Jay Cost writes that much of that much of the money spent for voter mobilization is ineffective in turning non-voters into voters.  Maybe he is right and I think that a lot of campaign season advertizing has many of the same limitations.  I'm numb from watching commercials about the New Hampshire Senate race and the Massachusetts Governor's race.  The commercial breaks on the local news have political ads back-to-back-to-back.  It is disorienting.  I think that a certain amount of repetition is needed just to keep your message fresh in the voter's mind, but at the margin, there has to be a better use for some of those dollars.  I remember watching the 2008 election and feeling that Obama would have done about as well if he had spent 30 million fewer dollars on ads.  Scott Brown was able to beat Coakley while being outspent in the air war.

I would suggest that if you wanted to turn non-voters into voters (or to turn some of the other side's voters who have weak or vaguely understood affiliations), some of that money is better spent between elections targeting low-to-medium information voters. The key would be shifting policy preferences (explaining how certain policies would directly benefit them and/or how those policies are consistent with their beliefs) and improving the party brand.  This approach would shift some significant amount of money to different media.  The ads that are being put out today tend to go to media with older-skewing audiences and to programs with the broadest appeal.  There are good reasons for that and those media buys should not be abandoned come election time.  The question is when more money isn't doing very much to advance your message.  It would make sense to spend money on media with large audiences that don't consume much right-leaning media and take only a passing interest in the old-model NBC/ABC/CBS/CNN-type news.  The ads on these media would also have to be different.  The audience does not know your policies or your presumptions.  Your slogans are completely meaningless to them.  Slow down, take a minute (or two) and explain.  Communicating with people who don't already share your frame of reference is hard.  Read this book.  Try to do the same thing with today's audiences who aren't already conservative.  Try writing an anti-cap-and trade ad for viewers of the Daily Show, a tax ad for Univision and an ad for market-driven health care reform for viewers of BET. 

Conservatives have developed great ways of talking to one set of Americans and pretty good ways of talking to another.  The right-leaning media offers conservatives a great way to communicate with tens of millions of Americans at length and at fairly low cost.  Conservatives are also okay at getting their message out through the "traditional" broadcast news media.  Those programs aren't exactly a favorable environment, but a combination of ability to work through the conventions of the programs and advertizing on those programs ensures that (if you are competent) your message will get out.  The problem is that the audience for right-leaning media is not a majority of the country and the size of the traditional broadcast news audience is shrinking and aging.  That leaves tens of millions of people who aren't getting talked to by conservatives in any kind of effective way.  Many of them end up as Democrats by default. 

Jay cost writes that, under present conditions "There is really only one reason to vote: you care about what happens."  True, but I think that one of the least understood communication problems in our politics is how little communication effectively takes place for millions of people and how many of the common place terms of debate are total nonsense (not wrong, just meaningless) to tens of millions of Americans.  If you want people to care what happens, meet them where they are and talk the language of everyday life.  It is harder than it sounds.

I know that there are reasons not to do this.  Money spent just before an election is fresher in the voter's minds.  But before you can reap you must sow.