As if the advertising industry didn’t have enough problems, now we need to show more  and more demonstrable results. There’s little accountability for our government, our banks, car makers, or mortgage brokers—but damnit, the ad industry is asked to show some, or else.

Why? Our clients want proof that advertising works.

But here’s the not-so-well-kept secret: No one knows if advertising works.

Actually, in its most base form, we know that advertising works. You can’t just open a business and not tell anyone about it, hoping they’ll show up. When you tell the world to do business with you, somebody will. So yes, advertising works.

We just don’t know how it works. Which is even worse.

Famously, John Wanamaker said, “Half the money I spend on advertising is wasted; the trouble is I don't know which half.” His department store is long gone, but that truism lives on.

Personally, I think Comcast wastes more than half. Comcast spends hundreds of millions of dollars on advertising. I’m a customer, but still they bombard me. I get direct mail from them every week—and last week, I found two different postcards on top of each other in the pile. They’re on TV,  radio, the internet, etc. There’s even a “ComcastCares” Twitter guy who responded to my customer service inquiry. Comcast is flooding the zone. So far, they haven’t tempted me with any new offers.

But let’s say I finally respond to a postcard. That one mailer gets all the credit. Boom! There’s your ROI--it must have been that shade of green on the postcard or that magic promotional offer that hooked me. Right?

Number crunchers in advertising are looking for pinpoint accuracy. Someone figured out it takes several viewings before a direct response TV spot produces a sale. But if more people order a Slap Chop during 3AM reruns of “Iron Chef,” well then they know where the media dollars need to go. Never mind if anyone saw it on YouTube first.

With the advent of new media, the formula gets even murkier. Sure, you can track how many people click through banners or visit web pages. Does a paltry clickthrough rate on a banner mean its useless. How about social media? Word of mouth? Blogs? Consumer generated content? Apps? All can play a role in the marketing mix—but no one knows exactly how.

And the quest for accountability is going even further—by analyzing your brain. Have you heard the latest buzzword--Neuromarketing? Millions of dollars are being spent hooking people up to electrodes and trying all manner of techniques to trigger their brain’s pleasure centers. It works—on marketers with way too much money. The idea of neuromarketing triggers your client’s pleasure center by attempting to prove how consumers get lured in. Are scientists going to perfect the art of advertising? Don’t bet on it.

We have access to more data, more statistics, more slicing and dicing of numbers and tactics than ever. And has advertising gotten any better through the years? Hardly. Has the creative work improved? Nope. Less wasteful? Occasionally. Is it still mostly wasted money? Absolutely.

But you will never, ever hear an agency exec get up in a presentation and say, “We don’t know if this’ll work. They won’t even say, “Well, we think it’s gonna work.” They’ll say, “It will work.” Every proposed campaign is a can’t miss. The producers of ‘Ishtar’ probably felt the same way.

If there’s one thing that doesn’t sell, it’s uncertainty. And that’s why the biggest charlatans in advertising act so confident. Of course, that absoluteness isn’t confined to the ad industry. In these days of Bernie Madoff, the permission to be a complete bullshit artist seems to come free with any order of 100 or more business cards.

We have to prove our worth, because so much advertising is worthless. In a desperate attempt to prove ourselves, ad agencies and their clients turn to any analytics, no matter half-assed or incomplete. And in the land of the blind, the one-eyed man is king. Get used to it. We need find a way to take analytics and interpret them for the creation of better work—not be a slave to imperfect numbers.

Or we could insist on conducting a real effectiveness test. Instead of what clients usually insist upon—watering down a concept and then blaming the agency for poor ROI--let’s compare results between two campaigns. One campaign produced exactly the way an agency recommends vs. one that looks the way it looks after the client messed with it.

Would that prove the superiority of the agency recommended work? I doubt it. Oh, wait—I’m in advertising. Absolutely it’ll work!