The ROI Revolution Blog

Introduction to Google's Ad Auction, Part 1: Quality Score

March 24, 2009

ctrpie.JPGLast Wednesday our agency's Google rep sent over a video put together by Google's Chief Economist, Hal Varian that very clearly explains Google's Ad Auction and Quality Score. The 9 minute video is definitely worth viewing at least once to get a better understanding of two large parts of the AdWords system that were, until now, pretty hard to quantify.


In this two part blog series, I will explain what Quality Score actually is, how the ad auction works, and how Quality Score determines your ads' positioning and costs per click using the video as my guide. Today's article will cover the ever elusive concept of what Google's Quality Score really is.

Google's reasoning for creating Quality Score is to make sure that the best interests of all parties included in the search experience (the advertiser, the user, and Google themselves) are taken into account when a search is performed on Google.com. Below, I've explained the motivation to appeal to each party's interest:


  • Advertisers want to show relevant ads so that users will click on their ads and land on their site, hopefully intending to purchase something.

  • Users want to see relevant ads so they can easily find what they're looking for in the shortest amount of time.

  • Google wants the best experience for both advertisers and users so that advertisers continue to use the AdWords program and so users continue to use Google.com and Google's other host of products.

Until recently, Google's "Quality Score" was an abstract concept with little to no definition. Advertisers wanted to get the highest Quality Score possible, but didn't really know what they were striving for.

Towards the end of 2008 and even more in Q1 of this year, Google has been slowly revealing what is included in Quality Score. The video explains that Quality Score consists of clickthrough rate (CTR), relevancy, and landing page quality. The reasons Google places emphasis on each of those components is outlined below:


  • CTR is the most heavily weighted component accounting for over 50% of Quality Score. Google looks to CTR as a means of getting users' feedback on ad text. If an ad has a high clickthrough rate, this must mean that many users are finding it relevant and useful to what they're looking for.

  • Relevancy is the second largest component. Google takes into account the relevancy of a keyword to an ad as well as to the user's search query. Relevancy is Google's way to display only useful ads to users. It also prevents advertisers from simply buying their way into an ad auction, when their product or service is irrelevant to the search.

  • Landing Page Quality is the third component of Quality Score. It holds the least weight, but is still a very important part of the equation. The video lets us know that, in Google's opinion, a landing page that is high quality contains relevant, original content, is easily navigable, and has quick load times, with minimum pop ups or pop unders. It should also be completely transparent about the nature of your business, how your site interacts with the visitor's computer, and how you intend to use the visitors' personal information.

It is very helpful for us to now see exactly what we should be working on in our clients' AdWords accounts to improve their Quality Score. We're also able to focus our time on areas where we feel we can be the most productive.


The high emphasis placed on CTR shows me that testing ad text and ad headlines in an effort to improve the relevancy to keywords and the user's query is of utmost importance. While focusing on improving the relevance of your keywords to your ads and your ads to high quality landing pages will certainly pay off, it is now clear what Google wants from us the most.

In tomorrow's article, I will use the video to explain how Google determines your Ad Rank as well as each advertiser's click cost. We'll also delve into how improving your Quality Score can help achieve higher positions for less cost.

Stay tuned...

Click here to read Part 2, AdRank and CPC

Take the next step. Ecommerce retailers spending at least $500/mo. in AdWords qualify for a free 20-minute Google AdWords account review.

Comments

parky said:

ROI Revolution you guys rock and I dont know where else to turn so I hope you dont mind me throwing this question out here, thanks for all your articles.

Situation is this:

Have a site with product pages that dont see enough traffic to each individual product page for a multivariate test (Google website optimizer) without waiting months for results. For example lets say I have 100 product pages each seeing about 200 views a week. I want to setup a simple multivariate test with different checkout buttons and locations. For this example lets use total of 10 combination's. If I run that multivariate test on even my most popular product page with only 200 views it will take forever to get solid results and to me Google was just never have enough solid data to calculate. Now if i could aggregate all the product pages into a test 100x200 so 20k views a week that is the kind of data i need GWO to see. So my questions is this:

Does the Javascript really care about the Test Page URL or even Conversion Page URL for that matter as long as the scripts are referenced. In other words I want to make 100 product pages look like one test page they will all have the same scripts on top of page and at bottom of page and then the checkout will always have the same conversion script. Has anyone tried something like this or any expert in JavaScript out there know if this will work? Any insight would be much appreciated.

-parky

March 25, 2009 4:18 AM

Shawn Purtell, Senior Web Analytics Engineer Author Profile Page said:

@parky:

The method you suggest should actually work just fine. Google Website Optimizer doesn't care about the test and conversion page URLs that you put in when creating an experiment - they only use those for the validation step.

If you put the same control, section, and tracking scripts on all product pages, you should be able to run your experiment just fine, and it will aggregate the data for all product pages. If a user views more than one product page in the same visit, they will see the same combination the entire time, so you shouldn't have any problems with consistency either.

Hope that helps!

March 25, 2009 8:08 AM

Dan PPC PROZ said:

apparantly there are two formulas, one for first page bid and one for adrank.

apparantly first page bid on new accounts is weighted more heavily on landing page relevancy than we originally thought, see my post...

