NOTE: In November 2008 I wrote a five post series on Google Adwords Quality Score. To make reading and linking easier, these posts are combined below. Enjoy.
Quality Score does three things for Google:
- It acts as a bozo filter to limit or prevent ‘undesirable’ ads and advertisers
- It acts as a ‘preferred customer program’ to reward top performing advertisers
- It provides a ‘secret sauce’ that ensures nobody knows how/why certain ads are run at specific times for certain prices.
The first two are rather straightforward. These are the aspects encompassed in the ‘improving everyone’s experience’ description and rationale Google generally gives for Quality Score.
But it’s the last one that has real impact on paid search marketers.
Quality Score is Google’s way of passing judgement on and rating a number of different aspects of your paid search campaigns.
This rating is then used to make value judgements about your suitability to advertise for any particular keyword at any particular time.
And to manipulate everything the concept of auction was supposed to tell you about bidding for keywords. Yes there is an auction going on, but it’s happening in an environment where everyone has a different multiplier on their money. Some are positive, some are negative.
Imagine placing bids on ebay when you had no idea the conversion rate that was going to be used to turn your dollars into the local currency of the seller. And what if when looking at the bids or relative order of other bidders, you had no idea what conversion rate had been applied to their bids. How would you bid in that environment? Quite differently than in one that was open and transparent, that’s for sure.
There is a lot we know about Quality Score, and a lot that Google just isn’t going to tell us.
What Drives Quality Score
A cornerstone of High Resolution PPC is the fact that there is a true but over-simplified view of just about every aspect of paid search marketing.
With Quality Score, the popular notion is that there is one single specific metric calculated based on a few simple variables and attached to each keyword in your Ad-Groups.
Google fosters this impression, but a careful reading of their materials (and the comments of some very knowledgeable folks) suggests it’s not that simple. There are a number of different Quality Scores or QS components which are calculated independently and used separately or collectively in different situations.
And these scores aren’t static. Quality Score is computed in real time for every search. The calculation is based not just the keyword but on the unique combination of search query, keyword, the text-ad selected, the searcher’s geography, and other variables.
While we don’t know everything about the Quality Score calculation(s), we can rank and summarize the main influencers:
- CTR is by far the largest factor, and considered at many levels – from the historic overall average CTR of your account, to the CTR of the Ad-Group the keyword is in, to the recent CTR of the specific query-keyword-textAd combination.
- Relevance is important – this requires you to keep tight topical and even literal groupings for your keyword within an Ad-Group and ensure that the specific terms (or clear & common synonmns) appear within each matching text ad and on the target landing page.
- ‘Other Factors’ are also considered although they probably play a generally minor role – these include the geography of the user (do you’re ads get higher CTR’s in FL, you’ll get a higher QS for FL searchers), the load time of your pages, the content on and linked to your landing pages, and more.
Around these basics there are a lot of details to chase down and act upon.
But the basic lessons should be learned first.
- The impact of Quality Score on your campaigns in enormous. Even without knowing exactly how it’s being calculated or applied, we need to understand the general goals of Quality Score and execute our campaigns accordingly. Selectively or occasionally doing these things isn’t going to to work.
- Quality Score rewards things you want to do anyway. Do not tolerate poor performing click-through-rates. Narrowcast your Ad-Groups from query to landing page. Treat your visitors with respect. Doing the basic right takes you a long way, and yet of the hundreds of accounts I review each year, very few uniformly get these things right.
Applying these lessons in a rather simple fashion could in many cases deliver excellent Quality Score results.
Want a quick-fix Quality Score strategy example? Try this:
- Go through your Ad-Groups, look at the text-ads that are running. Delete any ads getting CTRs 50% lower than your top performers.
- Go through the keywords in each Ad-Group. If there are keywords getting performing 2X worse than your average CTR, pause them or move them to a new ‘Rehab’ Ag-Group.
