Forum Moderators: martinibuster
I am an adwords advertister and it seem obvious that with conversion tracking in place that perhaps in the future publishers may have to maintain a basic level of conversions into sales for the advertisers on their sites.
Dosen't this seem like the logical next step to keep advertisters happy and the adsense program viable?
As an advertiser I'd love to be able to say I only want my ads to be displayed on publisher sites with x% conversion rate (into sales) for similar words.
could this be the next thing?
thoughts?
What I would like to see is that Google gives us the ability to pay on Conversion rather than on click.
Unfortunately people around here aren't very technical so they don't understand how that works.
They don't realise that 'conversion' doesn't have to be a credit card # .. it can also be PPV (pages per visitor). Got that term from you, btw.
The next lack of understanding is that people worry about advertiser fraud. What they don't realise is that rank is a function of CTR*CPC. If CPC is what you pay on a conversion, then fraud can't exist. If you stiff Google, then you don't rank and get no impressions.
These ideas are really good, unfortunately you are going to get a lot of resistance because of a lack of understanding how it all works. Have fun counting all the posts which say "yah but it's up to the Advertiser to convert..". It's crazy. Sometimes I wonder why I bother posting.
Also, of course, Google has to actually implement these things..
[edited by: blaze at 3:50 pm (utc) on April 13, 2004]
I wonder what the adsense (publishers) folks take on it is?
I am an adwords advertister and it seem obvious that with conversion tracking in place that perhaps in the future publishers may have to maintain a basic level of conversions into sales for the advertisers on their sites.Dosen't this seem like the logical next step to keep advertisters happy and the adsense program viable?
It might keep some advertisers happy, but it would tend to encourage highly commercial "created for AdSense" sites at the expense of quality editorial sites that are designed to serve the needs and interests of their readers even when an article or other content doesn't yield a direct commercial payoff. That, in turn, could result in lower-quality leads, and it would discourage the traditional mainstream advertisers who are used to choosing media on the basis of editorial quality and demographics.
I think you'd also find that quality publishers would be very skeptical about having their revenues determined by the competence or incompetence of advertisers.
Optimization is done by banner ad networks where advertisers dictate where their ads could run based on a number of metrics. In one ad agency for example, their VP said that optimization is based on a number of factors over and above click through rate; what those factors are they haven't revealed. Suffice to say that in the spirit of giving advertisers the best returns for their ads, it has become the norm among banner ad networks to show the ad only to the "best performing" publishing sites.
The question, of course, is how will Google define "best performing" publisher site for its advertisers
I don't think Advertisers choose where to advertise based on "editorial quality". That is only a possible indicator of demographics.
I think you'd also find that quality publishers would be very skeptical about having their revenues determined by the competence or incompetence of advertisers.
But if I am the one paying shouldn't I be able to place my own criteria on tmy continued patronage?
If only successful sites (as defined by the market) were allowed to carry advertising, then confidence in the adsense product would increase.
Separate the wheat from the chaff, as it were.
CTR is G's valuation of success, not the advertisters. For us it is only sales! The bottom line.
If the advertiser could decline placement on sites which did not produce conversions (sales or page views) then that kind of activity (other thread) would have no effect for anyone and therefore not take place.
If you think of it as a brand building tool it can be an okay deal. Although we do not get any conversions we are very happy with the number of impressions. since it does not seem G has the same CTR requirements for ad display as searches, we don't even care about the CTR (actually, in a way low is good!)
The good thing is to see that our ad gets 8k content site impressions per day! Just that many eyeballs seeing our name, even if the don't click is worth the little bit we pay (because of our lower bid)
That's a pretty good illustration of why publishers might not be enthusiastic about conversion-based payment.
I don't care what the sites do for the reader as long as it is generating conversions that I want (as opposed to only clicks). The "needs and interests of the reader" are not really my principle concern.
No, but the needs and interests of the reader are my main concerns as a publisher. And if advertisers or ad networks want to reach my highly targeted audience of people who spend money on travel, they'll need to pay according to a formula that I'm willing to accept. (I might accept a measure of risk when dealing with a handpicked affiliate program, but I won't do it with just any advertiser.)
If only successful sites (as defined by the market) were allowed to carry advertising, then confidence in the adsense product would increase.