March 28, 2009 5:36 AM

Erin Ewasyshyn, Strategy Manager - Paid Search Marketing Author Profile Page said:

@Dan, Thanks for reading and I enjoyed your post. After a discussion with our rep, I've found that landing page quality is not a bigger component of QS for newer accounts than for older accounts. It has to do with whether or not Google has evaluated the new account's landing page or not. If the new account sends traffic to a page Google hasn't seen, landing page quality will have no impact on QS at all until the site is evaluated. If the new account sends traffic to a page that Google has already crawled before and determined it not to be a good site, then the new account will have QS problems right from the beginning. I hope this helps!

March 30, 2009 1:41 PM

John Rodriguez said:

Erin, great post on QS. With limited hours in a day, it helps to know where you'll get the most bang for your buck.

It makes complete sense that CTR makes up the lion's share of QS.

Letting searchers "vote" on an ad's relevance goes a long way to ensuring that Google provides a positive experience to their users -- assuming that the actual landing page is relevant... LOL

From a business perspective, it makes a heckuva lot of sense as well -- ads with high CTR generate revenue. ;)

April 23, 2009 9:04 PM

Erin Ewasyshyn, Strategy Manager - Paid Search Marketing Author Profile Page said:

@John: Thanks for reading. I've also written an article on why we shouldn't simply let Google decide which of our ads are best (Optimize ad rotation) based on CTR, because while clicks are good for Google (and admittedly good for advertisers as well), we ultimately want *conversions* which is how most of us generate revenue! Check it out if you're interested: http://www.roirevolution.com/blog/2009/01/googles_recommendation_to_optimize_ads_and_why_its.html

April 25, 2009 10:54 AM

Alex said:

I am a newbiew and found your site while surfing.

Thank you for the great information.

I have one question on the info in the charts.

In the first chart you show Alison with a Max CPC of .05 and a Quality Score of 16.

I thought the range was 1-10?

September 11, 2009 2:51 PM

Erin Ewasyshyn, Strategy Manager - Paid Search Marketing Author Profile Page said:

@Alex: I had a similar question a few months back and passed it on to our Google rep. His reply was "The 1-10 scale you see in the UI is a proxy of your quality score--it's not your actual quality score, which has a much broader range of values than the integers between 1 and 10. Similarly, the '16' and '.5' are meant to illustrate the ads quality process, rather than reflect the actual metrics we use in the auction. In addition, bear in mind that the 1-10 score in the UI is your 'core' quality score, or QS1--a measure of your quality based on Google.com, exact match traffic. For each variation or match of that keyword, you'll have an independently calculated quality score specific to that variation. Hence if you're bidding on "red cars" and your reported QS is 7/10, that doesn't mean your QS will be 7/10 on the matching query 'used red cars.' Instead, your QS will reflect your performance on that specific query and other relevance factors. " I hope this helps!

September 16, 2009 12:01 PM

parky said:

ROI Revolution you guys rock and I dont know where else to turn so I hope you dont mind me throwing this question out here, thanks for all your articles.

Situation is this:

Have a site with product pages that dont see enough traffic to each individual product page for a multivariate test (Google website optimizer) without waiting months for results. For example lets say I have 100 product pages each seeing about 200 views a week. I want to setup a simple multivariate test with different checkout buttons and locations. For this example lets use total of 10 combination's. If I run that multivariate test on even my most popular product page with only 200 views it will take forever to get solid results and to me Google was just never have enough solid data to calculate. Now if i could aggregate all the product pages into a test 100x200 so 20k views a week that is the kind of data i need GWO to see. So my questions is this:

Does the Javascript really care about the Test Page URL or even Conversion Page URL for that matter as long as the scripts are referenced. In other words I want to make 100 product pages look like one test page they will all have the same scripts on top of page and at bottom of page and then the checkout will always have the same conversion script. Has anyone tried something like this or any expert in JavaScript out there know if this will work? Any insight would be much appreciated.

-parky

March 25, 2009 4:18 AM

Shawn Purtell, Senior Web Analytics Engineer Author Profile Page said:

@parky:

The method you suggest should actually work just fine. Google Website Optimizer doesn't care about the test and conversion page URLs that you put in when creating an experiment - they only use those for the validation step.

If you put the same control, section, and tracking scripts on all product pages, you should be able to run your experiment just fine, and it will aggregate the data for all product pages. If a user views more than one product page in the same visit, they will see the same combination the entire time, so you shouldn't have any problems with consistency either.

Hope that helps!

March 25, 2009 8:08 AM

Dan PPC PROZ said:

apparantly there are two formulas, one for first page bid and one for adrank.

apparantly first page bid on new accounts is weighted more heavily on landing page relevancy than we originally thought, see my post...

March 28, 2009 5:36 AM

Erin Ewasyshyn, Strategy Manager - Paid Search Marketing Author Profile Page said:

@Dan, Thanks for reading and I enjoyed your post. After a discussion with our rep, I've found that landing page quality is not a bigger component of QS for newer accounts than for older accounts. It has to do with whether or not Google has evaluated the new account's landing page or not. If the new account sends traffic to a page Google hasn't seen, landing page quality will have no impact on QS at all until the site is evaluated. If the new account sends traffic to a page that Google has already crawled before and determined it not to be a good site, then the new account will have QS problems right from the beginning. I hope this helps!

