Recently, I read a memo written by copywriting legend Gary Bencivenga to ETR’s own marketing expert Bob Bly. Here’s what he said: “As you know, you need strong copy to sell anything. But once I started writing my packages as magalogs and bookalogs (see “It’s Good to Know,” below), I found that a good rule of thumb was my ‘ABA’ rule.

I only half-jokingly told my clients that my copywriting style had become to write packages that resemble ‘Anything But Advertising.’ “I came to this ‘rule’ because I had always made it a practice to write two versions of a package, or at least try two radically different designs. “Every time I wrote a package in a magalog, bookalog, or newsletter look-alike format (i.e., one resembling and actually containing valuable editorial matter), that package would trounce the same exact copy in a traditional #10 (business-sized envelope) package. “So, in time, I stopped writing #10 and 6 x 9 envelope packages (since they always lost).

Yet, I still wanted to test two different packages on each major test . . . so instead of writing different packages, I started testing radically different headlines within the same editorial-style design. “And this yielded an amazing revelation. Over at least a dozen packages . . . radically different headlines (but with the same interior body copy) made no difference at all! All my clients told me that similar tests on other writers’ work yielded the same surprising result.

Well-written but radically different headlines on magalogs almost never made any appreciable difference in response. “This completely called into question the Ogilvy dictum that ‘your headline is 80% of the sale’ (although he said that referring to space ads in magazines and newspapers). In direct mail, all our tests have shown that with these editorial-style formats, the format, not the headline, drives the sale. “I came to realize this: In direct mail, your format IS your headline.” I’ve been thinking for about a month about what Gary said in that memo and have been comparing his experience with my own.

I have the advantage of having as my major client a marketer that tests hundreds of promotions a year. In the 12 or 13 years I’ve been working with them, I’ve seen the results of almost 1,000 tests. And I have seen headlines make a difference. Just recently, for example, a headline test for a financial newsletter doubled response rates. But this was for a headline that sat on top of a promotion that looked and smelled like a sales letter. I’ve tried to remember headline breakthroughs on “advertorial” formats but can’t recall anything significant.

In other words, I think Gary is right. The headline itself is unlikely to change response rates by any considerable degree when . . . the format is of an editorial nature, and the body copy is exactly the same In the copywriting program I helped design, I wrote a chapter about headlines and leads. In that chapter, I stated that the lead (which includes the headline, subheads, and everything in the body copy until you get to the point where the reader will think, “Oh, now I get it!) is 80% of the game.

In considering Gary’s astute observations about headlines, I’m glad I put it that way. Leads DO make a difference, even in advertising that has the look and feel of editorial. Over the time I’ve been actively testing leads (maybe 15 years), I’ve seen dozens of cases in which one lead trounced another. Increases of 50% to 75% are commonplace. Doubles and triples are not rare. It makes sense that when you’re dealing with advertorial promotions, heads don’t matter but leads do.

Here’s why: When you create a promotion that looks like editorial, most people are going to read beyond the headline simply because they see the promotion as serious and possibly beneficial. If they aren’t turned on by the headline, that’s OK. They’ll see what the copy looks like. Changing the lead can change everything — the promise, the point of view, the benefits, even the tone. If one lead isn’t working, it’s perfectly possible that another will.

Likewise, changing a headline on a promotion that looks like advertising SHOULD make a difference, because when people read something that is clearly advertising, they read it with a jaundiced eye. They scan the headline as if to say, “What kind of crap are they pushing here?” If the headline isn’t compelling, they’ll pass.

So I recommend Gary’s advice to you with the following modifications: With any sort of promotion, test the lead. The lead encompasses the essential sprit (the message) of the promotion. If you don’t get one lead right, you may do better with another one. With promotions that look like advertising, test headlines . . . because they WILL make a difference. With promotions that look like editorial, test formats rather than headlines.