It is precisely for this reason that I'm largely confused by Internet advertising payments. A lot of companies only want to pay out based on the number of clicks an ad gets. If what this article says is true (and I've come to this conclusion myself before), then the direct clicks as soon as someone sees an ad on a web page is largely irrelevant. What should matter are the impressions over time, allowing that product or service to grow in the consumer's mind.
1 - TV ads are more interruptive; to ignore them, you have to leave the room or change the channel. Most people just deal with sitting through them. On a website, you can easily ignore the ad. There are plenty of exceptions in digital advertising - pre-roll for web video, page takeovers, and other stuff that is unbelievably annoying, since your expectation as a web user is to not be interrupted.
2 - It depends on the key performance indicator for the campaign. Certainly, direct response campaigns want to measure clicks (or purchases/actions - the closest you can get to the sale, the better). But there are awareness-style campaigns on the web, and those advertisers pay by impression, not clicks. The problem is that with networks, you don't know the quality of the place your ad is running, so an impression is worthless. You'll only pay directly to a publisher you know is at a certain level of quality.
At a past company, I saw our media buyers put in ads at $170 CPM (cost per thousand impressions) for a particularly relevant publisher audience. The advertiser was in financial services, and the publication was uber-niche and uber-influential.
There are plenty of exceptions in digital advertising - pre-roll for web video, page takeovers, and other stuff that is unbelievably annoying, since your expectation as a web user is to not be interrupted.
I don't know if I'm unique, but every single time I'm presented with a pre-roll ad, I close the tab I started the video in. 100% of the time. I always wonder if stats on that type of thing are collected and given to the "right" people. (I'm 90% sure they are not, but I thought I'd ask, since you're an expert.)
Yes, those kinds of stats are collected. Ad networks, servers, and exchanges are some of those places that physics phds go when they don't go to Wall Street or, heaven forbid, academia. We're not at that level, but we do have a whole analytics department filled with statisticians.
No, you are not the norm; video pre-roll is a particularly strongly performing media type at the moment. It does radically depend on the audience & creative, of course. But if you're looking at the rates of a publisher, pre-roll is going to be one of the higher ones.
I wonder if particularly intrusive adverts might have a detrimental effect on a brand.
If I accidentally move my mouse over an underlined word and up pops a video with sound over the top of what I was trying to read, well I certainly feel pissed off with Vibrant Media, and I probably get annoyed enough with the website that I won't visit it again.
I'm unsure if it has any actual effect on the brand though.
I decided not to use Groupon just because of the fact that on one point pretty much every advert on the web featured their pink-colored deals for fast foods. I just got really, really annoyed.
People cling to clicks because it is the easiest thing to measure (clicks and then conversions) - and also because Google has trained them to do so with AdWords.
There is a movement to more multiple attribution style reporting - so when you get a sale you would see not only the click that generated it but the display ad that person saw that caused them to search on Google and finally click on your ad there.