March 30, 2009 1:41 PM

John Rodriguez said:

Erin, great post on QS. With limited hours in a day, it helps to know where you'll get the most bang for your buck.

It makes complete sense that CTR makes up the lion's share of QS.

Letting searchers "vote" on an ad's relevance goes a long way to ensuring that Google provides a positive experience to their users -- assuming that the actual landing page is relevant... LOL

From a business perspective, it makes a heckuva lot of sense as well -- ads with high CTR generate revenue. ;)

April 23, 2009 9:04 PM

Erin Ewasyshyn, Strategy Manager - Paid Search Marketing Author Profile Page said:

@John: Thanks for reading. I've also written an article on why we shouldn't simply let Google decide which of our ads are best (Optimize ad rotation) based on CTR, because while clicks are good for Google (and admittedly good for advertisers as well), we ultimately want *conversions* which is how most of us generate revenue! Check it out if you're interested: http://www.roirevolution.com/blog/2009/01/googles_recommendation_to_optimize_ads_and_why_its.html

April 25, 2009 10:54 AM

Alex said:

I am a newbiew and found your site while surfing.

Thank you for the great information.

I have one question on the info in the charts.

In the first chart you show Alison with a Max CPC of .05 and a Quality Score of 16.

I thought the range was 1-10?

September 11, 2009 2:51 PM

Erin Ewasyshyn, Strategy Manager - Paid Search Marketing Author Profile Page said:

@Alex: I had a similar question a few months back and passed it on to our Google rep. His reply was "The 1-10 scale you see in the UI is a proxy of your quality score--it's not your actual quality score, which has a much broader range of values than the integers between 1 and 10. Similarly, the '16' and '.5' are meant to illustrate the ads quality process, rather than reflect the actual metrics we use in the auction. In addition, bear in mind that the 1-10 score in the UI is your 'core' quality score, or QS1--a measure of your quality based on Google.com, exact match traffic. For each variation or match of that keyword, you'll have an independently calculated quality score specific to that variation. Hence if you're bidding on "red cars" and your reported QS is 7/10, that doesn't mean your QS will be 7/10 on the matching query 'used red cars.' Instead, your QS will reflect your performance on that specific query and other relevance factors. " I hope this helps!

September 16, 2009 12:01 PM

parky said:

ROI Revolution you guys rock and I dont know where else to turn so I hope you dont mind me throwing this question out here, thanks for all your articles.

Situation is this:

Have a site with product pages that dont see enough traffic to each individual product page for a multivariate test (Google website optimizer) without waiting months for results. For example lets say I have 100 product pages each seeing about 200 views a week. I want to setup a simple multivariate test with different checkout buttons and locations. For this example lets use total of 10 combination's. If I run that multivariate test on even my most popular product page with only 200 views it will take forever to get solid results and to me Google was just never have enough solid data to calculate. Now if i could aggregate all the product pages into a test 100x200 so 20k views a week that is the kind of data i need GWO to see. So my questions is this:

Does the Javascript really care about the Test Page URL or even Conversion Page URL for that matter as long as the scripts are referenced. In other words I want to make 100 product pages look like one test page they will all have the same scripts on top of page and at bottom of page and then the checkout will always have the same conversion script. Has anyone tried something like this or any expert in JavaScript out there know if this will work? Any insight would be much appreciated.

-parky

March 25, 2009 4:18 AM

Shawn Purtell, Senior Web Analytics Engineer Author Profile Page said:

@parky:

The method you suggest should actually work just fine. Google Website Optimizer doesn't care about the test and conversion page URLs that you put in when creating an experiment - they only use those for the validation step.

If you put the same control, section, and tracking scripts on all product pages, you should be able to run your experiment just fine, and it will aggregate the data for all product pages. If a user views more than one product page in the same visit, they will see the same combination the entire time, so you shouldn't have any problems with consistency either.

Hope that helps!

March 25, 2009 8:08 AM

Dan PPC PROZ said:

apparantly there are two formulas, one for first page bid and one for adrank.

apparantly first page bid on new accounts is weighted more heavily on landing page relevancy than we originally thought, see my post...

March 28, 2009 5:36 AM

Erin Ewasyshyn, Strategy Manager - Paid Search Marketing Author Profile Page said:

@Dan, Thanks for reading and I enjoyed your post. After a discussion with our rep, I've found that landing page quality is not a bigger component of QS for newer accounts than for older accounts. It has to do with whether or not Google has evaluated the new account's landing page or not. If the new account sends traffic to a page Google hasn't seen, landing page quality will have no impact on QS at all until the site is evaluated. If the new account sends traffic to a page that Google has already crawled before and determined it not to be a good site, then the new account will have QS problems right from the beginning. I hope this helps!

March 30, 2009 1:41 PM

John Rodriguez said:

Erin, great post on QS. With limited hours in a day, it helps to know where you'll get the most bang for your buck.

