The Death of CPM's

As we stumble our way across that vast digital frontier, we are guided by signposts and Sherpa guides, all of whom beckon us this way or that way, on toward our virtual valhallas of e-commerce. Everyone, it seems, has his or her own idea of what works and what doesn't. And if they don't have the answer, they certainly have an invoice they can mail you to charge for one.

Yikes. That was a mouthful.

But what it really means is that I'm getting pretty tired of all the pundits out there giving everyone wrong information about what does and doesn't work on the web. Especially when it comes to advertising.

It's been a good couple of years since the whole "advertising on the Internet" thing got rolling. And now that a few billion has been sprinkled around, the reality is finally setting in: very few people really know what's going on. The vast majority of lemmings -- like those Wall Street bozos that push Amazon stock up past $300 when there's no hope of seeing any profits there for the next four years -- cling to whatever the guy in the next booth says. It doesn't matter if it's true, but if everyone else is saying the same thing, who's going to know the difference?

Well, I know. And one of the things I know for sure is that the same thing is happening in online advertising.

For a couple of years, online pundits and gurus have been trying to convince us that the way to buy advertising on the web is by analyzing your buy with CPM's (Cost Per Thousands). The idea is that the lower the cost per thousand impressions, the less you pay for each impression. Sounds good, eh? That's because it is good -- if you're buying television, radio or print. When it comes to the web, it's a whole different story.

Online media is totally different than offline media, The trouble is that most of the people in online media fall into two categories:

1. Tech heads who have space to sell
2. Offline media reps who left their old jobs at TV Guide to jump on the Internet bandwagon.

Neither group (as a whole) seems to have caught onto the fact that blasting hundreds of thousands of eyeballs at an ad might work offline, but can be a disaster on the web.

The reason why it MIGHT work offline is that programming on TV or radio is broadcast at one time, in one place. The viewing public has no choice about when it sees the message. Online of course, the entire world tunes into what it wants, when it wants it. Another factor is that while there are relatively few channels to deliver the offline message, there are literally millions to deliver it online.

Put them all together and what you've got is a web strategy that doesn't rely on volume as much as it does on quality of traffic. And as I've pointed out before (in "Web Traffic and Genital Size", http://www.robfrankel.com/genital.html ), the web is great at attracting quality traffic, which is what you really should be after. I mean, who cares if 500,000 people click on your banner if none of them buy?

All of which is why the whole CPM concept -- for the most part -- is toast. What you want to buy is advertising that delivers quality people who are ready -- or at least highly inclined -- to buy what you've got. Sure, you'll get some leisure-suited ad broker to throw you CPM's of $2 or less. But he's going to sell you on that for two reasons:

1. CPM's are easy to unload. Targeted lead generation requires way more work.
2. Most of these guys are advertising reps for brokerages who have hidden agendas: they only sell the sites they represent. Which means your interests take a back seat to the brokers' and his clients.

In my book, there's a smarter way to go and it's not that difficult: Find your psychographically compatible sites and contact them directly. Ask them about their activity and impressions. But junk the whole CPM thing in favor of a flat 30 day fee. Believe me, web hosts would rather take a fat check now than the disappointing few cents they get from the ad brokers who burn them. Plus, when you negotiate directly with a site, you can firm up the deal with extra goodies (like extra graphics) and kill yucky stuff like banner rotation. Play your cards right, and you might even parlay the whole gambit into a month long sponsorship.

Of course, agencies and ad brokers will continue to sell their CPM models. And people will keep buying them. But I suppose that somewhere out there, people are still buying brand new eight-track tape machines, too.

Rob Frankel