Be wary of subpar data & subpar advice

October 1, 2004

Johnson’s World by Steve Johnson

“Dear.”

That was all there was. By convention, “dear” usually is followed by the name or title of the person to whom a letter is addressed. I marveled at this bold departure from etiquette. Donning my Sherlock Holmes deer-stalker cap, I weighed the possibilities. My first thought was that the letter must be from my wife, to the best of my recollection the only person who calls me “dear.” However, she stopped writing me letters after high school, so I knew she was not the perpetrator.

Who, then? In a flash of inspiration, I checked the signature line. I am quite certain the lady who wrote the letter has never, ever called me “dear,” and if she ever harbored romantic feelings for me, she has done a masterful job of concealing them.

Okay, enough of the gothic story line. The letter was part of a digitally printed mass mailing that incorporated variable information. It employed a simple database and the most basic of applications (probably Microsoft Word) to merge variable and static data.

The address at the top of the letter told the rest of the story. My company's name, street address, city, state and ZIP code were correct. Conspicuously absent, however, were my name and title. Since the “name” field in their database must have been empty, there was nothing variable to print in the salutation. MS Word thus automatically moved the static “Dear” and colon next to one another to eliminate awkward empty space.

The database rules

The database is king in variable printing. I say this often, and it may be the only area in which I agree completely with my fellow pundits. If the database is not up to snuff, the printing doesn't matter. Never mind how sharp your dots look or how accurately your color matches — if the database is poor, the variable printing also will be shoddy.

Here's the good news: Develop some expertise in the real world of variable printing, and you can help your clients salvage their jobs. My company often lands print jobs because we are the only ones who can output defective files. If you know how to manipulate a less-than-perfect database, you will gain a great edge over competitors who may have expensive variable software but fold when things get out of the ordinary.

In this example, the solution was simple. Conventional print files routinely are preflighted; a database should be also. A cursory glance would have revealed that the name field was empty in some records. If a complete database was not available, then a simple substitution command should have been employed. When the “name” field came up empty in a particular record, a static placeholder (such as “Print Professional” or “Graphic Arts Leader”) should have been programmed to insert. This would have resulted in “Dear Print Professional:” appearing at the top of my letter while “Dear Mr. Smith:” would have printed on a page generated from a complete record.

These holes happen all the time. Careless use of subpar data in variable printing is endemic to this fledgling segment of our industry. But the irony of this particular letter also sheds light on a larger issue.

Easier said than done

The no-named letter I'm referring to was sent from one of the country's most prestigious graphic-arts educational institutions. The person whose signature appeared on the letter is respected as an authority on digital and variable printing and has published extensively on the implementation of complex variable-printing strategies. In fact, this person has written about little else for the past decade.

It is ironic (but not surprising) that the topic of the letter was variable-data printing. The letter's text and attachments made repeated references to technology, machine prices, press models and software brands. Barely mentioned were client industry segments, and our customers' thoughts, needs, ideas, opinions and requirements weren't alluded to at all.

The harsh reality is that it's much easier to write about variable printing than it is to actually print something correctly. The people behind this letter have written books on the subject. Yet their letter, using the most basic form of variable data, was not executed correctly.

Variable printing is certainly worth investigating if you are not yet involved. But be careful who you take advice from. Take everything with a grain of salt and a healthy dose of skepticism. Do your homework. Be sure to verify proffered claims, anecdotes and statistics. If something sounds too good to be true, it probably is.

Now I have to go. I have a variable job to print.