It makes complete sense that CTR makes up the lion's share of QS.

Letting searchers "vote" on an ad's relevance goes a long way to ensuring that Google provides a positive experience to their users -- assuming that the actual landing page is relevant... LOL

From a business perspective, it makes a heckuva lot of sense as well -- ads with high CTR generate revenue. ;)

April 23, 2009 9:04 PM

Erin Ewasyshyn, Strategy Manager - Paid Search Marketing Author Profile Page said:

@John: Thanks for reading. I've also written an article on why we shouldn't simply let Google decide which of our ads are best (Optimize ad rotation) based on CTR, because while clicks are good for Google (and admittedly good for advertisers as well), we ultimately want *conversions* which is how most of us generate revenue! Check it out if you're interested: http://www.roirevolution.com/blog/2009/01/googles_recommendation_to_optimize_ads_and_why_its.html

April 25, 2009 10:54 AM

Alex said:

I am a newbiew and found your site while surfing.

Thank you for the great information.

I have one question on the info in the charts.

In the first chart you show Alison with a Max CPC of .05 and a Quality Score of 16.

I thought the range was 1-10?

September 11, 2009 2:51 PM

Erin Ewasyshyn, Strategy Manager - Paid Search Marketing Author Profile Page said:

@Alex: I had a similar question a few months back and passed it on to our Google rep. His reply was "The 1-10 scale you see in the UI is a proxy of your quality score--it's not your actual quality score, which has a much broader range of values than the integers between 1 and 10. Similarly, the '16' and '.5' are meant to illustrate the ads quality process, rather than reflect the actual metrics we use in the auction. In addition, bear in mind that the 1-10 score in the UI is your 'core' quality score, or QS1--a measure of your quality based on Google.com, exact match traffic. For each variation or match of that keyword, you'll have an independently calculated quality score specific to that variation. Hence if you're bidding on "red cars" and your reported QS is 7/10, that doesn't mean your QS will be 7/10 on the matching query 'used red cars.' Instead, your QS will reflect your performance on that specific query and other relevance factors. " I hope this helps!

September 16, 2009 12:01 PM

parky said:

ROI Revolution you guys rock and I dont know where else to turn so I hope you dont mind me throwing this question out here, thanks for all your articles.

Situation is this:

Have a site with product pages that dont see enough traffic to each individual product page for a multivariate test (Google website optimizer) without waiting months for results. For example lets say I have 100 product pages each seeing about 200 views a week. I want to setup a simple multivariate test with different checkout buttons and locations. For this example lets use total of 10 combination's. If I run that multivariate test on even my most popular product page with only 200 views it will take forever to get solid results and to me Google was just never have enough solid data to calculate. Now if i could aggregate all the product pages into a test 100x200 so 20k views a week that is the kind of data i need GWO to see. So my questions is this:

Does the Javascript really care about the Test Page URL or even Conversion Page URL for that matter as long as the scripts are referenced. In other words I want to make 100 product pages look like one test page they will all have the same scripts on top of page and at bottom of page and then the checkout will always have the same conversion script. Has anyone tried something like this or any expert in JavaScript out there know if this will work? Any insight would be much appreciated.

-parky

March 25, 2009 4:18 AM

Shawn Purtell, Senior Web Analytics Engineer Author Profile Page said:

@parky:

The method you suggest should actually work just fine. Google Website Optimizer doesn't care about the test and conversion page URLs that you put in when creating an experiment - they only use those for the validation step.

If you put the same control, section, and tracking scripts on all product pages, you should be able to run your experiment just fine, and it will aggregate the data for all product pages. If a user views more than one product page in the same visit, they will see the same combination the entire time, so you shouldn't have any problems with consistency either.

Hope that helps!

March 25, 2009 8:08 AM

Dan PPC PROZ said:

apparantly there are two formulas, one for first page bid and one for adrank.

apparantly first page bid on new accounts is weighted more heavily on landing page relevancy than we originally thought, see my post...

March 28, 2009 5:36 AM

Erin Ewasyshyn, Strategy Manager - Paid Search Marketing Author Profile Page said:

@Dan, Thanks for reading and I enjoyed your post. After a discussion with our rep, I've found that landing page quality is not a bigger component of QS for newer accounts than for older accounts. It has to do with whether or not Google has evaluated the new account's landing page or not. If the new account sends traffic to a page Google hasn't seen, landing page quality will have no impact on QS at all until the site is evaluated. If the new account sends traffic to a page that Google has already crawled before and determined it not to be a good site, then the new account will have QS problems right from the beginning. I hope this helps!

March 30, 2009 1:41 PM

John Rodriguez said:

Erin, great post on QS. With limited hours in a day, it helps to know where you'll get the most bang for your buck.

It makes complete sense that CTR makes up the lion's share of QS.

