Facts about Banner and Display Advertising | Hacker News Reader

Facts about Banner and Display Advertising

adpushup.com

13 pointsatulagarwal5 comments

CTRs may be low, but the post almost ignores impressions, which are much harder to quantify but still having some value[0]. Of course, with the proliferation of small devices, ad plug-ins, and the general 'banner ad blindness' of a population that is increasingly familiar with the Internet, it's all probably still dropping in value.

I disagree with those who assume it's all useless because they've never personally clicked on any ads ("After all it’s a $16 Billion market. Nobody would spend that much money, without any realistic returns"), but the trends are also pretty clear. Sometimes I feel like retargeting is the last great scheme to squeeze more conversion out of the whole model. But there will probably be at least a couple more...

[0]http://econsultancy.com/us/blog/10487-view-through-attributi...

I agree about the impressions point. High CTR or not, the Ad is indeed shown to the user, and to look it form a totally different PoV, the low CTR ads mean relatively low CPMs (i.e. eCPM, assuming CPC to be low). If looked from a pure branding or impression perspective, this might be desirable (although I'm not saying low CTR is good for the publisher OR the advertiser).

At www.optimine.com we are doing some interesting stuff to address the abysmal CTR of display ads.

The key is to measure the effect that your display ads are having on your paid search ads, ie measure the "brand effect" impact on purchase decisions.

The direct effect of banners ads is negligible, but that is missing the point. Nobody clicks on a Coca-Cola TV ad, but they aren't going to stop TV ads anytime soon.

Yes, which is why using other metrics to measure performance would be better (specially offline).

Our startup is also currently building a product which will help publishers improve ad performance by using continuous testing, machine learning and optimization.

Are any of those statistics actually true? I see TONS of ads all the time. In aggregate, aren't I more likely to click on a banner add than be accepted to Harvard and Stanford? Or are they saying for any one display add I see, I am more likely to blah blah blah than click the ad?

Reply on news.ycombinator.com