Quality score is a system AdWords uses to pass judgment on each of your keywords. They score every one, reflecting how well that keyword has done in the past and how well it’s expected to do in the future.
This score has impact. It determines how often your ads are shown, where they appear, and how much you pay for every click they generate.
The process by which these scores are determined is not clear – all we have is a vague set of explanations we’re asked to accept, and loose set of clues as to the behavior that they suggest in order to earn good or even great scores.
The power and complexity of quality score prompted me to spend a great deal of time over the past few years learning how it really works, determining the specific ways in which it impacts our results, and documenting the strategies and tactics that enable advertisers to better control their own fates and budgets.
Google was kind enough to assist, granting interviews and answering questions by email so many of the inscrutable elements of the system could finally be explained. The result is a book called ‘Quality Score in High Resolution‘, scheduled for first release in June 2011.
For the next few weeks, I’m going to discuss some of my most interesting discoveries and conclusions in a series of posts here on Search Engine Land. At the upcoming SMX Advanced in Seattle, I’ll present the ‘5 Biggest Surprises about Quality Score’ at the opening session of the paid search track.
Quality Score Is For Your Own Good
One of the things that struck me as I gathered and organized all the facts (and set aside a lot of myths) about quality score, was that broadly speaking the strategies and tactics that produce the best quality scores also produce the best economic returns. In other words, the steps Google is trying to get you to take to improve quality score really are in your own best interest.
More than anything else, quality score measures click-through rate – we all know that by now. Obviously, a higher CTR means you get more visitors and hopefully more revenue. But aiming to ‘increase CTR’ is a generic goal and one that doesn’t provide specific clues as to how to improve performance. But quality score is not driven by the ‘blended average’ click-through rate metrics that AdWords reports in the AdWords interface.
Instead, it’s determined by a more focused set of CTR measures:
- How each keyword performs with each different text-ad in its ad group,
- How the display URLs that appear in your ad copy perform across all the ads that share them,
- The aggregate and historic CTR performance of all keywords in the account, now and previously,
- How performance varies based on the unique search query matched to the keyword,
- Plus other similarly nuanced measures.
By understanding exactly what Google is measuring, the door opens to taking specific actions that can improve the metrics that really impact quality score. And produce better economic results too. A couple of points to consider:
- Paying attention to keyword and text-ad pairs, instead of thinking of those as two separate elements, forces you to create smaller, more tightly targeted ad groups because otherwise you can’t control which ads are matched to which keywords.
- Paying attention to aggregate and historical performance forces you to accept that a bunch of poor performing keywords in a ‘testing’ ad group can impact the results of your best performing keywords; and the damage done by a poor search manager can linger long after they’re gone.
- Paying attention to display URLs forces you to think about that generally ignored variable, the clues you’re sending to searchers, and the consistency of your visible URL assignments.
- Paying attention to geography forces you to think about the differences in how your product or service offering is being accepted in various parts of the country. If you never get an order from Arkansas, why are your ads showing there?
These are just examples. The point is that quality score, if we know how it really works, gives us a new perspective on the elements of our campaigns and ways we can think about and improve them.
AdWords Is Telling You What Not To Do
I think the biggest gift quality score gives is the idea – if you’re open to it – that you shouldn’t bid on every keyword that you think is somehow relevant to your target market. There are keywords, often many, that make sense to you but that despite your best efforts don’t earn respectable quality scores in your account. (Let’s say anything less 6 is not respectable.)
When this happens, there are three possible explanations:
- You have more work to do improving the attributes that really matter to quality score. Most people seem to assume this is their only option.
- AdWords could be making a mistake and assigning an incorrect quality score or applying some bad or mistaken information. That happens, but rarely, and if you can prove it you should take your case to your AdWords representative.
- The keyword in question just doesn’t work for you, and it should be paused or deleted. Prospects are telling you (it’s their actions that Google is measuring) that for whatever reason the queries being attracted or the ad copy you’ve written or the geographies you’ve targeted are not pleasing them. In effect they’re asking you to stop showing them uninteresting ads, and stop wasting your money – on the keyword as currently configured.
Thinking of it that way, maybe you’ll be inspired to find a better ad group organization and write better ad copy. Maybe you’ll review your search queries and try again with slightly different versions of the keyword with different match types. That can often work wonderfully.
But sometimes the right response is to say ‘thanks for pointing that out’ and turn off the keyword and spend your energy and dollars on the many more productive keywords in your account.
In the next installment, we’ll examine how quality score determines which ad auctions your keywords are eligible to enter, and the unexpected ways that increasing quality score can lower average positions and increase average CPC’s.
Opinions expressed in this article are those of the guest author and not necessarily Search Engine Land. Staff authors are listed here.