Letting searchers "vote" on an ad's relevance goes a long way to ensuring that Google provides a positive experience to their users -- assuming that the actual landing page is relevant... LOL

From a business perspective, it makes a heckuva lot of sense as well -- ads with high CTR generate revenue. ;)

April 23, 2009 9:04 PM

Erin Ewasyshyn, Strategy Manager - Paid Search Marketing Author Profile Page said:

@John: Thanks for reading. I've also written an article on why we shouldn't simply let Google decide which of our ads are best (Optimize ad rotation) based on CTR, because while clicks are good for Google (and admittedly good for advertisers as well), we ultimately want *conversions* which is how most of us generate revenue! Check it out if you're interested: http://www.roirevolution.com/blog/2009/01/googles_recommendation_to_optimize_ads_and_why_its.html

April 25, 2009 10:54 AM

Alex said:

I am a newbiew and found your site while surfing.

Thank you for the great information.

I have one question on the info in the charts.

In the first chart you show Alison with a Max CPC of .05 and a Quality Score of 16.

I thought the range was 1-10?

September 11, 2009 2:51 PM

Erin Ewasyshyn, Strategy Manager - Paid Search Marketing Author Profile Page said:

@Alex: I had a similar question a few months back and passed it on to our Google rep. His reply was "The 1-10 scale you see in the UI is a proxy of your quality score--it's not your actual quality score, which has a much broader range of values than the integers between 1 and 10. Similarly, the '16' and '.5' are meant to illustrate the ads quality process, rather than reflect the actual metrics we use in the auction. In addition, bear in mind that the 1-10 score in the UI is your 'core' quality score, or QS1--a measure of your quality based on Google.com, exact match traffic. For each variation or match of that keyword, you'll have an independently calculated quality score specific to that variation. Hence if you're bidding on "red cars" and your reported QS is 7/10, that doesn't mean your QS will be 7/10 on the matching query 'used red cars.' Instead, your QS will reflect your performance on that specific query and other relevance factors. " I hope this helps!

September 16, 2009 12:01 PM

parky said:

ROI Revolution you guys rock and I dont know where else to turn so I hope you dont mind me throwing this question out here, thanks for all your articles.

Situation is this:

Have a site with product pages that dont see enough traffic to each individual product page for a multivariate test (Google website optimizer) without waiting months for results. For example lets say I have 100 product pages each seeing about 200 views a week. I want to setup a simple multivariate test with different checkout buttons and locations. For this example lets use total of 10 combination's. If I run that multivariate test on even my most popular product page with only 200 views it will take forever to get solid results and to me Google was just never have enough solid data to calculate. Now if i could aggregate all the product pages into a test 100x200 so 20k views a week that is the kind of data i need GWO to see. So my questions is this:

Does the Javascript really care about the Test Page URL or even Conversion Page URL for that matter as long as the scripts are referenced. In other words I want to make 100 product pages look like one test page they will all have the same scripts on top of page and at bottom of page and then the checkout will always have the same conversion script. Has anyone tried something like this or any expert in JavaScript out there know if this will work? Any insight would be much appreciated.

-parky

March 25, 2009 4:18 AM

Shawn Purtell, Senior Web Analytics Engineer Author Profile Page said:

@parky:

The method you suggest should actually work just fine. Google Website Optimizer doesn't care about the test and conversion page URLs that you put in when creating an experiment - they only use those for the validation step.

If you put the same control, section, and tracking scripts on all product pages, you should be able to run your experiment just fine, and it will aggregate the data for all product pages. If a user views more than one product page in the same visit, they will see the same combination the entire time, so you shouldn't have any problems with consistency either.

Hope that helps!

March 25, 2009 8:08 AM

Dan PPC PROZ said:

apparantly there are two formulas, one for first page bid and one for adrank.

apparantly first page bid on new accounts is weighted more heavily on landing page relevancy than we originally thought, see my post...

March 28, 2009 5:36 AM

Erin Ewasyshyn, Strategy Manager - Paid Search Marketing Author Profile Page said:

@Dan, Thanks for reading and I enjoyed your post. After a discussion with our rep, I've found that landing page quality is not a bigger component of QS for newer accounts than for older accounts. It has to do with whether or not Google has evaluated the new account's landing page or not. If the new account sends traffic to a page Google hasn't seen, landing page quality will have no impact on QS at all until the site is evaluated. If the new account sends traffic to a page that Google has already crawled before and determined it not to be a good site, then the new account will have QS problems right from the beginning. I hope this helps!

March 30, 2009 1:41 PM

John Rodriguez said:

Erin, great post on QS. With limited hours in a day, it helps to know where you'll get the most bang for your buck.

It makes complete sense that CTR makes up the lion's share of QS.

Letting searchers "vote" on an ad's relevance goes a long way to ensuring that Google provides a positive experience to their users -- assuming that the actual landing page is relevant... LOL

From a business perspective, it makes a heckuva lot of sense as well -- ads with high CTR generate revenue. ;)

April 23, 2009 9:04 PM

Erin Ewasyshyn, Strategy Manager - Paid Search Marketing Author Profile Page said:

@John: Thanks for reading. I've also written an article on why we shouldn't simply let Google decide which of our ads are best (Optimize ad rotation) based on CTR, because while clicks are good for Google (and admittedly good for advertisers as well), we ultimately want *conversions* which is how most of us generate revenue! Check it out if you're interested: http://www.roirevolution.com/blog/2009/01/googles_recommendation_to_optimize_ads_and_why_its.html

April 25, 2009 10:54 AM

Alex said:

I am a newbiew and found your site while surfing.

