Quality Score Surprise! #1 Ad Position Doesn’t Improve QS

SMS Text

What I love about being an Advanced Experienced AdWords Expert as I have been for years now (and I am using the word LOVE sarcastically here) is that I keep finding out things that I have assumed for years are wrong.  And that if I read Dave Barry too much- and I am not making this up- I start to write like him.

It is only because I have teachers in the form of AdWords Dedicated Account Representatives that I have these frequent stunning realizations.

So here’s another myth by the wayside:

If you bid more, your cost per click will go down because your CTR goes up and your Quality Score goes up because QS is so dependent on CTR.

Ok maybe you didn’t follow that- so here’s the myth in visual form:



Here’s what the AdWords Lady said:

CTR is normalized in regard to quality score so ads in top positions which usually have a high CTR don’t have an unfair advantage. While higher positions tend to have a higher CTR, we expect higher positions to have a higher CTR.

For example, we know that when an ad appears in position 4 it has an x% CTR.  When that same ad moves up to position 1, with no other change, that same ad has a CTR of x+y% CTR.  When we calculate quality score, we remove the impact of the y variable so that we just have the intrinsic CTR of the ad, independent of position.  Therefore, while increasing bid may increase CTR, that increase alone should not improve the advertiser’s quality score.

The below blob posts give more insight into this topic as well.



That really got my goat- not only does it appear that my myth may have been true up until October of 2008, but it also appears I should be reading the Google AdWords blob more carefully. And that got my goat even more- I didn’t know Google had blobs. There’s a chance she meant “blogs” but since Google has Waves now, I think they really could also have Blobs. Or maybe I’m just making a stupid joke about her being dumb because I felt dumb. I don’t know, I’ll ask my therapist when she gets back in town.

So, since AdWords shows ads at a variety of positions and really only tells you the average position, they somehow gather a normalized CTR via some statistical magic I’m not educated enough to understand, and use THAT as that CTR to multiply your bid by. Voila!

Very cool. And smart, because if they didn’t normalize CTR, whoever was in the top position would have an unfair advantage.

It also means that you have to care more about the effectiveness of your ad copy again. Because THAT’s what determines your normalized CTR and quality score and required bid for a given position.

Ta da.

Brian Carter
Brian is author of The Like Economy: How Businesses Make Money With Facebook and Facebook Marketing: Leveraging Facebook's Features For Your Marketing Campaigns, How to... Read Full Bio
Brian Carter
Get the latest news from Search Engine Journal!
We value your privacy! See our policy here.
  • Yes, Q.S is highly dependent on the CTR, if your CTR are high on the average (I have CTR of 5% for my offline business).

    When your CTR and Q.S have good results, your Average CPC will also drop dramatically too.

    This will require time and great result in the long run.

  • Your point is accurate, but kind of the background for what I’m discussing here, which is whether or not the increased CTR you get from bidding higher and getting higher ad positions would increase QS and therefore lower your CPC.

  • Your point is accurate, but kind of the background for what I’m discussing here, which is whether or not the increased CTR you get from bidding higher and getting higher ad positions would increase QS and therefore lower your CPC.

  • I agree with you in that it makes sense. I always assumed they used a very similar method in normalizing as I believe they do with organic results.
    And thanks for making me launch my drink all over my computer with ‘the blob’ comment.

  • If this is completely true, can anyone explain to me how this works: Step 1: Start a brand new campaign and bid to the moon. Step 2: Wait few days. Step 3: BAM! my quality score goes up & the CPC drops down to only 1/5 of what it was to start with. This just happened on yet another campaign from 3 days ago. There’s got to be more here than meets the eye.

  • Brian, do I hear right? Are people propagating this myth in 2009!?! Are we getting less informed over time, not more?

    And what’s more… I can’t believe you suggest it might not have been a myth say as recently as 2008… well, rookie, not so fast. 🙂

    Dealing with QS and CTR being normalized for ad position, this fact (post introduction of Quality Based Bidding in Aug. 2005) is fairly easy to dig up having been shared by Googlers in forums and by presenters at SES in the Ads in a Quality Score World panel – see for example:

    This has basically been true since I think about three or four months into AdWords “Select”‘s rollout in 2002. For a very brief period at the start, you might have been able to do some kind of “#1 position trapping move” by bidding up to #1 and then locking that advantage in forever, taking advantage of your high CTR to “bar” competitors from outbidding you. Google did away with that quickly, normalizing the evaluation of CTR by ad position. In a sense, introducing that level of complexity to the formula was a small step towards building an increasingly opaque formula that didn’t disclose its exact workings… proto-proto-QS.

    Another link drop:

    Glad you pointed out this myth. Google may still have some practical challenges with normalizing CTR to ad position, but it’s clear they try to do so and have done since 2002.

  • Ah, good catch. Never heard that one. But I’m not a rookie… don’t know a constructive way to respond to that comment.

  • I’m still wondering why this strategy still works today if Google is truly normalizing this data…? My real world test shows the technique working just 3 days ago.

  • From my own results, I still think you get an unfair advantage in Pos 1, but much less so since the normalizing of positions and CTR by Google a while back. Therefore I don’t think the normalizing is completely accurate.

  • Brian, just kidding on the rookie part. The Internet makes it no fun at all to tease someone.

  • LMAO, ya it’s easy to be misunderstood 🙂 No worries. – B

  • Hey now play nice boys 😉

  • LOL

    (that comment was too short to post so here is some parenthetical text that I didn’t want to write)

  • Awesome article. But I am still getting very mixed ideas here. On one end of the spectrum, you guys are saying Google has normalize the positioning algorithms. But on the other end, I am hearing, no. 1is still better.

    So if I just started a ad campaign, which should I opt for to maximize my efforts?

  • What works for me is to bid on the higher end at first. Visibility is your goal at first. Then your success will be up to your ad quality, ad group segmentation, and other factors. It’s much harder to start with low bids.

  • Some posts you read and you say to yourself, ‘great information and an enjoyable read’. However, I must admit I found the comments to just as informative for a change and quite entertaining ‘rookie’ lol.

    Great info.


  • Brian,

    I’ve gotten the same message from Google about how they normalize QS depending on ad position so that lower CTRs don’t impact the QS as much when ads are in lower positions.

    At the same time many AdWords reps will recommend that you start off with higher bids initially when launching a new campaign or keywords to give them a “kickstart”.

    After reading your post I wonder if some reps are still back in 2002, but personal experience dictates otherwise where starting with higher bids initially does “shock” the campaign into performing better faster.

    I think that as a campaign ages, the normalizing effect might kick in more, but there are still wide fluctuations in more competitive industries where minimum first page bids play a larger role in getting a decent ad position, CTR and QS.

    I’ve done experiments where increasing bids to match minimum first page bid requirements actually work to improve QS over time presumably because CTR also increases. (And this is in instances where there are no keyword, ad or adgroup structure relevancy or landing page issues so bids and CTR are the main culprits in low QS)

    That said, I’m not convinced that CTR/ad positions are normalized for QS as well as Google claims. But do agree that better ad copy leading to a higher CTR is probably what Google is talking about and temporarily increasing bids initially just allows you to get more traffic to test your ads more quickly and make adjustments.

  • When studying the relationship between quality score and click through rate it is very important to account for position. We have modeled client data from various clients across a range of geographical regions. It is clear that quality score is assigned on a relative CTR by position. You can get get a good idea what CTR by position is the benchmark to achieve a quality score of 10. You can see some of these results at http://www.clicks2customers.com/c2cblog

  • Brian – don’t worry about Ahn-dee – he’s always like that.

    Wait – is this mike ON?!?