Back when I first started running campaigns I would do stupid things.  Stupid things like decide to run an ad or send a piece of direct mail because we thought it looked "good."

If I could, I would go back in time and slap some sense into myself.  I spent hundreds of thousands, even millions of dollars airing ads because we thought it looked good or the focus group seemed to like it.

These days, I live by the mantra "it doesn't matter if the candidate likes it or not.  Does it convert?"  As in convert voters or convert into sales.

After testing a lot of ads and a lot of direct mail, I have developed a sense of what does convert.

This ad, does not convert.

An obviously angry man, complaining about another supposedly angry man, does not convert.  There are ways to make it convert...but they ain't got it yet. 

Most campaigns do not test ads.  They may focus group an ad.  Or even people meter it at a mall test.

But I have found that focus groups are bullshit.  People just make shit up.  Half the time if you people meter the ad, then have a focus group discussion, the results will be the exact opposite.

A focus group is only useful to find out how people will talk about an ad around a water cooler.

The people meter is slightly more accurate.  We built a people meter with a built in lie detector--yep, just like the cops use--but even it is not as accurate as a good old fashioned randomized double blind--test group, control group.

Sure it takes time.  Sure you have to do some math.  But you find out whether the ad really converts before you drop a million bucks in TV time.

Brave New PAC may or may not be airing this ad.  But if they think a youtube viral is a test, they don't know how to use a graphing calculator.  (Which is what you need to do a proper test.)

If they tested it, it would have found its grave in the small, remote, isolated TV market they tested it in.  Or, that may be the reason it is on youtube.

If you want McCain to win, play the ad on youtube again and again.  Maybe it will encourage them to blow a couple million bucks on an ad that does not convert.

---