Thank you for the great information.

I have one question on the info in the charts.

In the first chart you show Alison with a Max CPC of .05 and a Quality Score of 16.

I thought the range was 1-10?

September 11, 2009 2:51 PM

Erin Ewasyshyn, Strategy Manager - Paid Search Marketing Author Profile Page said:

@Alex: I had a similar question a few months back and passed it on to our Google rep. His reply was "The 1-10 scale you see in the UI is a proxy of your quality score--it's not your actual quality score, which has a much broader range of values than the integers between 1 and 10. Similarly, the '16' and '.5' are meant to illustrate the ads quality process, rather than reflect the actual metrics we use in the auction. In addition, bear in mind that the 1-10 score in the UI is your 'core' quality score, or QS1--a measure of your quality based on Google.com, exact match traffic. For each variation or match of that keyword, you'll have an independently calculated quality score specific to that variation. Hence if you're bidding on "red cars" and your reported QS is 7/10, that doesn't mean your QS will be 7/10 on the matching query 'used red cars.' Instead, your QS will reflect your performance on that specific query and other relevance factors. " I hope this helps!

September 16, 2009 12:01 PM

parky said:

ROI Revolution you guys rock and I dont know where else to turn so I hope you dont mind me throwing this question out here, thanks for all your articles.

Situation is this:

Have a site with product pages that dont see enough traffic to each individual product page for a multivariate test (Google website optimizer) without waiting months for results. For example lets say I have 100 product pages each seeing about 200 views a week. I want to setup a simple multivariate test with different checkout buttons and locations. For this example lets use total of 10 combination's. If I run that multivariate test on even my most popular product page with only 200 views it will take forever to get solid results and to me Google was just never have enough solid data to calculate. Now if i could aggregate all the product pages into a test 100x200 so 20k views a week that is the kind of data i need GWO to see. So my questions is this:

Does the Javascript really care about the Test Page URL or even Conversion Page URL for that matter as long as the scripts are referenced. In other words I want to make 100 product pages look like one test page they will all have the same scripts on top of page and at bottom of page and then the checkout will always have the same conversion script. Has anyone tried something like this or any expert in JavaScript out there know if this will work? Any insight would be much appreciated.

-parky

March 25, 2009 4:18 AM

Shawn Purtell, Senior Web Analytics Engineer Author Profile Page said:

@parky:

The method you suggest should actually work just fine. Google Website Optimizer doesn't care about the test and conversion page URLs that you put in when creating an experiment - they only use those for the validation step.

If you put the same control, section, and tracking scripts on all product pages, you should be able to run your experiment just fine, and it will aggregate the data for all product pages. If a user views more than one product page in the same visit, they will see the same combination the entire time, so you shouldn't have any problems with consistency either.

Hope that helps!

March 25, 2009 8:08 AM

Dan PPC PROZ said:

apparantly there are two formulas, one for first page bid and one for adrank.

apparantly first page bid on new accounts is weighted more heavily on landing page relevancy than we originally thought, see my post...

March 28, 2009 5:36 AM

Erin Ewasyshyn, Strategy Manager - Paid Search Marketing Author Profile Page said:

@Dan, Thanks for reading and I enjoyed your post. After a discussion with our rep, I've found that landing page quality is not a bigger component of QS for newer accounts than for older accounts. It has to do with whether or not Google has evaluated the new account's landing page or not. If the new account sends traffic to a page Google hasn't seen, landing page quality will have no impact on QS at all until the site is evaluated. If the new account sends traffic to a page that Google has already crawled before and determined it not to be a good site, then the new account will have QS problems right from the beginning. I hope this helps!

March 30, 2009 1:41 PM

John Rodriguez said:

Erin, great post on QS. With limited hours in a day, it helps to know where you'll get the most bang for your buck.

It makes complete sense that CTR makes up the lion's share of QS.

Letting searchers "vote" on an ad's relevance goes a long way to ensuring that Google provides a positive experience to their users -- assuming that the actual landing page is relevant... LOL

From a business perspective, it makes a heckuva lot of sense as well -- ads with high CTR generate revenue. ;)

April 23, 2009 9:04 PM

Erin Ewasyshyn, Strategy Manager - Paid Search Marketing Author Profile Page said:

@John: Thanks for reading. I've also written an article on why we shouldn't simply let Google decide which of our ads are best (Optimize ad rotation) based on CTR, because while clicks are good for Google (and admittedly good for advertisers as well), we ultimately want *conversions* which is how most of us generate revenue! Check it out if you're interested: http://www.roirevolution.com/blog/2009/01/googles_recommendation_to_optimize_ads_and_why_its.html

April 25, 2009 10:54 AM

Alex said:

I am a newbiew and found your site while surfing.

Thank you for the great information.

I have one question on the info in the charts.

In the first chart you show Alison with a Max CPC of .05 and a Quality Score of 16.

I thought the range was 1-10?

