After months of keyword research, ad text testing, and landing page optimization, youa��ve carefully crafted a high-performing AdWords campaign, hitting your profit and volume metrics, and with that warm fuzzy feeling inside that comes with a job well done. Then it happens. Clicks plummet. The keyword that cost $2 for first position now costs $10 for tenth position. You didna��t make major changes to the account, so what happened? Then you see it. Your Quality Score has nose-dived, and keywords once listed as a�?greata�? with scores of nine or ten are now a�?poora�? at three and four. All that work building your account is now meaningless. Then you wake up a�� it was all a dream! Or was it?
Any search marketer in the business longer than a few months has no doubt experienced a Quality Score smackdown a�� a sudden decline in Quality Score (QS) across a few keywords or even an entire account. Regardless of the scope, a smackdown is a demoralizing and frustrating experience. You can, however, fight back a�� knowing what to do in a QS emergency may save your accounta��s life!
Why Does Quality Score Exist?
Before starting to massage your account to improve your QS, you need to understand the theoretical framework behind it. Googlea��s official explanation focuses on relevancy: a�?The AdWords system works best for everybody a�� advertisers, users, publishers, and Google too a�� when the ads we display match our users’ needs as closely as possible. Relevant ads a�� earn more clicks, appear in a higher position, and bring you the most success.a�?
This makes sense in theory. Google looks at every keyword and makes a relevancy judgment. High-relevancy keywords get a a�?greata�? QS rating, so-so keywords get a�?OK,a�? and bad keywords get a�?poor.a�? Simple, right? However, QS is also an effective way for Google to predict and maximize profitability. In other words, the most relevant ad doesna��t necessarily win; the most relevant and profitable ad wins.
Before Google, the Overture system ranked ads solely on the cost per click (CPC) advertisers were willing to pay. Google revolutionized paid search by creating a yield management system where ads were ranked by maximum CPC multiplied by clickthrough rate (CTR), then further altered the system by adding QS into the ranking algorithm Over time, CPC ranking changed.
2001 a�� Position = CPC
2002 a�� Position = CPC x CTR (effectively a cost per thousand, or CPM system)
2005 a�� Position = CPC x CTR x QS (CPM plus a statistical prediction of ada��s profitability)
By understanding that QS is really a way for Google to improve both profitability and relevance, you can save yourself a lot of headaches trying to guess why your score has dropped.
Manual Or Automatic Smackdown?
Two types of QS smackdowns exist, and responses should vary accordingly. A manual smackdown occurs when someone at Google reviews your account and determines it to be of very low quality. Basically, youa��ve been blacklisted, and all your keywords immediately receive the lowest QS possible. Minimum first page bids increase to at least $10. Deleting keywords or adding new keywords has no effect, nor do ad text changes or landing page adjustments. Generally you find that after receiving one or two clicks, your account shuts down for the rest of the day.
Why would Google manually blacklist an account? Usually, ita��s due to a clear violation of AdWords policy, such as advertising illegal products, double-serving (owning two or more sites but masking ownership and serving ads on the same keywords), abuse of user privacy (having a site solely to collect email addresses), or fraudulent ad claims. Blacklisting may result from competitor complaints, but sometimes Google has just audited your account and didna��t like what they saw.
The only recourse is a direct appeal to Google, usually through your account rep. If you discover a policy violation, correct it prior to asking your account rep for help, and make sure that everyone in your company understands this violation must not happen again. In other cases, you may have no idea why youa��ve been blacklisted, and neither will your account rep. Worse yet, you may find out that the reason is not even fixable. One client was told that their site content was a�?too similar to other advertisers,a�? and while it seems unfair for Google to prevent one advertiser from showing up while another with the same content continues to buy ads, this is the nature of the beast. Therea��s no recourse beyond continuing to plead your case.
Manual blacklists are rare, so most advertisers who have QS problems are dealing with the second type of smackdown a�� the automatic smackdown. Here, Googlea��s algorithms have determined there is something problematic about your ad campaign. An automatic smackdown can impact a few keywords or broadly affect your entire account. The good news is that automatic QS smackdowns are often correctable.
Factors You Can Control
Because Google doesna��t publish the QS algorithm, search marketers are left to infer relevant factors and have discovered that some QS factors are entirely out of your control. For now, leta��s focus on factors that you can actually manipulate.
Clickthrough Rate (CTR)
Virtually everyone agrees that CTR is the leading factor in QS calculations. Recalling the a�?relevancy plus profita�? concept, this makes sense a�� CTR is the best indicator that users find an ad relevant, plus ita��s the best way for Google to maximize profit. Other tactics that drive CTR (that some search marketers consider as separate QS factors) include: buying relevant, targeted keywords; creating targeted ad groups with targeted ad copy; and adding the keyword into your ad copy.
