p/g
Just so my response has at least some actual content, here is my number one tip for success on the content network:
* Advertise using extremely targeted ad groups, with a smallish number of keywords which are all about exactly the same thing as the type of page you'd like to have your ads appear on. Do not water down this list with marginally related keywords. Resist the temptation to do so. And of course, create a brilliant ad with a compelling message or call to action, which is also about the exact same thing as the type of page on which you'd like that ad to appear.
AWA
I think the absolutely most important thing you can do as far as the Content Network is to realize that it's a complete different kettle of fish, and treat it accordingly. That means separate campaigns, separate bids, separate reports, and separate analyses. If you lump it in with your search and search network efforts, you'll never ever get an idea if it works or not, and you might as well pour your money down a rat hole (or send it to me)
Right now, I'm having actually worse headaches with the quality of the Search Network than I am with Content. I really resent having to take 90% chaff in order to get the 10% wheat that I really want.
How do I know what is the optimum number of keywords.. which variations help, which hurt in defining my target topic? If I am a newspaper... lets say the Chicago Tribune and am targting pages about chicago news, am I better off with a single keyword "chicago news" or should I add "chicago local news" "chicago illinois news" or how about "windy city news"... or would the last one mess the ad group up? There is just no way to know. I wish you just enter a single keyword per content ad group but when I do this I get very few impressions compared to what I get when I hit it right. The problem is.. you never know when you hit it "right".
Also, I wish the site targeted ppc would get out of beta and open to the rest of us who missed it. They keep rolling out new features but yet this one never seems to be one of them.
Block continents from ever showing your ads. Use both an opt-in (i.e. location targeting options) but an opt out method of 'I never want my ads to appear in these countries/continents'.
Better understanding of what theme Google has decided your keywords are in. Maybe some example sites your ad would appear on based upon your current keyword selection.
concerning this one:
~~~~~~~~~~~~~~~~~~~~~~~
See what sites are sending you traffic.
See what sites are sending you conversions.
~~~~~~~~~~~~~~~~~~~~~~~
I do have an idea to take this one step further...
I don't want to sit around and monitor 1000's of sites, I can't, I just don't have the time.
If I could optionally configure one of these:
google conversion tracking (GCT) event (like a sale or signup)
google analytics (GA) funnel goal (like navigating to a page)
to be monitored for the content network sites that bring me traffic...
allowing me to set a threshold that shuts off sites that fail to meet the criteria...
so if I defined my "hurdle" as a GCT sale and set my threshold to 200...
or...
(as an affiliate) if i set my "hurdle" to a GA page visited [like my jump to merchant php page - meaning I handed them off to one of my merchants... i can't directly monitor sales as an aff] and set my threshold at 100...
then any site that failed to meet my criteria settings...
like they sent me 200 clicks/visiotrs and didn't make a sale...
and/or...
like they sent me 100 clicks/visitors and none of them left my affiliate site to go to my merchants...
then the system would automatically get added to my criteria-blocked sites...
then have the spanked sites listed for me to see...
and give me a chance to let any one site (or sites) back in later, in case I change things to better improve my conversions or whatever...
call it... der auto spanker
so i set criteria and let the chips fall. if a bunch get blocked, i can reconsider my hurdles, let them spanked ones out of their time out corner and start again...
now, many sites will never have enough clicks to get spanked... right, and if I was watching them, same thing... i don't care about the small clicks here and there, but if a site becomes a major traffic source for me within the content network, i need a rude nanny hired to spank the ill-behaving kids for me...
[edited by: RhinoFish at 2:24 pm (utc) on April 20, 2007]
Well, I want G to create this as a feature that allows me to programmatically monitor the sites within the content network that are sending me clicks.
Anyone else have any feedback? Harsh criticism welcomed!
Anyone else think the Der Auto Spanker would motivate them to do more with the content network? Or not?
No hard feelings AWA, but RhinoFish just described the exact reasons why we stopped Advertising Via Google.
This would be a fantastic addition
allowing me to set a threshold that shuts off sites that fail to meet the criteria...
I think this idea is brilliant and fair. I'd even suggest taking it a step further. Not just shutting out poor performers, but giving refunds. That sounds overly optimistic, but if G wants to encourge people back in, or to start up, it's reasonable. It could set limits on the number of clicks to constitute a trial run, say, 100 or whatever.
Of course it has to be noted that site designs change. Adsensers do, occasionally, revamp the site, increase its quality, or test new placement and color schemes. So G. would want some type of way to monitor significant site changes, and then "reset" a site within the Content Network if/when that happens.
G needs to develop a better innovative or 'beta environment' in the sense that it encourages more testing with different free trials (within certain limits).
p/g
Ok, I suggested that the adwords account user could, at any time, take "blocked" sites back out of the "time-out corner" to give them another chance. But your idea of G also making recommendations to release those being spanked is excellent! In my list of blocked ones, G can put a recommended to try again (green light) and base it on aggregate data they collect from other advertisers plus the site changes you suggested. Let us choose in the end, but sure, give us more automated, algorithmic data to help us decide. Love it - thanks for the input.
Any more?
I'd especially like anything negative anyone sees in this idea - for publishers, advertisers and/or G itself - something like this has to work for all parties for it to be truly effective.
anyone have any thoughts, good or bad, on my idea above?
I built this into a product several years ago as POC code. It worked extremely well. So well that it caused the worst of the worst publisher sites to jump out like rats on a sinking ship. That's the problem with it though. When it works as designed it helped the advertisers but was not good for the network profits. So the code was mothballed.
The moral of the story: There is to much money for Google to loose if they cleaned up all the rats.
JAG
While that statement is true, I believe it isn't the whole picture. They'd lose money from exterminating the rats, but then more people would come if it was a rat-free place.
Maybe it's more analogous to a bar bouncer than a nanny... if a bar is too raucous, people stop coming.
There's more to be gained by having a clean content network than by taking rat money.
It's voluntary for us to feed the rats. I choose not to. But I'd rather have them exterminated - there are plenty of GREAT sites out there for me to get high quality traffic - if the rats can be put down.
But I'd rather have them exterminated - there are plenty of GREAT sites out there for me to get high quality traffic - if the rats can be put down.
No doubt that is the preferred scenario. It's just hard to do in a network that does zero screening in the sign up process.
JAG