September 11, 2009 2:51 PM

Erin Ewasyshyn, Strategy Manager - Paid Search Marketing Author Profile Page said:

@Alex: I had a similar question a few months back and passed it on to our Google rep. His reply was "The 1-10 scale you see in the UI is a proxy of your quality score--it's not your actual quality score, which has a much broader range of values than the integers between 1 and 10. Similarly, the '16' and '.5' are meant to illustrate the ads quality process, rather than reflect the actual metrics we use in the auction. In addition, bear in mind that the 1-10 score in the UI is your 'core' quality score, or QS1--a measure of your quality based on Google.com, exact match traffic. For each variation or match of that keyword, you'll have an independently calculated quality score specific to that variation. Hence if you're bidding on "red cars" and your reported QS is 7/10, that doesn't mean your QS will be 7/10 on the matching query 'used red cars.' Instead, your QS will reflect your performance on that specific query and other relevance factors. " I hope this helps!

September 16, 2009 12:01 PM

parky said:

ROI Revolution you guys rock and I dont know where else to turn so I hope you dont mind me throwing this question out here, thanks for all your articles.

Situation is this:

Have a site with product pages that dont see enough traffic to each individual product page for a multivariate test (Google website optimizer) without waiting months for results. For example lets say I have 100 product pages each seeing about 200 views a week. I want to setup a simple multivariate test with different checkout buttons and locations. For this example lets use total of 10 combination's. If I run that multivariate test on even my most popular product page with only 200 views it will take forever to get solid results and to me Google was just never have enough solid data to calculate. Now if i could aggregate all the product pages into a test 100x200 so 20k views a week that is the kind of data i need GWO to see. So my questions is this:

Does the Javascript really care about the Test Page URL or even Conversion Page URL for that matter as long as the scripts are referenced. In other words I want to make 100 product pages look like one test page they will all have the same scripts on top of page and at bottom of page and then the checkout will always have the same conversion script. Has anyone tried something like this or any expert in JavaScript out there know if this will work? Any insight would be much appreciated.

-parky

March 25, 2009 4:18 AM

Shawn Purtell, Senior Web Analytics Engineer Author Profile Page said:

@parky:

The method you suggest should actually work just fine. Google Website Optimizer doesn't care about the test and conversion page URLs that you put in when creating an experiment - they only use those for the validation step.

If you put the same control, section, and tracking scripts on all product pages, you should be able to run your experiment just fine, and it will aggregate the data for all product pages. If a user views more than one product page in the same visit, they will see the same combination the entire time, so you shouldn't have any problems with consistency either.

Hope that helps!

March 25, 2009 8:08 AM

Dan PPC PROZ said:

apparantly there are two formulas, one for first page bid and one for adrank.

apparantly first page bid on new accounts is weighted more heavily on landing page relevancy than we originally thought, see my post...

March 28, 2009 5:36 AM

Erin Ewasyshyn, Strategy Manager - Paid Search Marketing Author Profile Page said:

@Dan, Thanks for reading and I enjoyed your post. After a discussion with our rep, I've found that landing page quality is not a bigger component of QS for newer accounts than for older accounts. It has to do with whether or not Google has evaluated the new account's landing page or not. If the new account sends traffic to a page Google hasn't seen, landing page quality will have no impact on QS at all until the site is evaluated. If the new account sends traffic to a page that Google has already crawled before and determined it not to be a good site, then the new account will have QS problems right from the beginning. I hope this helps!

March 30, 2009 1:41 PM

John Rodriguez said:

Erin, great post on QS. With limited hours in a day, it helps to know where you'll get the most bang for your buck.

It makes complete sense that CTR makes up the lion's share of QS.

Letting searchers "vote" on an ad's relevance goes a long way to ensuring that Google provides a positive experience to their users -- assuming that the actual landing page is relevant... LOL

From a business perspective, it makes a heckuva lot of sense as well -- ads with high CTR generate revenue. ;)

April 23, 2009 9:04 PM

Erin Ewasyshyn, Strategy Manager - Paid Search Marketing Author Profile Page said:

@John: Thanks for reading. I've also written an article on why we shouldn't simply let Google decide which of our ads are best (Optimize ad rotation) based on CTR, because while clicks are good for Google (and admittedly good for advertisers as well), we ultimately want *conversions* which is how most of us generate revenue! Check it out if you're interested: http://www.roirevolution.com/blog/2009/01/googles_recommendation_to_optimize_ads_and_why_its.html

April 25, 2009 10:54 AM

Alex said:

I am a newbiew and found your site while surfing.

Thank you for the great information.

I have one question on the info in the charts.

In the first chart you show Alison with a Max CPC of .05 and a Quality Score of 16.

I thought the range was 1-10?

September 11, 2009 2:51 PM

Erin Ewasyshyn, Strategy Manager - Paid Search Marketing Author Profile Page said:

@Alex: I had a similar question a few months back and passed it on to our Google rep. His reply was "The 1-10 scale you see in the UI is a proxy of your quality score--it's not your actual quality score, which has a much broader range of values than the integers between 1 and 10. Similarly, the '16' and '.5' are meant to illustrate the ads quality process, rather than reflect the actual metrics we use in the auction. In addition, bear in mind that the 1-10 score in the UI is your 'core' quality score, or QS1--a measure of your quality based on Google.com, exact match traffic. For each variation or match of that keyword, you'll have an independently calculated quality score specific to that variation. Hence if you're bidding on "red cars" and your reported QS is 7/10, that doesn't mean your QS will be 7/10 on the matching query 'used red cars.' Instead, your QS will reflect your performance on that specific query and other relevance factors. " I hope this helps!

