Traffick - The Business of Search Engines & Web Portals
Blog Categories (aka Tags) Archive of Traffick Articles Our Internet Marketing Consulting Services Contact the Traffickers Traffick RSS Feed

Posts Tagged ‘adwords’

Five Quality Score Myths Busted

Thursday, July 29th, 2010

In search marketing, it’s easy to get people obsessed with a particular metric: you express it on a scale of zero to 10. (Remember Toolbar PageRank?)

It’s only been a few months now that a significant number of search advertisers have even figured out how to look up Quality Scores on their Google AdWords keywords. (You have to “customize columns” and “show Quality Score”; it’s not shown by default.)

What Toolbar PageRank was to dabblers in SEO, Quality Score is to practitioners of paid search. In this running-of-the-bulls-like environment, the main casualty amidst the din of the hooves in this virtual Pamplona appears to be the truth.

Quality Score multiplied by your bid still determines where you rank on the page against competitors in the advertising auction, so it’s indeed important. High Quality Score and awesome ROI ¬†are key, all else being equal.

But let’s dial it down a notch and check out exactly what Quality Score is, and what it can or cannot do. Here are five common myths about Quality Score, with some suggested alternatives closer to reality:

Myth: AdWords used to be all about rewarding higher CTRs, but now, the formula is incredibly complex.

Reality: Keyword Quality Score is based on keyword “relevance,” which is primarily about CTR. Prominent Googlers have stated that other elements of the formula are mostly “just other cuts at CTR.” Keywords develop a history, so good performance can raise your Quality Score over time as statistical confidence grows. You can also be rewarded or punished by the overall (mostly CTR) performance of various account elements: the whole account, a campaign, ad group, or ad. No one outside Google knows the precise formula. “Other relevancy factors” might include semantic analysis that assesses commercial intent or intent related to better-performing keywords on your campaign or similar competitors’ campaigns. These various other factors will generally be trumped by a very strong CTR relative to other advertisers on the same keyword.

Myth: Quality Score is like old-fashioned SEO. You should make a list of keyword elements to “optimize,” and your score will likely improve.

Reality: This myth is circulated so SEOs and other clever vendors can sell you traditional keyword optimization, including tweaking various page elements of your landing page. It bears little resemblance to the marketing strategies that have always made sense in paid search. There was virtually no change in behavior or performance when Google shifted from a less complex version of quality scoring to the new one called Quality-Based Bidding in 2005.

Myth: You should be doing multivariate landing page testing to improve your landing page and Web site Quality Score.

Reality: The purpose of landing page and Web site quality (rolled out as a second kind of Quality Score in December 2005, following initial keyword Quality Score in August 2005) is mainly to punish particularly poor customer experiences or shady business models. Landing page and Web site Quality Score is not discretely broken out from overall Quality Score and is only updated every few weeks (unlike keyword Quality Score). So, you can’t use multivariate testing on landing pages to gauge Quality Score response, period. Google has noted that landing page and Web site quality effects tend to be “binary” (either you’re in the 98 percent of advertisers who are “greenlighted,” or you’re not).

Improving your Web site’s load speed, customer engagement, and information scent may translate back into a small boost in this Quality Score. Going forward, these factors could have a bigger impact on overall Quality Score. These are the kinds of things you should be doing anyway.

Myth: Make heavy use of exact match on your keywords to boost CTRs, and thus Quality Scores.

Reality: Google states that Quality Scores are normalized by match type, and regardless of match type, the system tries to predict how it would do if all comparable keywords in the auction were of the same match type (probably exact match). You should continue to pay close attention to matching options for various other reasons. Exact match is not a ticket to Quality Score bliss, and will sharply narrow your reach. You should, however, make use of negative keywords (exclusions) where it makes sense. Consumers and Google both love this, and it can translate into higher CTRs and higher Quality Scores.

Myth: The Content Network is killing my CTRs. I’m paranoid that this is killing my Quality Scores, so I disabled the Content Network.

Reality: Quality Scores are calculated separately for search and for the Content Network. You’re safe to place a much lower emphasis on CTRs in network placements than you do for search.

Those are the basics. Down the road, I’ll cover some more advanced debates about quality-based bidding.

This column originally appeared at ClickZ on January 29, 2010. Reprinted by permission.

