A different look at disappearing advertisers

from the AdWords side

[ronburk]

I do some AdWords for physical goods/services unrelated to my AdSense work. Today, I checked on a campaign I started several days ago. I suddenly realized that I had screwed up and failed to disable the Content network for my campaign (the Content network having consistently delivered me total crap ROI compared to Search for this particular field).

What shocked me was the tiny number of impressions, compared to what the Search network was getting. Back when I did test the Content network, I came to expect that the Content network would always return a relatively large blast of relatively useless impressions.

These are fairly specific, niche keywords I'm talking about, but there is no shortage of websites in the topic area, and I am shocked to see them get 2 orders of magnitude less impressions on the Content side.

Has Google seriously tightened up the criteria for displaying ads on AdSense sites? That would explain at least some portion of "disappearing" advertisers.

This inspires me to go make a thorough search for advertisers my competitors might be getting that I'm not (and analyze their content, if I find any).

I guess I'll also test the Content network on some other campaigns to see whether this is an anomaly or not.

[jomaxx]

Apologies but I'm not sure what you think this particular case demonstrates.

[ronburk]

It demonstrates that an AdWords ad that would have been splatted across dozens of (probably questionable quality) AdSense websites a couple of years ago is now being displayed on only one or two -- that's my inference of the situation.

[ann]

It does sound interesting. I think I will check out some of my competitors sites in much the same way.

Ann

[jomaxx]

Yeah, but what mechanism would cause your ad to only match a small number of websites? I never see PSAs, so it doesn't seem as if every advertiser is experiencing the same thing.

I don't do AdWords, and I hesitate to extrapolate from a sample size of one campaign, so I'm not going to try to suggest possibilities. But I wonder if there's a more mundane explanation than a huge change in Google's matching algorithm.

[ann]

Well,
I don't do adwords either but on looking at my competitors I found, interesting enough, that they are getting the exact same ads I am seeing...mostly low paying. I guess my biggest site is about the wrong widget. :(

My newest site is getting some great ads and high paying ones BUT I haven't yet built up the traffic for it. The middle site is doing soso in the traffic area but low paying ads.

I guess my original niche was just not destined for stardom. :)

Ann

[Green_Grass]

IMO , Google has increased slightly CPC for content side also in the last 2 months or so. The change is minor maybe 3-5% upwards. But it does reduce impressions if the Contents side bids are not upped. If an advertiseer is running things on remote, he/she may suddenly find a drop in impressions and clicks.

[netmeg]

I took Content off most of my clients' AdWords accounts, and am slowing adding it back in for some of them. I am noticing many smaller impressions as well, even in niches like medicine and sporting goods, where those ads used to get splattered across everything that moved. I assumed it was because I set the max CPC for Content considerably lower than it is for Search.

[ronburk]

what mechanism would cause your ad to only match a small number of websites?

I don't understand the question. Google controls where my ads get displayed, so they would be the mechanism that would cause this. Why, whether, and under what conditions is open to question and maybe what you meant. Certainly something as trivial as eliminating display on websites returning less than the mean CTR would produce a large reduction relative to what I used to see.

it does reduce impressions if the Contents side bids are not upped

Thought of that, but since my Content ads are showing a respectable average position of 4, I discounted it. Still... I'll double my bid for a few days and look for an effect. Actually, I'll triple it -- no point in being timid if you're looking for cause and effect.

[ronburk]

on looking at my competitors

Just curious, Ann -- how many competitors do you figure you have?

For some of the ads on my site, a Google search of the ad text shows they've been displayed on upwards of 600 other websites. Multiplying that by all the different ads on my site and, even after subtracting the intersections, it's looking like more research than I can do a thorough job on without some kind of automation to help me.

[ann]

HeH,

Over 9 million.

I only spot checked and came up with that as a quasisump. Eyeball method. :) Same tired ads.

Ann

[ronburk]

Over 9 million.

Yowsa! OK, I now have a new step during keyword research. Find some AdSense ads targetting the keyword. Google for the exact text. If more than 5 figures returned, just move along...

:-)

[ann]

Just finding the right niche itself is serendipitous.

Ann

[ronburk]

Didn't have to wait long. Tripling the CPC made the Content impressions go from 2 digits to 3 digits before the day closed out. CTR is crap -- which is actually good news (leaves me hoping it may deliver a reasonable conversion rate).

Oddly, my average Content position dropped significantly with the high CPC. That one, I'm still pondering.

This is all very interesting to me. It sure seems like things have changed significantly about ad placement from a couple years ago. The puzzle is how exactly that effect comes through to me as an AdSense publisher (and how to take advantage of it).