He shows a series of posters for a ski resort, with dramatic photos of snowboarders upside-down in mid-leap. The headlines, all caps, appear upside-down at the bottom of the page. This, he alleges, conveys "…the subliminal message that there are no limits at the resort."
Quite a feat, considering there's no such thing as subliminal advertising techniques that have any effect. But let's be fair; maybe he meant they convey that impression. Fine. Exactly how does he know that anyone besides him is getting the no-limits message from his upside-down type? Saying “this is what happens” doesn’t make it so.
Oh, and by the way, setting a line of type in all caps reduces readership. So does putting it upside down. I know this from testing, not speculation. It's a lesser sin for short headlines, which these are. But ... subliminal? Please.
Next he shows a shopping bag for a lingerie store. On the bag is an admittedly cool illustration of a bra, with copy set in calligraphy of varying sizes swirling about to form the cups. He says it tells "…an evocative story in an engaging manner." Evocative the story may be. I wouldn't know. I was unwilling to rotate the bag (or myself) to read the message.
It doesn't matter how cool the type looks if no one can read it. If you turn it, swirl it, reverse it, set it in caps, use goofy fonts, erratically change fonts or sizes or shape it into a bra, no one will read it. It's too much work — and no one cares enough about what an advertiser has to say to work for it.
Do not think that by making text challenging to read you will motivate people to dig in and decipher it. That would be like saying diners will be more motivated to eat a meal if you make it smell bad.
Right after I fired him, one designer hurled this accusation at me: "You refuse to admit that design is the most important part of advertising."
Guilty as charged.