Google Changes Ad Policy in Response to France Case

Wednesday, July 14th, 2010

Among several other policy updates related to permissible advertising categories and claims, Google has quietly updated its policy to allow “radar detection databases and software,” specifically in France and Poland.

French Navx AdWords Banning Case – Comments

Monday, July 5th, 2010

Google will appeal the French regulator’s decision in this Navx case… and I can see why.

There are so many products and services banned from advertising with Google, third parties have a field day documenting all of them. Google’s own help files keep evolving, and provide a real sense of the breadth of the issues they must consider when deciding whether to accept advertising.

The French decision appears harsh in that it claims Google “abused its dominant position” to shut off an advertiser account. Yet if Google leans too far to the permissive side, it has to worry about being liable for facilitating violations of the law.

Due to its size, Google will inevitably face claims of favoritism and caprice when it comes to banning advertiser accounts (or in practice — the way it normally plays out — allowing Landing Page and Website Quality scores to sink so low that the final keyword quality scores are very low, which may affect not only minimum bids but “eligibility” for each and every potential keyword auction). The only way to avoid such claims is for Google to keep publishing its policies, and to be as transparent as possible as to what’s banned, and what isn’t.

Unfortunately Google has been transparent about around 50% of the process, and opaque about the other half. The various deceptive and discouraged practices Google refers to in its Landing Page and Website Quality guidelines are very much open to interpretation.

Despite Google in some sense bringing this type of scepticism on itself, the question remains as to whether they have indeed “abused their dominant position” in any given case. On the whole, its stock price is high because Google is able to exploit its dominant position generally, to keep ad prices high under an auction regime that is tuned to extract de facto reserve prices.

In attempting to solve several distinct classes of problems — economic issues, user experience issues, legal and policy issues, enforcement issues, etc. — by tossing them all into the same Quality Score cauldron, Google has in some sense made some problems (appear to) disappear from view while guaranteeing that they’ll keep coming back to life, like any undead life form you try to bury. As it turns out, you can’t take a subjective policy decision on whether to accept an advertiser’s money, hope the industry will uncritically accept your characterization of it as a “largely technical” calculation of “quality score,” and have the resulting debate and due process just disappear.

That being said, there is a flaw in the French regulator’s attitude and approach, insofar as it seems to rely on excessive due process being required of a publisher in its advertising policy decisions. All publishers might be second-guessed as to their decisions to accept or not accept certain advertisements. The law here can be quite complex, and it usually leans towards the publisher having the right not to take certain advertisers’ dollars, unless the refusal can be taken as overtly discriminatory or anti-competitive.

But was it so in this case? In this case, presumably Google was acting in good faith to refuse advertising from a service that helped users contravene traffic laws. They do much the same on many similar issues, including products that help people beat drug tests.

Google (or any publisher) will miss the mark on some of these calls: that’s only normal. As a private company, they shouldn’t be assigned the burden of creating an excessive layer of regulatory bureaucracy in their private dealings, given that there are plenty of regulators on the outside to fulfill that role.

Perhaps the French regulator is happy that the nation’s investment in traffic cameras will now be negated, and public safety further jeopardized by people playing with their smartphones in their cars. Maybe that’s their call. But Google banning ads because they thought they were for something prohibited hardly counts as abusing one’s “dominant position” as a publisher.

The European regulator’s reaction is unsurprising, however. Silicon Valley companies think more than anyone else in the world about “scale,” which means sucking unnecessary extra steps out of all business models. I’ve begun to write about how Google necessarily plays a role as a quasi-regulator of many things, and I’ve termed that role “the guvernment“. The “guvernment” of Google attempts to run cheaply to scale; Google always tries to ensure that it can quantify and codify what it can to remove capricious judgment and excess steps (and cost) from the process of passing editorial (etc.) judgment.

That model is not well understood by old-school regulators, who relate better to government in its more cumbersome, weighty role. Surely, though, part of it is professional jealousy. Google has gotten pretty good at exacting a hefty “tax” from advertisers who run afoul of “the guvernment”. Abusing its dominant position? It takes one to know one.


Traffick - The Business of Search Engines & Web Portals


Home | Categories | Archive | About Us | Internet Marketing Consulting | Contact Us
© 1999 - 2013, All Rights Reserved