Mail merge and printing

Mail merge is the process of combining a standard document with a list of data to produce many personalised copies of that document — one for each record in the list — in a single operation.

$$\text{Main document}+\text{Data source}⟶\text{one merged document per record}$$

The three components.

1. Main document. The document containing the text that is the same in every copy — the body of a letter, the layout of a label, the wording of a certificate — together with merge fields marking the places where varying data is to appear. It is written and proofread once.

2. Data source. A table holding the data that varies between copies. Its first row contains the field names (Name, Address, RollNo, Marks) and each following row is one record, that is one recipient. It may be a Word table, an Excel worksheet (the usual choice), an Access database, Outlook contacts, or a delimited text file.

3. Merge field. A placeholder inserted into the main document, displayed as «Name», which is replaced during the merge by the corresponding value from the current record. Fields must be inserted with Insert Merge Field; typing the angle brackets by hand produces ordinary text that never merges.

Applications. Circular letters to students or customers, address labels, envelopes, certificates, mark lists, salary slips, and personalised bulk email.

Both settings concern the order in which pages emerge when more than one copy of a multi-page document is printed.

Take 3 copies of a 4-page document.

Collated — each complete copy is printed in turn:

$$1,2,3,4|1,2,3,4|1,2,3,4$$

Three finished documents come out, each in the right order and ready to staple.

Uncollated — all copies of each page are printed together:

$$1,1,1|2,2,2|3,3,3|4,4,4$$

Four stacks emerge, and the pages must be sorted by hand into three documents.

When collated is wanted — almost always for documents to be read or bound: reports, project submissions, question papers, handouts. The printer does the sorting.

When uncollated is wanted — when the pages are to be separated rather than kept together: printing 40 copies of a single-page notice for distribution, or a set of forms where all copies of each form go to a different office. Uncollated is also slightly faster on some printers, since the page image is sent once and printed repeatedly.

A note on single-page documents: the setting makes no difference at all, which is why it goes unnoticed until someone prints thirty copies of a booklet uncollated and spends an afternoon sorting.

Step 1 — Prepare the data source. In Excel create a worksheet with the field names in the first row — Name, Address1, City, PIN, Amount — and one customer per row for all 100. Requirements: no blank rows, no title line above the headings, no merged cells, and consistent data in each column. Save and close the workbook.

Step 2 — Create the main document. In Word, type the letter with the text common to every customer, leaving the personalised parts empty for now.

Step 3 — Start the merge. Mailings → Start Mail Merge → Letters.

Step 4 — Select the recipients. Mailings → Select Recipients → Use an Existing List. Choose the Excel file, then the worksheet, and confirm that the first row contains column headers.

Step 5 — Insert the merge fields. Place the cursor where each item belongs and use Insert Merge Field to place «Name», «Address1», «City», «PIN» and «Amount». Address Block and Greeting Line insert a conventionally formatted address and salutation in one step, and Address Block suppresses lines that are empty.

Step 6 — Refine the list if required. Edit Recipient List allows records to be filtered (only customers of one city), sorted (by PIN, for postal discount), or individually unticked.

Step 7 — Add rules if the wording must vary. Rules → If…Then…Else can insert "Sir" or "Madam" from a Gender field, or a different closing paragraph depending on the amount.

Step 8 — Preview. Preview Results and step through the records. Check the first, the last, and any with unusual data — a very long name, a missing address line, an amount that appears as 1500.000000 (which needs a numeric formatting switch).

Step 9 — Complete the merge. Finish & Merge, then either:

• Edit Individual Documents — creates one document holding all 100 letters, which can be checked, corrected and saved as a record before printing
• Print Documents — prints directly
• Send E-mail Messages — emails each customer, using an email-address field

Step 10 — Save both source files. Keep the main document and the Excel list. The next letter reuses the same recipient list, and the merge can be run again at any time.

