Along with max cost per click, Google AdWords employs a “quality score” to determine where ads will rank for certain keyphrases. The quality score factors in click-through rates, ad quality, and landing page quality with respect to the keyphrases being bid on. This system has serious flaws. It is common for ads that have little or no history that are circulating for very specific keyphrases not to show because of a low quality score. This means some of the most relevant ads won’t show for specific queries because they haven’t had a chance to establish a history to demonstrate their quality. Automating the analysis of ad text and landing pages leaves a large margin of error that AdWords doesn’t compensate for. Worse yet, the system often assigns higher quality scores to ads that are not relevant. This is more common in cases of geographically targeted keyphrases. Sites that have ads circulating for specific cities might have ads circulating for products and services they don’t offer because their landing pages are specific to the location being searched. For instance, a real estate site for Pleasantville, New York, might have a high enough quality score to have ads circulating for search terms such as “pleasantville new york daycare.” Even though they are obviously not relevant to daycare services.
Setting higher max CPCs can offset low quality scores, but for specific terms that may have only a few ads circulating, the max CPC might be well over a dollar and still not show. The max CPC should be closer to ten cents for keyphrases like this.
The Problem
AdWords treats terms that are broad matched the same as terms that are exact matched. This means that if a query broad matches a term being bid on it is treated on the same level as a query that matches exactly the term being bid on (Either term could be bid on as a broad match. The distinction here is how the query in fact matches the term being bid on, not the matching set for the term). This is not a good way to have the most relevant ads showing. If an advertiser takes the time to bid on a term, it is likely that their site is more relevant to the term than a site that is broad matching the term. This is because people can’t possibly think of every variation that their ad might show for when they broad match a term.
AdWords should start treating these types of matches differently. Not only would it improve the relevance of their ads, it would give small advertisers more of an advantage in AdWords. Claims about the benefits of the “long tail” in PPC advertising would actually have some support.
Source: http://purevisibility.blogspot.com/2006/10/adwords-quality-score.html
