SEOJaimie
01-30-2006, 05:58 PM
The following is a letter I sent to Google AdWords management. My complaint is not that Google cannot match things perfectly, or well, always, but rather that there is no way to evaluate this except in retrospect, and inconveniently.
Let me know what you think. Does this make any sense? It's really a feature request more than anything.
Dear Google Management,
I am a programmer and a search engine marketer developing a content portal and managing an advertising campaign for a prominent firm in NYC. And while your search PPC is a fine tool to that end, your content match product has a fundamental flaw that makes it very problematic. What I ask for is completely possible and until it is offered I cannot continue in the content network. Other vibes I get suggest that other people are having the same problems.
In order to evaluate where my ads are placed, I am told I must look at my server logs. And, while I am a programmer, and fully capable of looking at Apache logs, this is far from optimal. They may or may not even have accurate information depending on whether referrers are accurate. It's just painful, in general, to do this. It's not summarized, and I don't feel I should be the one to write the tools to report which sites are generating traffic. Logs are not relational, or meant for this type of reporting at all.
Your system placed me on, apparently, completely irrelevant spots. I acknowledge this as a risk, but my bounce rates and page views per visitor were so low, it looked like click fraud. I have requested a review of the traffic, and I expect that it will be done.
Why can I not, at least in (recent) retrospect, look at the sites I am appearing on in a simple report? As a programmer, I can assert that this information is in a database somewhere, and it can be provided in a somewhat current report. That way I can evaluate the quality of those sites and, in a convenient interface (hopefully), evaluate whether or not to exclude them in the same interface.
The problem is, this isn't an option, so advertisers must pay to evaluate that your system is providing an inferior product, if it is. And I know this result is not universal, and I'm sure that in areas that aren't a niche, the system must perform well, but in this case it performed extremely poorly. So while I run the test, I have no convenient way to evaluate it until a month or so, after collecting data and mulling over it.
Furthermore, it has come to my attention that after I signed up, you added a new feature to price content and search ads separately. To that end I state the following. I realize that it had to be at least substantially motivated by other totally disillusioned customers receiving inferior results from the content network.
Without thesefeatures, Google runs the risk of providing irrelevant results in AdWords, which is something I have run in to, and something I know AdWords has been designed to avoid from the ground up.
I propose an interface to see a snapshot of which sites you appear on, and an easy way to blacklist sites you do not wish to appear on.
I really love Google AdWords, but this is a feature that I see as not only totally possible, but something that will increase relevance.
Regards,
Jaimie
Let me know what you think. Does this make any sense? It's really a feature request more than anything.
Dear Google Management,
I am a programmer and a search engine marketer developing a content portal and managing an advertising campaign for a prominent firm in NYC. And while your search PPC is a fine tool to that end, your content match product has a fundamental flaw that makes it very problematic. What I ask for is completely possible and until it is offered I cannot continue in the content network. Other vibes I get suggest that other people are having the same problems.
In order to evaluate where my ads are placed, I am told I must look at my server logs. And, while I am a programmer, and fully capable of looking at Apache logs, this is far from optimal. They may or may not even have accurate information depending on whether referrers are accurate. It's just painful, in general, to do this. It's not summarized, and I don't feel I should be the one to write the tools to report which sites are generating traffic. Logs are not relational, or meant for this type of reporting at all.
Your system placed me on, apparently, completely irrelevant spots. I acknowledge this as a risk, but my bounce rates and page views per visitor were so low, it looked like click fraud. I have requested a review of the traffic, and I expect that it will be done.
Why can I not, at least in (recent) retrospect, look at the sites I am appearing on in a simple report? As a programmer, I can assert that this information is in a database somewhere, and it can be provided in a somewhat current report. That way I can evaluate the quality of those sites and, in a convenient interface (hopefully), evaluate whether or not to exclude them in the same interface.
The problem is, this isn't an option, so advertisers must pay to evaluate that your system is providing an inferior product, if it is. And I know this result is not universal, and I'm sure that in areas that aren't a niche, the system must perform well, but in this case it performed extremely poorly. So while I run the test, I have no convenient way to evaluate it until a month or so, after collecting data and mulling over it.
Furthermore, it has come to my attention that after I signed up, you added a new feature to price content and search ads separately. To that end I state the following. I realize that it had to be at least substantially motivated by other totally disillusioned customers receiving inferior results from the content network.
Without thesefeatures, Google runs the risk of providing irrelevant results in AdWords, which is something I have run in to, and something I know AdWords has been designed to avoid from the ground up.
I propose an interface to see a snapshot of which sites you appear on, and an easy way to blacklist sites you do not wish to appear on.
I really love Google AdWords, but this is a feature that I see as not only totally possible, but something that will increase relevance.
Regards,
Jaimie