- Visit your landing page. Think like a prospect and fix anything that would stop you from understanding, trusting, or moving forward.
To learn more about Quality Score, and hear a more detailed approach to applying the deep facts to improving your campaigns, attend our Quality Score Webinar today (Tuesday Nov 25) at 12:00 EST.
Quality Score Final Thoughts
Quality Score is the secret sauce in Google Adwords. It plays a huge rule in nearly everything advertisers care about; when and where ads run, how ads rank, and what ads cost-per-click.
Quality Score – along with their Broad Match and Automatic Matching formulas – give Google a huge set of levers and dials to play with at will.
As they do, our ads and ad-budgets jerk around like marionettes.
I’m left with two conclusions:
- The lack of transparency is astounding. Everything Google is doing is reasonable and legitimate from a business perspective. They’re optimizing their product to maximize their revenue, and trying to make their customers feel good so they spend a lot of money and are happy about it. But advertisers can’t and don’t know what’s going on in the black box of Quality Score. We’ve got some clues, there has been more clarity recently than historically, but the playing field we’re on is far from level.
- Paid search managers must prioritize Quality Score management. This means a lot of things as we’ve discussed; small tight keyword groupings, focused text-ad and landing-page copy, paying attention to the published Quality Score numbers themselves, starting new campaigns slowly, not letting losers hang out, even in dark corners of your campaign, and much more. It all adds up to an increase in workload, responsibility, and the need for specialized tools to have any chance to real success
I used to think of Quality Score as an ‘other factor’ in campaign management and success.
Now I think it’s one leg in the three-legged stool of the PPC process.
Campaign organization, Bidding, and Quality Score must all have equal and appropriate attention to make paid search really work.
(Although a more complete picture is the Target-Value-Satisfy-Understand model of ‘High Resolution PPC’ with Quality Score being a piece of the Valuation component.)
Where Are The Quality Score Tools?
As a final point, given this realization, I must say that the tools for helping manage the importance of Quality Score are sorely missing. Right now the Quality Score number itself is available only inside of Adwords – although it is now in the API so we can expect third party vendors to support it soon.
But the broader issues of focus, alignment and relevance between components, and the impact Quality Score has on bidding and position is almost entirely unsupported or assisted by the tools on the market.
The ClickVariance variable in ClickEquations does help identify AdGroups with keywords that are too diverse from a performance perspective, which is a start in the right direction.
That makes the reality of taking advantage of whatever understand we’ve gained about Quality Score very difficult. Today it will require a lot of manual effort and hours of work.
But from a ClickEquations viewpoint it’s an opportunity we’ll address.
Quality Score Q&A
In our Quality Score Webinar with Bryan Eisenberg (If you missed it, you can now watch the replay) there were way too many questions to answer during the event.
This is the first of several posts in which these questions will be answered. We’ll split the answers between here and TheGrok.com, but keep linking to more as they’re posted.
Have more? Put ‘em in the comments. Disagree with any of the answers. Comment please!
Q: What does ‘removing ads from the bottom’ mean?
A: I think the point you’re referring to was part of the discussion of text-ads. Since most people run 2-4 versions of their ad to test for better CTR and conversion rate, it’s a good idea to regularly remove the ads getting lower CTR (and/or Conv Rate) and add new ones in an attempt to create a new ‘kind of the hill’.
Q: What about long tail keywords?
A: The only thing that matters about keywords relative to Quality Score is the CTR they generate and their relevance to the ads, queries, and landing page. The concepts of head and tail don’t factor in.
Q: How does Google determine if a landing page is relevant? Someone might actually find the page useful but still bounce back to Google to click another ad.
A: The primary determinant of relevance is semantic – do the words on the page match the words in the query and keyword purchase, either literally or at least contextually. Bryan mentioned the idea of Google measuring bounce rates and using the fact that someone came back and did another search or clicked another ad as one of their clues, but that is likely less significant. If a page has good relevance but many users bounce that’s better than if it has no relevance and users bounce.