Spooling — from Simultaneous Peripheral Operation On-Line — is the technique of placing output intended for a slow device into a temporary file on disk, and then feeding it to the device at whatever rate the device can accept.

Why it is needed. A printer is enormously slower than a processor. Printing a page may take several seconds; in that time the CPU could execute billions of instructions. Without spooling the program would have to hand over each line of output and wait for the printer to accept it, so the application — and the user — would be blocked for the whole duration of the print.

What happens instead. The application writes the entire print job to a spool file on disk as fast as the disk will take it, which is a fraction of a second. It is then finished and the user regains control immediately. A separate part of the operating system, the print spooler service, reads that file and sends it to the printer in the background at the printer's own pace.

The print queue. Because jobs are held on disk, several can wait at once, and the queue is the list of them. From it a job may be:

• paused and later resumed
• restarted from the beginning
• cancelled and removed
• reordered by priority

The queue also allows one printer to serve many users on a network, each job being printed to completion in turn rather than the outputs becoming interleaved.

Why the computer does not lock up. Three things together: the transfer to disk is fast; the spooler runs as a background process so the CPU continues with foreground work; and the operating system's multitasking shares the processor between the spooler and the application. The user can close the document, or even the program, and printing continues — the job no longer depends on the application, because the output is already safely on disk.

A practical consequence. Cancelling a job from the queue may not stop printing at once, because pages already transferred are held in the printer's own memory and will finish. And if the spool folder's disk fills up, printing fails with an error that appears to be a printer fault but is really a disk-space problem.

Two independent faults, with different causes.

Fault 1 — «Name» appears instead of the data.

There are three possible causes, and they are distinguished by what the document does when Preview Results is toggled.

(a) The field was typed rather than inserted. The characters « and » can be typed from the keyboard or pasted, producing text that looks exactly like a merge field but is ordinary text. Word has no reason to replace it, so it prints literally.
Test: switch Preview Results on. A genuine field changes to data; typed text does not.
Remedy: delete the typed text and insert the field properly with Mailings → Insert Merge Field.

(b) Preview Results is simply switched off. In this state Word deliberately shows the field names so you can see the structure — and if the merge is completed from here the output is still correct.
Test: look at the merged output rather than the main document.
Remedy: switch Preview Results on; nothing else is wrong.

(c) Field codes are being displayed rather than their results, which shows as { MERGEFIELD Name }.
Remedy: press Alt+F9 to toggle field codes off.

A fourth possibility if the data genuinely fails to arrive: the data source is not attached, or the field name in the document no longer matches a column heading in the source — for instance the Excel heading was changed from Name to Student Name after the field was inserted. Edit Recipient List shows whether the source is connected and what its columns are called; Match Fields repairs a mismatch.

Fault 2 — blank lines in the addresses.

The cause is empty fields in the data source. A merge field with no value produces nothing, but the paragraph mark or line break you typed after it in the main document remains — so a customer with no second address line still gets that line's worth of empty space, and the city and PIN are pushed down.

$$\text{Address1}\to \text{Address2 (empty)}\to \text{City}⟹\text{a gap where Address2 should be}$$

Three remedies, best first.

1. Use Address Block instead of inserting the address fields one by one. It is designed for exactly this: it assembles the address and suppresses any line whose fields are all empty. This is the correct solution and needs no per-record work.

2. Use an If…Then…Else rule where a specific line must be conditional: Rules → If…Then…Else — if Address2 is not blank, insert Address2 and a new line, otherwise insert nothing.

3. Fill the gaps in the data, if the values genuinely exist and were simply omitted. Never "fix" it by putting a space in the empty cells — Word then treats the field as non-empty and the blank line returns.

The general lesson. A merge fault is almost always in one of two places: the link between document and data (fields not inserted, source not attached, names not matching), or the shape of the data (empty cells, blank rows, headings not in row 1). Preview Results, stepping through the first and last records and one deliberately incomplete record, catches nearly all of them before anything is printed.