To the advertiser who's selling fuzzy German widgets, it doesn't matter if a "Widgets of the World" site has an overall conversion rate of 0.1% because most of the pages are on obscure widgets that don't get targeted ads. The only thing that matters is the conversion rate on the fuzzy German widgets page.
At the same time, for the new advertiser who's selling smooth Maori widgets, the fact that the site's Maori Widgets page has a low conversion rate doesn't necessarily mean the page is worthless--it just means it may not have been displaying the right ads because nobody has ever bought a content ad for Maori widgets before.
CTR is G's valuation of success, not the advertisters. For us it is only sales! The bottom line.
The publisher's job is to deliver prospects. It's the advertiser's job to complete the sale. If you can't convert the prospect, don't expect the publisher to accept responsibility for your high prices, poor selection, bad Web design, unreliable e-commerce server, or--just as important--your poor or deceptive ad copy.
I mention "poor or deceptive ad copy" because it isn't unusual. Example: Not long ago, I was looking for information on "[dog breed] rescue." I searched Google and saw an AdWord for "[dog breed] rescue." When I went to the advertiser's page, I found that it wasn't about canine rescue or even about the dog breed--it was just a sell page for pet products. Would an ad like that have a high conversion rate, or even an acceptable conversion rate? I should think not; anyone searching for information on [dog breed] rescue would be infuriated by such a sleazy ad, and it's unlikely that someone searching for that information would decide to interrupt his search to buy pet products. Now, should a publisher (or Google, for that matter) be expected to subsidize that advertiser's dishonesty and stupidity? No way.
(Side note: I complained about the deceptive ad to Google, and they didn't see any problem. But that's a topic for another forum and another day.)
For starters, the advertising would become COMPLETELY saturated - every dummy who can't "really" afford to advertise will be able to slap up a crappy website and their ads will overlap OURS, the respectable businesses who have a real product, real service, etc. GOOD companies would STOP using AdWords because they wouldn't want to be lumped in with the losers.
Second problem is control - Google has no way of measuring whether you make a sale or not unless YOU configure it to do so. In order for them to have that level of control, they would have to host the site on their network and force you to use a shopping cart THEY built/integrated into their hosting, and would have to make it so that the site hosted on their network can't have any outbound links - otherwise you could just get the user there and redirect them somewhere else.
This is the only way a pay per conversion model would work for Google. Problem is that now Google would know EXACTLY how much you're making, who your customers are, etc, and I'm not comfortable with giving that information to anyone.
[edited by: digitalv at 6:26 pm (utc) on April 13, 2004]
i was hoping this would not turn into another us vs. them thread. I really think something like what I have outlined is on the way weather you or i like it or not.
I take a little umbridge at your (I think implied) suggestion that my ad having a low CTR is somehow dishonest. But maybe that is not what you meant.
Currently my assesment (which is only based on anecdotal evidence) is that adsense is in a downward spiral. advertisers are opting out more and more.
If it is to be saved, something has to be done.
[edited by: nyet at 6:32 pm (utc) on April 13, 2004]
you are missing my point. I am not suggesting pay per conversion. I am suggesting a way to opt in or out of advertising on publisher sites *depending* on how that site produces *conversions* for other advertisters with related keywords.
a way to *not* show ads on "loser" sites. the likely result would be *higher* CPC on the "good" sites. Sites which "served their readers" AND their advertisers would be rewarded and those that did not do BOTH would not be.
This is basically what the new bid adjustment scheme tries to do, I just wan to be able to have a little more control myself because I think I know my business more that G's computer does.
I am suggesting a way to opt in or out of advertising on publisher sites *depending* on how that site produces *conversions* for other advertisters with related keywords.
But how would Google get an unbiased view of conversions from advertisers without introducing complex integration with each of their adwords clients?
I imagine if this part was simple and/or easy they would have done it already since it would, as you point out, make an interesting feature if implemented correctly.
- Tony
[edited by: Dreamquick at 7:04 pm (utc) on April 13, 2004]
I take a little umbridge at your (I think implied) suggestion that my ad having a low CTR is somehow dishonest. But maybe that is not what you meant.
I mentioned dishonest ad copy as just one of the reasons why an ad might not convert.
Currently my assesment (which is only based on anecdotal evidence) is that adsense is in a downward spiral. advertisers are opting out more and more.