Q: Is there a way to check the Quality Score of your competition?
A: No you can’t see their score on an individual keyword basis, or figure out their CTR(s). But you can certainly assess the relevance of their text ads and landing pages. Finding keywords that have low relevance – because they tend to be broad matched and lumped into a more general group – and then tightening up your relevance to that exact word/topic, would be a way to get an advantage.
Q: What’s a good click-through-rate?
A: As we mentioned in the Webinar, their is no real answer for this given the wide range of keywords, queries, ads, and situations. Long ago Google wouldn’t run ads with less than 1% CTR for very long, and while that is no longer true it is rare that less than 1% is a very good CTR. For brand-terms on the other hand I’ve seen 30-40% common in some cases. Just depends is the real answer. Testing some reasonably broad types of text-ads should help you find the range for any keyword. But writing good text-ads is pretty hard for many people.
Q: Does the Quality Score of one account in My Client Center effect the other accounts in that Client Center?
A: No. All Quality Score issues are constrained on one Google Account.
Q: If QS suffers when keyword and query aren’t tightly aligned, should you use Exact and Phrase match early in a campaign and delay Broad until a good QS is established?
A: No because QS is only calculated ‘as if’ all keywords were exact match. Not sure exactly how this works but it suggest Google is trying to not penalize you because they match a broader set of queries to your keywords.
TIP FROM PARTIPANT: Quality Score is in the Keyword/Placement Report in Google. I had said it was only under the pop-up in the management window.
Q: If you have two ads in an AgGroup, does the QS as displayed reflect the lower of the two? Should you just run one until ‘established’?
A: Probably not, more likely an average. Running one wouldn’t help because as soon as you ad another you’re in the same situation (although you may have some good history built up, but that probably counts much less than whatever is currently running.)
Q: Do the other search engine use a Quality Score?
A: There are other search engines? Hm. Actually you’ve stumpeded me. Comments?
Q: Does testing ads frequently reduce Quality Score?
A: In theory this makes sense – as you’re restarting the history and calculation, but my guess is that if a keyword in an AdGroup has a good history, you get the benefit of the doubt for a while until a new text-ad proves lousy. Google wants to encourage testing. I would never recommend a full new set of text-ads, rather leave known winners and fold new ones into the mix until they prove themselves.
Q: Does using the keyword in the title and description and URL produce a high Quality Score?
A: Repeating a keyword in the text-ad and landing page and yes even the URL (not often possible certainly does produce a part of the relevance that delivers a good Quality Score. But remember that CTR is the most important attribute, there are other components too (like load time etc.) and almost certainly the relevance calculation is somehow more complex. So nothing other than doing everything right (with everything being a set of things and measures we don’t know) will guarentee a high Quality Score
Q: What is the most optimized landing page in terms of Quality Score?
A: One that is perfectly relevant to the keyword, search query, and text in the ad, and gets people to stay (low bounce rate) and probably move forward (another click if not a conversion). Beyond that, nobody knows.
Q: Can high impression low Quality Score keywords impact low impression keywords?
A: Yes. Their are overall Quality Score for an Adgroup and for the Account components which impact everything. This is why improving or removing low CTR keywords and text-ads is important to overall and broad Quality Score.
Q: How many impressions should an ad get before evaluating it’s CTR?
A: It depends upon margin of error you’re willing to tolerate. A simple calculation for the margin-of-error in any sample size can be calculated as 1/SquareRoot-of-Sample-Size. So, with 100 impressions your margin of error is 1/10 or +/-10%. For an ad with a CTR of 7%, this is clearly an unacceptable margin of error. For a brand keyword/ad where CTR is in the 40-50% range, this margin of error is probably acceptable. A good general guideline is that brand ads should probably see something on the order of 100-150 impressions and that general ads should see at least 200-300 impressions before making any meaningful decisions. More impressions yield even greater confidence levels.