There are two important caveats to this a�?CTR is kinga�? philosophy. First, dona��t become obsessed with CTR to the point that you forget about the metrics driving your business. An ad that says, a�?Ia��ll pay you $5000 if you click here!a�? may get a huge CTR, but when business objectives are based on cost per conversion, it will win the CTR battle and lose the PPC war. Second, too much targeting can get you in trouble. In theory, you could use a one-to-one ratio of ad groups to keywords, creating the most targeted ad text possible. But in addition to bringing excess complexity to your AdWords account, this can also negatively impact QS. Ia��ll discuss this shortly.
Landing Page Relevancy
Search marketers have always been concerned about optimizing landing pages from a conversion rate perspective, but QS requires that we stick our nose into site design on an entirely different level. In Googlea��s QS guidelines, landing page factors are broken into four groups:
- Relevancy a�� a�?Users should be able to easily find what your ad promises.a�?
- Originality a�� a�?Feature unique content that can’t be found on another site.a�?
- Transparency a�� a�?Your site should be explicit in a�� the nature of your business; how your site interacts with a visitor’s computer; and how you intend to use a visitor’s personal information.a�?
- Navigability a�� a�?The key a�� is making it easy for users to find what they’re looking for.a�?
As with CTR, an obsession with satisfying Googlea��s AdsBot can also result in bad business decisions. I recommend building landing pages that drive the highest conversion rate for your business, and then assess your QS landing page score (found by hovering over the magnifying glass next to a keyword in the AdWords interface and clicking the details and recommendations link).
If you find that your QS is a�?poora�? and your account performance appears to be negatively impacted by this rating, go back and make changes to your landing page until you see your QS ranking improve.
Keyword, Ad Group, And Account History
History a�� at the keyword, ad group, or account level a�� has always been an important part of the AdWords system. Think about this from a Google profitability perspective. If two advertisers each bid $10 on the same keyword, but one has 10,000 clicks with a 25% CTR and the other has 1 click with a 30% CTR, it is safer for Google to serve the advertiser with thousands of clicks, because of the greater ability to statistically predict the profit from serving them. QS also takes history into consideration. Many advertisers note that most new keywords show up as a�?poora�? across the board, but over time, they improve to a�?OKa�? or a�?great.a�? In essence, this is an a�?SEM sandboxa�? that puts new keywords into a holding pattern until Google can better predict their performance.
Recall how earlier I suggested creating a one-to-one keyword to ad group ratio could actually hurt QS? Herea��s how. If you have 5,000 keywords and 5,000 ad groups, the odds that Google will be able to get ad-group-level predictability on most of your ad groups drop dramatically. As a result, Google may just assign a a�?poora�? QS to most, resulting in the keywords never being served. By contrast, if you consolidate those 5,000 ad groups into 100 still-targeted ad groups, you give Google more ad-group-level data to work with, which may bring an improvement in ad-group-level QS and more clicks.
Other Factors To Consider
There are an array of other factors that could be involved. For example, with geographic location, if you are a Chicago auto repair shop but try to advertise in Seattle, you may receive a geo-specific low QS. If youa��ve made past QS mistakes on a display URL, that URLa��s historical performance may affect QS so much that you should consider a different URL going forward. Even the number of words in a keyword query may be a problem, as Google reps have stated that keywords longer than four words, or tokens, are assumed to have low QS. However, these are such nuanced considerations a�� and have relatively small impact compared to others a�� that the effort you put into thinking about them should be minimal.
Factors Beyond Your Control
On the other hand, some factors impacting QS are almost completely out of your control, which can be the most frustrating part for advertisers.
Not all money is equally green to Google, which explicitly identifies site types that will automatically receive poor landing page quality scores and others that are likely to do so. The a�?definitely punisheda�? sites include data collection sites, arbitrage sites (a�?made for AdSensea�? or MFA sites), and malware sites. The a�?likely punisheda�? sites include eBook sites, get-rich-quick sites, comparison shopping sites, travel aggregators, affiliates (generally also including lead generation sites). If your site falls into one of these categories (especially the a�?definitesa�?), the best response to a QS smackdown may be to move into a different business.
Historical Keyword-Specific Performance
When Google determines keyword-specific QS, it considers the historical performance of all advertisers who have ever tried to advertise on that keyword. Translation: if other advertisers have failed to create relevant, high-CTR campaigns on a specific keyword in the past, you will likely find a big QS hurdle in front of you trying to buy this term.
For example, leta��s say you want to buy the keyword a�?Britney Spears photosa�? to drive people to your Britney Spears photo archive. This may be the most targeted keyword possible, but because thousands of advertisers have tried to monetize this phrase with random ads like a�?Who do you like better, Britney Spears or Jewel?a�? Google assumes that most ads on this keyword will have a very low CTR. Youa��ve done nothing wrong a�� indeed, youa��ve tried to be very targeted a�� but you are still penalized for other peoplea��s mistakes.
Putting It All Together
You can definitely drive yourself crazy trying to game the QS algorithm. As with most things, the best course of action is to run your business first and worry about QS only when you have to. It turns out that many factors contributing to high QS a�� like high CTR, relevant landing pages, protection of user information, and good ad text a�� are also best practices to help you grow the profit from your AdWords account anyway. Do all of that as you should and QS worries may become a thing of the past.