I haven't seen that for my topic; I suspect the apprehension or bailout factor varies quite a bit by category. One thing I have seen is an increase in "name brand" advertisers in my category, and I think that's where the real future of content advertising lies. As I've said in other (and much older) threads, AdSense is a godsend for well-established mainstream advertisers who have obtained leads from magazine ads and direct mail up till now. (A while back there was a post in the AdWords forum by an advertiser who said AdSense leads would be a bargain at 50 times the price. That may have been hyperbole, but I don't think there's much doubt that a click on an ad in an online article for French barge cruises, Caribbean villa rentals, gourmet chocolate, etc. is cheaper than the cost of an inquiry obtained from an ad in CONDE NAST TRAVELER, THE NEW YORKER, or GOURMET.)
If it is to be saved, something has to be done.
Well, Google is doing something by introducing variable discounts based on whatever mysterious formula it may be using. Is that enough? It's too early to tell, since the formula has been in place for less than two weeks.
I think an "AdSense Select" program would be another step in the right direction. (Another forum member stickymailed me with the comment that Google could charge $100 to consider sites for an AdSense Select program; that would defray the cost of the editorial review while discouraging quick-buck, churn-sites-out-by-the-dozen publishers from applying.)
I also think the market will come to the rescue. For example, Overture could introduce a product that offered the best features of AdSense without the problems (assuming that AdSense doesn't evolve, which it almost certainly will). And in industries with enough commercial potential, vertical ad networks could appear. I think advertisers and publishers will have more choices in the future. We're still in the earliest days of what promises to be a huge industry.
Currently my assesment (which is only based on anecdotal evidence) is that adsense is in a downward spiral. advertisers are opting out more and more.
You mean Adsense on CONTENT SITES only? Because you get to choose which channels you get your ads to run on.
I'd like Google AdWords to allow an adverstiser to exclude SE traffic, ie have a campaign on content sites only,
But how would Google get an unbiased view of conversions from advertisers without introducing complex integration with each of their adwords clients?
maybe the same way they are doing it now....
"How smart pricing works
We are constantly analyzing data across our network, and if our data shows that a click is less likely to turn into business results (e.g. online sale, registration, phone call, newsletter sign-up), we may reduce the price you pay for that click. You may notice a reduction in the cost of clicks from content sites."
I just want to be the one to define "less likely to turn into business result"
I might define it differently that someone else. For example for my company the average page views per visitor is our benchmark, not $ conversions.
when their machine defines it for me, they are making assumptions.
Perhaps in the same way that I cannot use keywords that don't perform over time, a publisher should be dropped if (over time) thier site does not "perform" for the advertisers. In whichever way the advertisters define performance.
I might define it differently that someone else. For example for my company the average page views per visitor is our benchmark, not $ conversions.
How would Google and the publisher know that your benchmark was legitimate? For that matter, how could Google and the publisher know whether the click resulted in a conversion according to your benchmark? Who'd be auditing the system at your end?
I just want to have some input into how those stats are applied to my ad. All business models are not created equal.
so If i were selling widgets I would want to say "show my ad only on content sites in which there is a 1% conversion rate related to the word 'widget', and for those sites I am willing to pay X." If the site generates 5% conversion rate, I am willing to pay 3X, etc.
Now, lets say I don't sell widgets, but I sell complicated advice about how to make a better widget to widget makers. I only get 20 conversions per year. So the number of conversions is not a good measure for my business. So I use the average page views of my site (site penetration) as a benchmark to measure relevance and 'success'. Well I would then want to say "show my ad only on content sites related to the words 'make better widgets' which generates and average of 5 page views on the related advertister's site, and I am willing to pay X" on sites which generate 8 averpage views per user I am willing to pay 3X.
publisher sites which don't generate 'success' for any advertister, end up making very little money and no advertisters opt-in to them. Natural selection runs its course and they are effectively removed from the system.
Sites which produce for the advertisters are rewarded and make more money.
Whenever I check a refferal link in our logs for a publisher site, I'd say about 30% of the time it is either totally lame, or a site which seems to exist ONLY to participate in the Adsense program. we have to get rid of those somehow.
I think that this is what smart pricing is trying to do right now, but their machine is deciding *for me* what success means. I want to decide.
