It's Not Nice to Annoy People: Online Ad Trends via Doubleclick

[andrewgoodman]

The July newsletter from Doubleclick contained some interesting material, including this online ad trend report:

http://www3.doubleclick.com/market/2004/07/dc/news2.htm?sssdmh=dm6.81008&c=0407_smr&id_lead=newsletter&id_source=newsletter_0407

Average CTR's for "all ads" is stable at 0.56%, which represents a 17% decline year over year. Clearly one can't trust aggregate figures, especially not from Doubleclick, but the fact that even they admit rather lukewarm performance is not heartening, especially for those who think banners still work.

"Rich media" CTR's are 5X higher than non-rich media. The orthodoxy in the industry appears to suggest that we must up the ante to regain the lost ground of banner CTR's. But the fact that rich media is out-CTR'ing flat banners 1.17% to 0.23%, to me, might reflect the fact that rich media is much more in-your-face as opposed to genuine interest in the content of the ads. In-your-face ad formats have historically had auspicious starts while ruining the playing field for those trying to get noticed next year, because they get noticed at first, but soon turn users off.

At first, some believe, people either click out of novelty or in a vain attempt to turn off the animated commercial. Longer term, people become better trained to ignore the ads or to stop visiting sites that insist on showing them.

Post-click activity does seem to be improving, probably reflecting improved targeting as well as better landing page optimization. It seems there might be hope, even for those who insist on bothering people with banners.

But looking at the full report, there is a disturbing trend towards larger ad formats. Skyscrapers and leaderboards (giant top-of-page ads) have apparently gained ground as a proportion of all ads served. It appears that those in the ad industry are still pursuing a "scorched-earth policy," convincing themselves that an anvil-to-the-head approach equates with effectiveness, and choosing to ignore the significance of Google's impending IPO to the tune of 50X Doubleclick's current market valuation.

Comments?

[AdamJewell]

especially for those who think banners still work.

That seems to imply banners don't work, when in fact they can work extremely well. What matters most is, just like search, relevancy. Keyword buys, banner buys, rich media buys & content-targeted ads can all work and work very well if they are highly targeted.

It also depends on what the definition of a campaign that "works" is. If it is so narrowly defined as to be the sales revenue generated based on the immediate response after the click then even lots of paid search campaigns don't work these days in market where bid prices are escalating rapidly.

Rich media ads can work very well if they are use appropriately. Appropriately is not hijacking the users screen for 10 seconds or more and making it impossible to do anything while some animated thing dances round on the screen like the ads one might find on a weather related site.

The context in which the ads are displayed needs to be relevant. The beauty of rich media ads is that they can offer the user the opportunity to expand and interact with the ads if they choose to without leaving the site on which the ad is displayed, really creating a "brand experience" or "brand interaction".

Trends toward larger ad units that command attention through interruption is not a healthy one & it is likely CTR associated with those ads will plummet as well. Until there is a greater understanding that online advertising, media buying, and media placement is not the same as buying an ad in a magazine we'll continue to see large intrusive online ads.

Any online ad (banner, rich media, whatever) or search campaign can be a success. It just needs to be relevant. It's that simple.

If presented with 2 media plans:

1) $100,000 for very highly targeted ads & media & $50,000 agency fees to get it all up and running (agency fees at 50% of media cost)

2) $1,000,000 in media fees, $100,000 for the above placements and $900,000 for stuff that is not targeted very well, but added in to make the agency fees look less, & $100,000 in agency fees (agency fees at 10% of media cost)

Which is likely to happen when these are presented to a client?

#1) $50,000 in fees for a $100,000 buy, are you crazy? We shouldn't have to spend more than $10,000 for that so cut your fees or we won't do it.

#2) hmmmm a 1mil buy, agency fees at 10%, looks like a great deal, lets do it! The client feels like it is getting a better deal here and getting 10X as much media without having to pay the agency much more, when the meat of the campaign is virtually all in #1 and the better investment is by far #1.

I think this type of thinking is one of the core reasons we see lots of irrelevant and aggressive ads including rich media. They can be very expensive and help make agency fees look less significant. Maybe if the way we view media buys and fee structures changed, we'd get better results and there would be less incentive to use a "scorched earth" policy to try to do anything to bump up the numbers (or at least CTRs).

It's not what you pay, its what you get.

[McFox]

Doubleclick's biggest weakness is the reputation they have as spyware.

The massive popularity of anti-spyware utilities such as Adaware (http://www.lavasoftusa.com/software/adaware/) and Spybot Search & Destroy (http://security.kolla.de/) is testimony to the unpopularity of this type of strategy.

For Doubleclick to survive at all, they must reinvent themselves as the new, whiter than white, repentant advertiser, imho.

McF