Interesting Patents on Google Quality Score

Google provides sophisticated methods of determining Google AdWords rankings. The complex set of algorithms that determine how good an ad is collectively known as Quality Score. A good ad can rank high without necessarily requiring high budgets.

Unlike organic SEO whose algorithms are kept under wraps, Google AdWords Quality Score is more transparent probably because of the fact that advertisers deserve to know. As paying customer, one demands to know how to achieve optimal results. Knowing the idea behind the mechanics of Quality Score provides guidelines on best practice pay-per-click search engine marketing.

A new set of pending patents generally abide by user behavior in determining ad quality. More than just click through rates, other factors that may make up Quality Score are also revealed, as enumerated by Bill Slawski:

How many times a user selects a given ad in a given session

How can a visitor click an ad more than once? If I am based in Hong Kong searched for “hotels” and click on one of the ads but they don’t return specific packages for Sydney so I try another search query using “sydney hotels” and clicking on the ad from the same company that feature rooms from Sydney.

If I click on more than one ad on a given set of ads, does it mean that the first ad displayed is inappropriate and therefore receive lower Quality Score? Is there a qualifying time difference between the two ad clicks before the first can be considered low quality and that the second click isn’t accidental?

A duration of time, from an ad result selection, until the user issues another search query. This may include time spent on other pages (reached via a search result click or ad click) subsequent to a given ad click.

This is related to the second question I posed in the first item. There is a considerable effort in perusing a page referred to by an ad. Therefore time duration must be considered between clicking an ad and issuing the next search query. Time is mostly spent on landing pages, forms or at the shopping cart section and unless the transaction is successful and the visitor doesn’t perform search within the session, time is measured.

A ratio of the time, from a given ad result selection until the user issues another search query, as compared to all other times from ad result selections until the user issued another search query.

It is a common perception that the shorter the time spent by a visitor on an ad’s landing page and consequent pages, the less interesting the advertiser’s landing page is. So a comparing this time duration with all other times spent on ads before the user issued another search query is important in measuring the quality of ads. Was the ad related to the search query? Was the ad able to overcome the vagueness of the search term and presented its best guess?

Time spent, given an ad result selection, on viewing other results for the search query, but not on the given ad result.

There are moments when it is difficult to determine which ad is compelling enough to click. This isn’t entirely exclusive to pay-per-click ads; on certain occasions, organic results present uncharacteristically unrelated links to pages. (Remember the link that used to be shown at the bottom of the page that says along the lines, “Don’t like the quality of search results?” inviting users to make a feedback?) On some cases, users refine their search queries by making longer phrases in an attempt to go out of the ambiguousness.

I think this is a more important factor than the previous ones. It’s because instead of looking at one ad, disliking it and going back to search results, the array of ads displayed aren’t attractive enough to prompt the user to click on anyone of them.

How many searches (i.e., a unique issued search query) that occur in a given session prior to a given search result or ad selection;

This determines how related search results are compared with the search queries. As long as users don’t see the ads fit to what they are looking for, they’ll continue to issue a different query or give up searching. We can’t rely on Google to deliver the best ads for specific search phrases. Factors such as pairing of ads and search phrases or keywords in the copy contribute to the overall quality of the ads. The ads or landing pages might look perfect but when paired with the wrong set of keywords (broad match, exact match or otherwise), the Quality Score could probably get hit.

How many searches that occur in a given session after a given search result or ad selection.

Ideally, when we perform search and select which ads to click, that’s it. We leave the search results page and focus on the landing page of our selected ad, hoping to locate what we are seeking. To our frustration, searching for “shoes” yield a variety of results and not the mountain climbing shoes we’re hoping to see. So within the same session, we refine our search phrases and land more specific ads. Why a given session? Generally, it is thought that a search query or series of queries are related when done in a given session; new sessions usher in an entirely different one.

From this factor, we can hypothesize that using singular terms (non-brand) is one of the worst moves a search marketer can do on his pay-per-click campaigns.

Indeed, it is very interesting to study visitor behavior as a basis of deciding our ad model, as Bill puts it:

If you use paid advertisements through Google, these patent applications may be worth delving deeper into. It is pretty interesting to see all of the user behavior considerations that may go into determining a score and placement for an ad.