solution.
DON’T send composed documents, such as mail-merged Microsoft Word, PDF, Postscript, PCL, or other types of print-image files.
Do you have questions about how to prepare data files? Please call us! We’ll be happy to advise you on approaches to keep your costs low and maximize your results.