September 16, 2009 12:01 PM

parky said:

ROI Revolution you guys rock and I dont know where else to turn so I hope you dont mind me throwing this question out here, thanks for all your articles.

Situation is this:

Have a site with product pages that dont see enough traffic to each individual product page for a multivariate test (Google website optimizer) without waiting months for results. For example lets say I have 100 product pages each seeing about 200 views a week. I want to setup a simple multivariate test with different checkout buttons and locations. For this example lets use total of 10 combination's. If I run that multivariate test on even my most popular product page with only 200 views it will take forever to get solid results and to me Google was just never have enough solid data to calculate. Now if i could aggregate all the product pages into a test 100x200 so 20k views a week that is the kind of data i need GWO to see. So my questions is this:

Does the Javascript really care about the Test Page URL or even Conversion Page URL for that matter as long as the scripts are referenced. In other words I want to make 100 product pages look like one test page they will all have the same scripts on top of page and at bottom of page and then the checkout will always have the same conversion script. Has anyone tried something like this or any expert in JavaScript out there know if this will work? Any insight would be much appreciated.

-parky

March 25, 2009 4:18 AM

Shawn Purtell, Senior Web Analytics Engineer Author Profile Page said:

@parky:

The method you suggest should actually work just fine. Google Website Optimizer doesn't care about the test and conversion page URLs that you put in when creating an experiment - they only use those for the validation step.

If you put the same control, section, and tracking scripts on all product pages, you should be able to run your experiment just fine, and it will aggregate the data for all product pages. If a user views more than one product page in the same visit, they will see the same combination the entire time, so you shouldn't have any problems with consistency either.

Hope that helps!

March 25, 2009 8:08 AM

Dan PPC PROZ said:

apparantly there are two formulas, one for first page bid and one for adrank.

apparantly first page bid on new accounts is weighted more heavily on landing page relevancy than we originally thought, see my post...

March 28, 2009 5:36 AM

Erin Ewasyshyn, Strategy Manager - Paid Search Marketing Author Profile Page said:

@Dan, Thanks for reading and I enjoyed your post. After a discussion with our rep, I've found that landing page quality is not a bigger component of QS for newer accounts than for older accounts. It has to do with whether or not Google has evaluated the new account's landing page or not. If the new account sends traffic to a page Google hasn't seen, landing page quality will have no impact on QS at all until the site is evaluated. If the new account sends traffic to a page that Google has already crawled before and determined it not to be a good site, then the new account will have QS problems right from the beginning. I hope this helps!

March 30, 2009 1:41 PM

John Rodriguez said:

Erin, great post on QS. With limited hours in a day, it helps to know where you'll get the most bang for your buck.

It makes complete sense that CTR makes up the lion's share of QS.

Letting searchers "vote" on an ad's relevance goes a long way to ensuring that Google provides a positive experience to their users -- assuming that the actual landing page is relevant... LOL

From a business perspective, it makes a heckuva lot of sense as well -- ads with high CTR generate revenue. ;)

April 23, 2009 9:04 PM

Erin Ewasyshyn, Strategy Manager - Paid Search Marketing Author Profile Page said:

@John: Thanks for reading. I've also written an article on why we shouldn't simply let Google decide which of our ads are best (Optimize ad rotation) based on CTR, because while clicks are good for Google (and admittedly good for advertisers as well), we ultimately want *conversions* which is how most of us generate revenue! Check it out if you're interested: http://www.roirevolution.com/blog/2009/01/googles_recommendation_to_optimize_ads_and_why_its.html

April 25, 2009 10:54 AM

Alex said:

I am a newbiew and found your site while surfing.

Thank you for the great information.

I have one question on the info in the charts.

In the first chart you show Alison with a Max CPC of .05 and a Quality Score of 16.

I thought the range was 1-10?

September 11, 2009 2:51 PM

Erin Ewasyshyn, Strategy Manager - Paid Search Marketing Author Profile Page said:

@Alex: I had a similar question a few months back and passed it on to our Google rep. His reply was "The 1-10 scale you see in the UI is a proxy of your quality score--it's not your actual quality score, which has a much broader range of values than the integers between 1 and 10. Similarly, the '16' and '.5' are meant to illustrate the ads quality process, rather than reflect the actual metrics we use in the auction. In addition, bear in mind that the 1-10 score in the UI is your 'core' quality score, or QS1--a measure of your quality based on Google.com, exact match traffic. For each variation or match of that keyword, you'll have an independently calculated quality score specific to that variation. Hence if you're bidding on "red cars" and your reported QS is 7/10, that doesn't mean your QS will be 7/10 on the matching query 'used red cars.' Instead, your QS will reflect your performance on that specific query and other relevance factors. " I hope this helps!

September 16, 2009 12:01 PM

Post Your Comments

© 2002-2014 ROI Revolution, Inc. All rights reserved.